HCI Bibliography Home | HCI Conferences | HFS Archive | Detailed Records | RefWorks | EndNote | Hide Abstracts
HFS Tables of Contents: 87-187-288-188-289-189-290-190-291-191-292-192-293-193-294-194-295-195-2

Proceedings of the Human Factors and Ergonomics Society 38th Annual Meeting 1994-10-24

  1. HFS 1994-10-24 Volume 2
    1. MEDICAL SYSTEMS AND REHABILITATION: User-Centered Design in Medicine/Productivity and Ergonomics in Rehabilitation [Lecture]
    2. MEDICAL SYSTEMS AND REHABILITATION: Panel
    3. ORGANIZATIONAL DESIGN AND MANAGEMENT: Panel
    4. ORGANIZATIONAL DESIGN AND MANAGEMENT: Community Ergonomics: A Theoretical Model for Rebuilding the Inner City [Symposium]
    5. ORGANIZATIONAL DESIGN AND MANAGEMENT: Potpourri: A Diversity of ODAM Interests [Lecture]
    6. ORGANIZATIONAL DESIGN AND MANAGEMENT: Macroergonomic Concepts and Tools [Lecture]
    7. PERSONALITY AND INDIVIDUAL DIFFERENCES IN HUMAN PERFORMANCE: Personality and Individual Differences in Human Performance [Lecture]
    8. SAFETY: Safety Potpourri [Lecture]
    9. SAFETY: Product Label Design and Consumer Behavior [Lecture]
    10. SAFETY: Panel
    11. SAFETY: Arnold M. Small Lecture in Safety [Symposium]
    12. SAFETY: Visual Information Processing and Driver Performance [Lecture]
    13. SAFETY: Human Factors Issues in the Maritime Industry [Symposium]
    14. SAFETY: Safety Potpourri II [Lecture]
    15. SAFETY: Perception and Understanding of Highway Signs [Lecture]
    16. SAFETY: Decision Making in Technical Environments [Symposium]
    17. SPECIAL SESSIONS: Alternative Format
    18. SPECIAL SESSIONS: Demonstrations
    19. SPECIAL SESSIONS: Posters
    20. SYSTEM DEVELOPMENT: Development Activity at the System Level [Lecture]
    21. SYSTEM DEVELOPMENT: Steps in the System Development Process [Lecture]
    22. SYSTEM DEVELOPMENT: Tools in the System Development Process [Lecture]
    23. SYSTEM DEVELOPMENT: Panel
    24. SYSTEM DEVELOPMENT: Developing Human Factors Guidelines for Intelligent Vehicle/Highway Systems: Third Progress Report [Symposium]
    25. TEST AND EVALUATION: Panel
    26. TEST AND EVALUATION: Test and Evaluation Methodology [Lecture]
    27. TEST AND EVALUATION: Test and Evaluation of Product Design [Lecture]
    28. TEST AND EVALUATION: Test and Evaluation Involving Higher-Order Functions [Lecture]
    29. TEST AND EVALUATION: Test and Evaluation of Performance-Shaping Functions [Lecture]
    30. TRAINING: Panel
    31. TRAINING: Training on the Technological Frontier [Lecture]
    32. TRAINING: Insights into Learning and Interaction [Lecture]
    33. TRAINING: Investigating the Unique Contribution of Feedback in Teams: Implications For Training [Symposium]
    34. TRAINING: Computers in the Training of Complex Tasks [Symposium]
    35. TRAINING: Data-Based Training Studies [Lecture]
    36. VISUAL PERFORMANCE: Visual Search Performance [Lecture]
    37. VISUAL PERFORMANCE: Matching Visual Displays to Information-Processing Capabilities [Lecture]
    38. VISUAL PERFORMANCE: Stereoscopic Display [Lecture]
    39. VISUAL PERFORMANCE: Panel
    40. VISUAL PERFORMANCE: Vigilance, Memory, and Mental Workload [Lecture]
    41. VISUAL PERFORMANCE: Interpreting Graphic Displays [Lecture]

HFS 1994-10-24 Volume 2

MEDICAL SYSTEMS AND REHABILITATION: User-Centered Design in Medicine/Productivity and Ergonomics in Rehabilitation [Lecture]

Improving Ultrasound Systems by User-Centered Design BIBA 705-709
  Arlene F. Aucella; Thomas Kirkham; Susan Barnhart; Lawrence Murphy; Kris LaConte
A series of user-centered design methods were applied during the development of an ultrasound system. The methods included hospital site visits, interviews with sonographers, videotapes of examinations and usability testing with simulations of the console and displays. Based on user feedback and preferences, over one-hundred design changes were made to both the console and control panel. The resulting system accommodates users through an adjustable control panel and display monitor. In addition, the control panel and corresponding operator interface software reflects frequency-of-use and sequencing-of-use of functions, leading to more efficient interaction with the system.
Users as Designers: How People Cope with Poor HCI Design in Computer-Based Medical Devices BIBA 710-714
  Jodi Heintz Obradovich; David D. Woods
In this paper, we examine how users interact with a computer-based infusion device adapted for terbutaline infusion to treat preterm labor in women experiencing high-risk pregnancies. This study examines: (1) the HCI deficiencies in the device as related to this context of use, (2) how the device characteristics increase the potential for error, and (3) the tailoring strategies developed by users to insulate themselves from failure. Interviews with nurses and bench tests of the behavior of the infusion device in different conditions identified several HCI deficiencies: complex and arbitrary sequences of operation, mode errors due to poor differentiation of multiple operating modes intended for different contexts, ambiguous alarms, getting lost in multiple displays, and poor feedback on device state and behavior.
Integrating Ergonomics into the Rehabilitation Process: A Multidisciplinary Approach for Successful Return-to-Work BIBA 715-718
  Elizabeth A. Mayne; Julie M. Sawyer
This paper presents a system for successfully returning an injured employee to work through a multi-disciplinary approach. It details a process that interfaces physical and occupational therapists, rehabilitation specialists and ergonomists, among others. The aims of an effective and efficient occupational rehabilitation process are to ultimately reduce the societal costs of health care. The factors that contribute to these costs are many, and can be influenced by the behavior of the injured worker, the employer, the doctors and other health care providers. This paper discusses the complementary functions of the ergonomist and the therapist: the ergonomist, through workplace evaluation and redesign, considers the external factors that influence an individual's performance; the therapist evaluates and aims to improve the internal factors that affect an individual's performance. A match or mismatch between the physical demands of a job and an injured employee can then easily be detected by comparing the results of an ergonomic assessment and patient evaluation by a therapist. The authors conclude by discussing problems and benefits encountered through this approach.
An Investigation of the Relationship of Disability and Productivity of Visually Impaired Worker BIB --
  Jacob Jen-Gwo Chen; Lawrence J. H. Schulze; Min-Der Ko; Aziz U. Huq

MEDICAL SYSTEMS AND REHABILITATION: Panel

Risks, Incidents, and Errors in Medicine BIB --
  Marilyn Sue Bogner; Dennis I. Serig; Denise C. R. Benel; Colin F. MacKenzie; Steven K. Howard

ORGANIZATIONAL DESIGN AND MANAGEMENT: Panel

The IEA and Ergonomics Internationally BIBA 719-723
  Hal W. Hendrick; Andrew Imada; Waldemar Karwowski; Ogden, Jr. Brown; Ian Noy
The conception, development and growth of the IEA and ergonomics internationally is reviewed. The current structure and functioning of the IEA is described. Recent activities of the IEA's two largest committees, Education and Training and Science and Technology, are summarized. A brief overview is provided of the IEA federated societies and current international ergonomics. With these data as background, the panelists will discuss current trends and issues within the IEA and ergonomics internationally.

ORGANIZATIONAL DESIGN AND MANAGEMENT: Community Ergonomics: A Theoretical Model for Rebuilding the Inner City [Symposium]

Community Ergonomics: A Theoretical Model for Rebuilding the Inner City BIBA 724-728
  Michael J. Smith; Pascale Carayon; John Smith; William Cohen; Jerry Upton
There is a critical need for various forces that are working to improve the inner city to coordinate their efforts and to develop cooperative approaches. This includes governments at all levels, social agencies, educational systems, the business community, financial institutions, religious groups, ethnic groups, police, universities and inner-city residents. The diversity in values, approaches, goals, desires and legal requirements makes such a systematic integration very complex and difficult. Several human factors and ergonomics theories for the management and control of complex systems can provide insight into effective means to address these issues. General principles can be established to promote "community ergonomics."
Empowerment Zones and Enterprise Communities BIB --
  A Albert Gore (Invited)
Crusade to Emancipate American Inner Cities BIB --
  Reggie White
Community Ergonomics: An Emerging Theory and Engineering Practice BIBA 729-733
  John H. Smith; Michael J. Smith
The context and condition specific relationship of certain human populations and their communities to technology, the environment, and numerous interactions which contribute to conditions of economic-social alienation, are contributing negatively to community quality, productivity and work organization of the American city. The need is for systems methods to resolve and prevent the occurrence of negative design factors which contribute to disputes, conflict and social impairment of individuals and institutions. This paper will advance the practice of Community Ergonomics to improve human interactions, evolution, law, and in planning for positive organization of groups for self-regulation and control of community resources and work.
Community Ergonomics: Past Approaches and Future Prospects Toward America's Urban Crisis BIBA 734-738
  William J. Cohen; John H. Smith
The purpose of this paper is to examine past and current urban programs and to propose a new human engineering approach to solve entrenched problems. The application of a systematic engineering philosophy is central to a process that is focused around the concepts of work. As such, a new application of industrial engineering and human factors is proposed. The creation of stable, worthwhile jobs is crucial, but only in relation to an entire system which includes (among many others) existing governmental programs, educational systems, law enforcement financial institutions, and especially, people's desires and needs towards bettering their lives. This approach is neither entirely governmental nor privatized. Private industry will be created and encouraged to utilize a non-traditional workforce. Benefits from governmental programs will be utilized to promote an entrepreneurial spirit that can succeed outside of the "underground economy". In addition, efforts will be coordinated to help people become increasingly independent from hand-outs. The ultimate goal of community ergonomics is to reinvent the "fit" between people, their cities, and society. The development of people and jobs together is vital to this effort. In order for this approach to succeed, programs must propose both empirical research and practical applications.
Community Ergonomics: Data Collection Methods and Analysis of Human Characteristics BIBA 739-743
  Leah Newman; Pascale Carayon
This paper presents methods that can be used in community ergonomics to examine the Community-Environment fit. An example is discussed that illustrates the use of survey questionnaire in assessing the fit between the community and its banking environment.
Community Ergonomics: A Conceptual Model and a Pilot Test in Knoxville, Tennessee BIB --
  Jerry Upton; Lauren Tullock; Michael J. Smith
Future Directions in Community Redevelopment BIB --
  A Henry Cisneros (Invited)

ORGANIZATIONAL DESIGN AND MANAGEMENT: Potpourri: A Diversity of ODAM Interests [Lecture]

Response Surface Analysis of Effects of Situation Constraints on Communications Media Choice in Organizations BIBA 744-748
  Barrett S. Caldwell; Shiaw-Tsyr Uang
Past research on communications media use in organization has frequently focused on conditions such as the technical, economic, psychological, and social factors that influence use of media in organizations. However, few studies deal with situation factors. The purpose of this paper is to examine the relationships between situation factors and utilization of communication media in organizations. These relationships between situation variables and media use are contrasted against previous media choice theoretical models (Social Presence, Information Richness, and Social Information Processing). In order to investigate the main and interaction effects of three situation constraints (urgency of message, the amount of message content, and distance between communicators), eight situations were analyzed through response surface methodology, using surveys collected from 1072 voice mail users in Wisconsin. The results strongly indicate the necessity of integrating situation factors into media choice models.
Long-Term Evaluation of a Compressed, Rapidly Rotating Work Schedule BIBA 749-753
  Michael J. Paley; Leslie Beth Herbert; Donald I. Tepas
A current trend in alternative work schedules is to increase the length of the workday while decreasing the number of days in the work week. These "compressed" work schedules are popular amongst shiftworkers, in part because of longer periods of non-work days. The current research is an attempt to add to the short list of long-term evaluations of compressed work schedules. A change in work schedules from an 8 hour rotating shift schedule to a rotating 10 hour day/14 hour night operating schedule in a fire department was studied. Multiple measures were used to address a variety of outcomes associated with the new schedule. From the significant increase in satisfaction with their work schedule, it is clear that the new shift schedule is well liked by the fire fighters. Reasons for the satisfaction include significant changes in usual sleep length, a decrease in being tired or sleepy at work and an increase in free time benefits. Other results are discussed in the paper. In conclusion, the use of a compressed work schedule is beneficial when workers are allowed to sleep on the job, however, generalizations to other workplaces must be limited. Use of these popular work schedules should be limited to applications which incorporate a systematic evaluation of the new work schedule.
Self-Management of Rest Breaks by VDT Users BIBA 754-758
  Robert A. Henning; Anna M. Ortega; Eric A. Callaghan; George V. Kissel
Video display terminal (VDT) users can benefit from frequent, short rest breaks in terms of improved productivity and well-being. However, VDT users report that scheduled breaks can seriously disrupt some tasks. This laboratory study tested if performance feedback would promote better self management of discretionary rest breaks. Undergraduate typists (N=31) entered lines of randomized words for 65 min. A mandatory rest break was administered whenever discretionary rest breaks did not total 30 s every 10 min. Typists in the experimental condition received feedback indicting how their discretionary breaks compared to a criterion. Typists in the control condition received no such feedback. Mood and musculoskeletal discomfort were assessed before and after the work period, followed by a questionnaire about the break system. Performance measures included keystroke rate, error rate, and correction rate (backspace use). Measures of heart rate and heart rate variability were also collected. Both the number of full-length mandatory breaks and correction rate were lower in the feedback condition. No significant differences in mood, musculoskeletal discomfort, physiological response, nor acceptance of the break system were found. These results suggest that self management of discretionary breaks as well as keystroke performance are improved by feedback, with no untoward effects on well-being.
Developing a Better Design Process: Redesigning the Organization to Produce More Ergonomic Designs BIBA 759-763
  Christopher A. Dockery; Thomas Neuman
This paper explores the development of a systematic process for effectively considering in the product design process the ergonomics requirements of all who will interact with a product during its entire life cycle. Because of its very broad scope relative to ergonomics, the process has huge organizational and management implications. Focus in this paper is given to: the steps and rationale involved in developing the process so that ownership and buy-in would result along with a sound and valid process that would be likely to be used; organizational and management implications for communicating and implementing the process including concurrent engineering, accountability, and resource availability. Results of application of the process are illustrated by several case studies from divisions of a large corporation. Organizational structures and relationships that encourage successful partial implementations are explored. Deficiencies in the traditional product design process are explored from motivational and perspective limitation standpoints.

ORGANIZATIONAL DESIGN AND MANAGEMENT: Macroergonomic Concepts and Tools [Lecture]

High-Involvement Ergonomics: A New Approach to Participation BIBA 764-768
  Ogden, Jr. Brown
High involvement ergonomics is introduced as a macroergonomic approach to the implementation of technology in organizational systems which requires that workers be involved in developing and implementing the technology. It has been proposed that work organizations move toward a more involved or commitment approach to the design and operation of the organization. Several types of involvement approaches are discussed, the managerial role in high involvement organizations is examined, and conclusions are drawn regarding the implementation of high involvement ergonomics.
Quantitative Approaches to Team Communication and Performance: A Harmony of Systems Analysis Tools BIBA 769-773
  Barrett S. Caldwell
Effective team performance in dynamic, high-consequence environments requires an interdisciplinary integration of human interactions with technological systems components. Several systems analysis approaches have substantial applications to the study of team communication and performance. This paper examines cybernetics and sociotechnical approaches to information exchange and coordination in human-technology systems. One component of cybernetics, based on quantitative feedback control engineering tools, can be effectively integrated with group communication and coordination principles from sociotechnical research. The author suggests ways in which these systems frameworks can be utilized and consolidated to support a more harmonious and quantitatively grounded examination of human performance in complex systems.
Macroergonomics in the Context of Vendor-Customer Technology Transfer BIBA 774-778
  Wayne Fisher
The transfer of a system-level product from vendor to customer typically requires that the customer change both business processes and organizational structures as part of the system implementation process. A macroergonomic approach was used to analyze the transfer of an image based check processing system to customer sites and identify opportunities to improve the transfer process. The analysis identified a lack of knowledge of the organizational, administrative and managerial changes that would be introduced by the new technology as a potential risk to the project implementation schedule. The analysis indicated that establishing a knowledge base covering five non-technical areas (general system familiarity, worksite preparation, human resources, operations and system support) and transferring that knowledge base to the customer's implementation team would minimize the potential schedule risk. A specialized consulting role was created to transfer the knowledge base to the customer's implementation team through a series of seminar sessions tied to specific implementation milestones. Each session focused on assisting the customer's implementation team in developing an action plan to allocate and schedule the resources necessary to carry the project through to the next milestone. The structure of the knowledge base and the presentation format for the consulting service were verified during initial trials at customer sites. A curriculum of classroom instruction and apprenticeship was then developed to train the consultants. Qualitative feedback from customers indicated that the service was perceived to provide value to the participants.
Reengineering a Large Technology-Based Organization BIBA 779-783
  Heidi Ann Hahn; Richard P. Bastian; Pamela R. French
Los Alamos National Laboratory is a large science- and technology-based organization that has recently undertaken a major reengineering effort in an attempt to maintain its competitiveness in a rapidly changing world. Although the initial thrust was toward altering processes, management systems, and culture, it was decided that a major reorganization was needed as well. Along with the reorganization came a flattening of the management hierarchy and an almost complete change in the composition of the senior management group. This restructuring has had profound effects on the organization, requiring a new approach to management by the senior executives, and, at least initially, has swamped process and systems changes that will ultimately determine the success or failure of the effort. This paper documents the journey -- how the structure and behaviors of the senior management group had to change to work in the new environment, trickle-down effects of reorganization, levels of decision-making, integrating processes and systems, and communicating with employees -- and identifies some lessons learned that might be useful to other large technical organizations contemplating such changes.

PERSONALITY AND INDIVIDUAL DIFFERENCES IN HUMAN PERFORMANCE: Personality and Individual Differences in Human Performance [Lecture]

Human Factors Issues for Resolving Adverse Effects of Human Work Underload and Workload Transitions in Advanced Transportation Systems BIBA 784-788
  Thomas G. Ryan
A workshop was conducted in which the specific purpose was to build on earlier work by the National Research Council, Federal Government agencies, and the larger human factors community to: (1) clarify human factors issues pertaining to degraded performance in advanced transportation systems due to human work underload and workload transition; and (2) develop strategies for resolving these issues.
Designing an Intelligent Tutoring System Interface to Account for Personality Differences BIBA 789-792
  Pauline K. Cushman
Intelligent Tutoring Systems have been designed for a variety of purposes. Much of the design effort has been aimed at the actual subject matter. Often ignored has been the critical nature of the interface. If the way people interact with computers is directly related to their personality, then systems should respond differently to different people. This paper describes the design of an interface for an Intelligent Tutoring System that, given the student's personality, will make adjustments in the style of interaction.
Technology, Teacher Style, and Classroom BIBA 793-795
  Anita Ambardar
The purpose of this program of research was to investigate the impact of computer/technology in student teaming and motivation to learn in an urban school setting.
   Specifically, the following questions were raised:
  • 1. DOES THE USAGE OF COMPUTER/TECHNOLOGY ENHANCE STUDENT PERFORMANCE AND
        MOTIVATION TO LEARN IN AN URBAN CLASSROOM SETTING?
  • 2. HOW ARE THE TEACHERS INTEGRATING TECHNOLOGY IN THEIR INSTRUCTIONAL
        PRACTICES?
  • 3. WHAT FACTORS INFLUENCE TEACHERS TO USE TECHNOLOGY IN A CLASSROOM? Fourteen teachers from fourteen classrooms participated in the project. Four hundred and eighteen students were part of this project. The students involved were from grade 1 through seventh grade. Apple two GS, IBM computers and commercially available software were used for the project.
       The instructional program involved three phases -- training workshops, in-class demonstrations and observations/follow up.
       The results showed that the use of technology in reading/writing connection enhances student performance and students motivation to learn. The teachers' beliefs and teaching style proved to be significant in integration of technology in classroom instructions. The factors, the teachers considered important were training of teachers, time for preparation and technical assistance.
  • Is Motion Sickness Hereditary? BIBA 796-800
      Thomas M. Yanus; Frederick V. Malmstrom
    Motion Sickness surveys were administered to 77 male graduate students and 95 male and female members of an Air Force Reserve medical unit. Results of Survey 1 indicated significant and sizeable correlation's between sons and their natural fathers for severity of motion sickness symptoms of fatigue, headache, and nausea. Stepwise multiple regression indicated that the father's nausea accounted for 59.4% of the total variance of their son's nausea during travel. Results of Survey 2 indicated sizeable and significant correlations between respondents and both natural parents for percentage of reported symptoms of fatigue and headache. Stepwise multiple regression indicated that both parents' reported percentage of fatigue (26.6%) and headache (33.6%) reliably predicted the respondents' percentage of fatigue and headache symptoms but not the nausea symptoms. In addition, there were no significant correlations between self-reported symptoms of fatigue, headache, and nausea. Findings suggest (1) hereditary factors may be a significant contributor to motion sickness, both in severity and frequency of symptoms, and (2) fatigue, headache, and nausea are largely independent of each other. Indications are that motion sickness be considered a syndrome rather than a unitary disorder.

    SAFETY: Safety Potpourri [Lecture]

    Intelligibility of Synthesized Voice Messages in Commercial Truck Cab Noise for Normal-Hearing and Hearing-Impaired Listeners BIBA 801-805
      H. Boyd Morrison; John G. Casali
    A human factors experiment was conducted to assess the intelligibility of synthesized speech under a variety of noise conditions for both hearing-impaired and normal-hearing subjects. Modified Rhyme Test stimuli were used to determine intelligibility in four speech-to-noise (S/N) ratios (0, 5, 10, and 15 dB), and three noise types, consisting of flat-by-octaves (pink) noise, interior noise of a currently produced heavy truck, and truck cab noise with added background speech. A quiet condition was also investigated. During recording of the truck noise for the experiment, in-cab noise measurements were obtained. According to OSHA standards, these data indicated that drivers of the sampled trucks have a minimal risk for noise-induced hearing loss due to in-cab noise exposure when driving at freeway speeds because noise levels were below 80 dBA. In the intelligibility expert neat, subjects with hearing loss had significantly lower intelligibility than normal-hearing subjects, both in quiet and in noise, but no interaction with noise type or S/N ratio was found. Intelligibility was significantly lower for the noise with background speech than the other noises, but the truck noise produced intelligibility equal to the pink noise. An analytical prediction of intelligibility using Articulation Index calculations exhibited a high positive correlation with the empirically obtained intelligibility data for both groups of subjects.
    Analysis of Accident Data and Fatal Risk for Occupational Use of Extension Ladders BIBA 806-810
      Christine T. Wood; Roger L. McCarthy; Jeya Padmanaban; Roman R. Beyer
    This paper presents the results of analyses of injury and fatality accidents associated with the occupational use of metal extension ladders. Data contained in seven different data bases differing in geographic representation and level of severity of injury were analyzed to identify the type of accidents that occur and their frequency. In addition, the risks of fatality and of electrocution fatality associated with occupational use of extension ladders were estimated and compared with the fatal risk for various occupations. The overall fatal risk for extension ladder use by workers is less than the overall fatality risk for the construction industry.
    The Time Delay to Start Evacuating upon Hearing a Fire Alarm BIBA 811-815
      Guylene Proulx
    An experiment was designed to observe the evacuation drills of occupants in four mid-rise apartment buildings. Each mid-rise building contained approximately 100 apartments, with 1 to 4 persons per apartment. All printed fire safety procedures stated that upon hearing the fire alarm all occupants should leave the building or move to an area of refuge, which means the movement of 100 to 200 people for each drill. The buildings chosen for this project were characterized as mixed-occupancy buildings; that is, they included adults, children, seniors and people with handicaps.
       Occupants received a memo that an evacuation drill would take place during the upcoming week. The evacuations were planned and carried out with the full participation of the local fire departments. The evacuation drills were recorded on video-cameras located throughout the buildings. This paper presents the results regarding the time at which each occupant started to evacuate which varies between 30 seconds to over 24 minutes. Most occupants who heard the fire alarm started their evacuation approximately 2 1/2 minutes after the alarm activation. The impact of the alarm system, the location of alarm-bells, training and pre-evacuation actions are discussed.
    Ergonomic Analysis of Extension Ladders BIBA 816-820
      Roman R. Beyer; Roger L. McCarthy; Christine T. Wood
    Although the most serious accident mode associated with extension ladder use has always been "falling off the ladder" there has been increasing attention, recently, on "contact with electric current" or electrocution accidents primarily with aluminum ladders. The U.S. Consumer Product Safety Commission (CPSC) has made several recommendations to address this problem from a technical approach. They have proposed insulating and/or isolating the ladder electrically or replacing it with a fiberglass ladder. Long extension ladders are heavy, and proposed recommendations by the CPSC to make aluminum extension ladders nonconductive include replacing them with fiberglass extension ladders, which will definitely result in a heavier ladder. A heavier ladder has implications for the characteristics of the user population and for types of injury patters.
       Occupational accident data show that ladder associated accidents directly related to ladder weight, such as overexertion, are more likely. If ladder weight were increased for example, by using exclusively nonconductive fiberglass ladders, overexertion would be a greater concern. A statics model was developed that uses a variety of variables relevant to the ladder erection process to determine the amount of force required of users to erect various types of extension ladders. Anthropometric data on the male and the female adult population were compared to the ladder user requirements. The percentages of males and of females capable of erecting 12.2-meter (40-foot) extension ladders were determined. Only three percent of female adults have the strength and reach height to erect 12.2-meter (40-foot) fiberglass extension ladders. Mandated use of 12.2-meter (40-foot) fiberglass extension ladders would greatly restrict the numbers of females who could perform jobs requiring the use of such ladders.

    SAFETY: Product Label Design and Consumer Behavior [Lecture]

    Type Form Variables: Differences in Perceived Readability and Perceived Hazardousness BIBA 821-825
      N. Clayton Silver; Paul B. Kline; Curt C. Braun
    This investigation assessed the type form variables that influence the readability of warnings and the subjective hazardousness of the product. The variables used were font type, point size contrast between the signal word and the main body of the warning (signal word-text size difference), and point size of the signal word. Fifty undergraduate students rated 36 insecticide labels that contained a warning which varied across all levels of the three variables. The composite variable "perceived readability" was formed from the averaged ratings from questions regarding readability and saliency. Another composite variable "perceived hazardousness" was formed by averaging the ratings of hazardousness of the product and carefulness in usage. Results for perceived readability and perceived hazardousness were analogous. Results showed that Century Schoolbook was perceived as more readable than Bookman or Helvetica. Moreover, greater perceived readability was obtained when there was no difference between the point sizes of the signal word and the body of the warning. Perceived readability and perceived hazardousness decreased as the signal word-text size difference increased. There was greater perceived readability and perceived hazardousness when the signal word was printed in 14-point type than in 12- or 10-point types. Implications for warning design are discussed.
    Behavioral Compliance with Warnings: Effects of Stress and Placement BIBA 826-830
      Amy Barlow Magurno; Michael S. Wogalter
    Research on the effectiveness of warnings has tended to focus on internal design aspects including variables such as the inclusion of various pictorials, color, and signal words. Only a few studies have examined the influence of warning-related variables that are external to the design of the warning itself although there have been some exceptions such as research on the effects of social influence and cost of compliance. Another potentially important external factor with respect to warning effectiveness is stress. Stress has been shown to influence the quality of decision-making and judgment formation in other domains. The present research examined the effects of stress and warning placement on compliance behavior. Participants were assigned randomly to one of four conditions in a 2 (Stress: lower vs. higher) x 2 (Warning Placement as a posted sign vs. within a set of task instructions) between-subjects design experiment. In the higher stress condition, participants were given a time limit to complete the task, and during the task the experimenter stood immediately adjacent to the participant, appearing to be measuring and timing the participant's performance. Thus in this condition there were both time-pressure and social-evaluation stress. In the lower stress condition, participants were given as much time as they needed to complete the task and the experimenter stood at a distance, out of the participants field of view. Participants performed a chemistry task in which they weighed and measured various chemical substances that appeared potentially hazardous, but were actually safe. A warning directing participants to wear mask and gloves while performing the task was present in one of the two locations. Compliance with the warning (wearing of protective equipment) was significantly higher among participants under lower stress and who were exposed to the within-instructions warning. The findings add to knowledge about the effects of external warning factors by showing that stress, such as that evoked in the present experiment, affects the extent to which warnings are complied with. Implications of these results are discussed.
    The Interaction of Signal Word and Color on Warning Labels: Differences in Perceived Hazard BIBA 831-835
      Curt C. Braun; Lori Sansing; N. Clayton Silver
    Previous research has examined the connoted hazard of various colors and signal words separately. The purpose of the present research was to examine the interaction of signal words and colors. A sample of 30 undergraduates rated the perceived hazard of 105 signal word/color combinations printed in specific hazard colors. Of the colors used, red conveyed the highest level of perceived hazard followed by orange, black, green, and blue. There were significant differences among the signal words which were grouped into three different hazard level categories. High hazard words conveyed significantly more hazard than moderate and low hazard words. Likewise, moderate hazard words conveyed significantly more hazard than low hazard word group. More importantly, however, it was noted that a signal word such as DEADLY connoted less hazard when printed in green than red ink. Implications for warning design are discussed.
    Enhancing Comprehension and Retention of Safety-Related Pictorials BIBA 836-840
      John W. Brelsford; Michael S. Wogalter; James A. Scoggins
    Because of their relatively universal information transmission potential, pictorials have been suggested as a common means of safety communications across heterogeneous groups of users and uses. The present study used a training paradigm designed to enhance comprehension and retention of pharmaceutical and industrial-safety pictorials. Manipulated were time of testing (prior to training, immediately following training, and after a one-week delay), content of instruction (supplying the associated verbal label vs. the verbal label plus an extra explanatory statement), and difficulty level ("easy" vs. "difficult" to understand pictorials according to comprehension rates in earlier studies). Using an incomplete factorial mixed-model design experiment, the results showed substantial training effects. There was little change in scores between the test immediately after training and the test after a one-week delay (and the final test scores did not differ between participants who took or did not take the immediate post-training test). Easy pictorials were comprehended (both initially and following training) better than difficult pictorials, although the latter showed the most dramatic increase in understandability after training. Additionally, the instruction content manipulation (adding the explanatory statement to the verbal label) -- which had been expected to influence the degree of encoding -- had no effect on retention. The substantial gains in understanding the more difficult pictorials suggest that brief training, as little as giving the pictorial's verbal meaning once, can have a large impact in facilitating comprehension for pictorials that would otherwise not be understood by many people.

    SAFETY: Panel

    The Proposed OSHA Standard Related to Ergonomics BIBA 841
      David J. Cochran; Barbara A. Silverstein; Deborah Berkowitz; Gregory A. Worrell; Thomas J. Albin; William S. Marras; Thomas J. Armstrong; Todd R. Brown
    This roundtable discussion will be devoted to the proposed OSHA Standard related to Ergonomics which is scheduled to be published the first week in October. The discussion will be timely and important. This is a very controversial topic. The participants have been selected from different backgrounds and opinions.

    SAFETY: Arnold M. Small Lecture in Safety [Symposium]

    Arnold M. Small Lecture in Safety BIB --
      Jerry R. Duncan; R. Todd Brown
    Design of Jobs for Control of Cumulative Trauma Disorders of the Upper Limbs BIB --
      Thomas J. Armstrong

    SAFETY: Visual Information Processing and Driver Performance [Lecture]

    Age and Driver Time Requirements at Intersections BIBA 842-846
      Neil Lerner
    Current highway design models for required sight distance a stop-sign controlled intersections assume that the perception-reaction time (PRT) required is 2.0 seconds. That is, a 2.0 second interval to perceive, evaluate, decide, and initiate a response, is adequate to cover the range of time it takes real drivers to do this. This experiment evaluated the adequacy of the 2.0 second PRT assumption, including specific consideration of older drivers, who are known to experience relatively greater difficulty at intersections. Subjects in three age groups (20-40; 65-69; and 70+ years old) drove their own vehicles (fitted with a computer-controlled video-based data collection system) over a route that included 14 stop-controlled intersections. At each stop sign, they were required to make ratings of "road quality;" this broke visual search, and provided an opportunity for the experimenter to precisely define the initiation of search and the initiation of forward movement (thus defining PRT). The 2.0 second PRT assumption was found to work reasonably well for all age groups, and corresponded to roughly the with percentile PRT for all subjects. PRTs for older subjects were slightly (but significantly) briefer than for younger drivers. Reasons for not observing a slowing of intersection PRT with advancing age are discussed. The findings are also compared to gap acceptance data from another experiment. Even though the present experiment did not find objective evidence of older drivers requiring longer decision times, older subjects nonetheless demanded longer gaps in traffic in order to judge it safe to enter traffic.
    Scaling of Relative Velocity between Vehicles BIBA 847-851
      Errol R. Hoffmann; Rudolf G. Mortimer
    Data are presented on the ability of drivers to perceive and scale the relative velocity between their own and a lead vehicle. Experiments were carried out on four groups of subjects using Ekman's ratio-rating method. Only when the subtended angular velocity of the lead vehicle exceeded about .003 rad/s were the subjects able to scale the relative velocity. The threshold subtended angular velocity obtained in the experiments was very much affected by the ability of subjects to use the concept of a ratio-engineering students found this a simpler task than did subjects from the general population. The result of this was that the values used by engineering students were closer to the real values. The relative velocity was perceived non-linearly, with a Stevens' power law exponent of about 0.8. It was found that linear models gave as good a fit to the data. The implications of the results of the experiments are (i) traffic flow models that include human visual characteristics must consider the 'dead zones' in response produced by thresholds of subtended angle change and subtended angular velocity. (ii) it may be necessary to consider the non-linear relationship between perceived relative velocity and actual relative velocity (iii) In overtaking, the driver will not be able to scale the speed of the oncoming vehicle as the subtended angular velocity will be below the threshold value a the time of making the decision to overtake.
    Novice and Experienced Drivers' Looking Behavior and Primary Task Control while Doing a Secondary Task BIBA 852-856
      Tapio Nieminen; Heikki Summala
    This study is addressed to time-sharing and primary task control during a secondary task as a function of driving experience. After about 1.5 h of test driving, when well-adapted to the experimental car, 23 novices (less than 5,000 km of driving) and 26 experienced drivers (more than 150,000 km) were asked to change a cassette in a cassette player on an ordinary two-lane road. The task was replicated three times. The results showed no difference between novice and experienced drivers in time-sharing (glance length at the in-car task and at the road), lateral position-keeping (lateral displacement as a function of time at in-car task) or control in relation to oncoming traffic. The only difference occurred in speed control, experienced drivers keeping their speed level constant while novices slowed down somewhat during the secondary task. These data showed, in a supervised experimental setting, a similar linear relationship between time spent on an in-car task and lateral displacement both for novice and experienced drivers, and a similar median time gap of about 2 s to an oncoming vehicle at the moment when both novice and experienced drivers shifted their gaze from the in-car task to the road.
    Driver- and Situation-Specific Effects on Assistance Systems for Speed and Distance Control BIBA 857-861
      Gunther Nirschl; Ralf Eck
    AICC (Autonomous Intelligent Cruise Control) systems, as investigated in the European PROMETHEUS and DRIVE projects, should support the driver in longitudinal vehicle control. The presented study analyzed driver- and situation-specific effects, which have to be considered when developing or evaluating AICCs. The investigations focused on assistance systems, which inform or warn a driver in case of inadequate speed or distance. Messages to the driver are transferred either acoustically or visually or via active control devices, such as an "active gas-pedal". Driving experiments were performed, whereby subjects' distance keeping behavior was monitored. As a first step before experimental evaluation, relevant traffic situations for AICC application were identified and classified. The results of the subsequent situation analysis reveal driver- and situation-specific characteristics. Typical effects are found in frequency and duration of various AICC situations, in distance distribution when following, and in minimum distance when approaching a preceding vehicle. From situation analysis, first consequences were derived showing off capabilities and limitations of AICC systems on principle and of an "active gas-pedal" in particular.

    SAFETY: Human Factors Issues in the Maritime Industry [Symposium]

    Human Factors Issues in the Maritime Industry BIBA 862
      Susan G. Hill; Anita M. Rothblum
    As a result of the Exxon Valdez accident in March, 1989, considerable attention has been focused on the role of the crew in maritime safety. It is generally recognized that human error contributes to the majority of the accidents in the aviation and nuclear industries, and recent analyses of marine safety data suggest a similar level in maritime applications. The problem areas also appear to be similar, i.e., fatigue and stress effects on performance, mismatches between levels of automation and skill, poor or inadequate procedures, and human frailty with regard to following procedures.
       This session presents four papers which present research and discussion of human factors issues related to work in the maritime environment. These papers present a breadth of topics, including the mariner's interaction with automation, issues of manning, methodology and results concerning "live-aboard" work concepts, and collection of human factors-related data from marine casualties. The papers presented are either directly sponsored by, or associated with, work being carried out by the U.S. Coast Guard, which is responsible for marine safety in U.S. waters. This symposium presents human factors research and applications in the marine environment; however, the issues and results may be useful across a wide variety of transportation and safety areas.
    Gathering and Recording Human-Related Causal Data in Marine and Other Accident Investigations BIBA 863-867
      Susan G. Hill; James C. Byers; Anita M. Rothblum; Richard L. Booth
    One of the missions of the U.S. Coast Guard (USCG) is to promote safety at sea through the prevention and mitigation of marine accidents. One means of preventing accidents is through the thorough investigation of the causes of marine casualties, the analyses of which can illuminate needed safety improvements. A study was conducted with the purpose of learning more about the USCG casualty investigation process, analyzing the data entry process, and making recommendations for improvements to the current computer system and the casualty investigation process. The study identified four major factors in the USCG system, particularly related to the reporting of human-related causes, which may have broad application to other safety professionals who contemplate the use of similar automated reporting and analysis systems. These widely applicable factors were:
  • the reliability and completeness of the data can be affected by the
       investigators' understanding of the purpose and scope of the accident
       database;
  • the collection of human factors data can be overlooked and oversimplified;
  • the taxonomy/classification scheme affects data collected as well as data
       reported;
  • the computer interface used for data entry can affect the reliability,
       validity, and completeness of the human factors data.
  • Mariner Performance Using Automated Navigation Systems BIBA 868-872
      Robin A. Akerstrom-Hoffman; Myriam Witkin Smith
    Electronic Chart Display and Information Systems (ECDIS) technology has recently emerged as a promising aid to maritime navigational safety and operational efficiency. ECDIS is likely to have multiple effects on the tasks bridge personnel must perform. However, careful consideration must be given to safety in adopting this new technology. A shiphandling simulator-based evaluation of some human factors aspects of the use of ECDIS is presented. The issues discussed include the effect on navigational safety and workload as a result of introducing ECDIS to the bridge, and the mariner's preference for and use of chart information and navigational data. Experienced mariners each made multiple port arrivals and departures as the lone watchstander on the bridge: navigating a planned route, responding to harbor traffic, and managing the preparations for annual or departure. During "baseline" transits, conventional methods were available to the mariner. During the test scenarios, one of two prototype ECDIS systems was also added to the badge. Under certain conditions, ECDIS reduced the mariner's workload for navigation. In addition, ECDIS showed a potential to increase safety as measured by a smaller cross-track distance from the planned route and by a larger proportion of time available to devote to collision avoidance and "look out".
    An Evaluation of a Coast Guard "Live-Aboard" Concept: Can Crews Adapt to a Restricted Living and Operational Environment? BIBA 873-877
      Antonio B. Carvalhais; Donald I. Tepas; Michael J. Paley
    The present evaluation was conducted to assess whether crews can cope and adapt to a restricted living and work environment. Two United States Coast Guard small boat stations with similar environmental conditions, mission profiles, and workload levels were selected for this evaluation. At one station, shore-side facilities were replaced with a 50-foot live-aboard boat (NORCREW). The second station, maintained shore-side facilities (POMP). Survey and daily log data on human factors variables, which have been used to predict adaptation to work environments, were collected from 18 crew members. Analysis of these data did not reveal any significant adverse psychophysiological effects associated with the live-aboard concept. Data revealed time-of-day effects consistent with conventional thought and chronobiological theory which leads the authors to conclude that the live aboard concept does not appear to disrupt circadian cycles. Overall, the consistent failure to detect any differences between NORCREW and COMP on human factors variables clearly suggests that further consideration of this concept is warranted.
    Identifying Clumsy Automation at the Macro Level: Development of a Tool to Estimate Ship Staffing Requirements BIBA 878-882
      John D. Lee; Jennifer Morgan
    Automation promises increased operating efficiency and suggests the opportunity to reduce the number of people required to operate commercial ships safely; however, clumsy automation may degrade safety and performance, rather than enhance it. This paper distinguishes between clumsy automation at the macro level and clumsy automation at the micro level and discusses macro level issues. Specifically, macro level clumsy automation refers to the failure to consider the broad implications of automation induced crew reductions. Clumsy automation may reduce workload and facilitate crew reductions during routine sailing, but these crew reductions may leave the vessel more vulnerable to unacceptable workload peaks during unusual and emergency situations (fires, unscheduled maintenance, rough seas, reduced visibility). In these situations, tasks not accommodated by the automation may overwhelm a crew that has been reduced by the introduction of automation. Currently, no systematic procedure exists to estimate the changes that automation implies for staffing levels and skill requirements. Avoiding the effects of clumsy automation depends on identifying techniques to broaden the study of automation to include the effects on the entire crew during all phases of ship operation. This paper describes a computer-based tool to help identify potential overload situations that would be difficult to anticipate through intuitive assessments of how automation and other factors affect crew requirements.

    SAFETY: Safety Potpourri II [Lecture]

    Cognitive Strategies and Their Transformations as a Factor of Control Systems Safety in Emergencies BIBA 883-887
      Valery F. Venda; R. Duane Shelton; A. A. Spikov; Cory Jessau; S. A. Chachko
    In accordance with the laws of ergodynamics (Venda and Venda, 1994), it has been suggested that, when the condition in a control system goes from a normal situation to an emergency situation, human performance efficiency dynamics should be wave-like, first demonstrating a decrease due to different functional-structures, (normal and emergency), then transform from one to the other. Special emergency experiments were organized and it was found that operators used three different cognitive strategies that transformed from one to the other. This process caused the wave-like dynamics of performance efficiency. Operator's performance efficiency first decreased and, only after cognitive strategy transformations were completed, the efficiency then increased. The processes were then reconstructed and studied at the regional training centers in Krasnoyarsk (Siberia, Russia) and Kiev (Ukraine).
    Components of Perceived Risk: A Reconciliation of Previous Findings BIBA 888-892
      Stephen L. Young; Kenneth R. Laughery
    People's perceptions of the risk associated with a product, in large measure, determines the degree of caution they will exhibit with regard to that product. Previous attempts to examine risk and to determine what factors are important in the composition of risk have produced disparate findings. Specifically, some studies have demonstrated that likelihood of injury plays an important role in risk perceptions (e.g., Slavic, Fischhoff & Lichtenstein, 1980), and other research shows that severity of potential injury plays the foremost role (e.g., Wogalter, Desaulniers & Brelsford, 1986; 1987). It was not until risk was conceived of in multidimensional terms that a coherent picture of risk seemed possible. The present study uses principal components analysis (PCA) to see if stable multidimensional solutions could be extracted from two qualitatively (and significantly) different item lists (as demonstrated by Young, Wogalter & Brelsford, 1992). Two sets of subjects rated the two different item lists on the same rating questions. The results were collapsed across subjects and submitted to a PCA. The solutions that emerged from the two item lists were strikingly similar. Each produced three components, with similar variable loadings and magnitudes. The results demonstrate dearly that risk, conceived of multidimensionally, is not affected by the products, technologies or activities under consideration. Rather, risk is a construct that could be tapped in order to give people a proper appreciation of the hazards associated with products and technologies that they encounter every day.
    Escape Worthiness of Vehicles with Passive Belt Restraint Systems BIBA 893-897
      Edmundo Rodarte; Jerry L. Purswell; Robert Schlegel; Richard F. Krenek
    There are a variety of conditions that can exist in the post-crash environment which make rapid escape necessary for survival or to avoid further injury. These include a post-crash fire, the vehicle going into the water, or avoiding being struck in a secondary collision. The National Highway Traffic Safety Administration (NHTSA) has defined this parameter vehicle escapeworthiness. It has been estimated in past research performed by the author for NHTSA that escapeworthiness becomes important in up to 7% of all vehicle crashes. Since escapeworthiness research was performed in the early 1970's, the advent of passive shoulder belt systems has made it necessary to again review the impact of this development on escapeworthiness. In particular, the inability of the occupants to release the passive restraint because the door cannot be opened after the crash, coupled with the inability to release the passive restraint due to its design or a lack of experience, or knowledge of how to release the passive restraint while the door is closed, creates a serious problem. Thus, the present study was performed to investigate the impact of passive restraint systems on the time required to escape from the vehicle under various conditions of available escape routes, and physical condition of the occupants. The experimental design included the variables of age, gender, escape route, level of incapacitation and type of passive restraint system. The times to effect an escape as well as the method of escaping were determined through videographic analysis of all escape trials. The findings demonstrated that the use of passive restraint systems increased the time to escape significantly, ranging from 37 to 65 percent for the respective conditions. This difference may determine whether a person survives or not after some post-crash conditions. The results have significance for the design of passive restraint systems for easy release, while at the same time not creating an incentive for some users to routinely leave the passive restraint unfastened.
    Oh! Say, Can You Hear that Train Coming to the Crossing? BIBA 898-902
      Rudolf G. Mortimer
    How effective is the horn on trains in alerting motorists, cyclists and pedestrians of the approach of a train to a highway-railroad crossing? Road users become aware of trains approaching crossings either by seeing them, hearing the warning horn or by warning devices at the crossing. Auditory warnings have some advantages over others, but practical problems of sound transmission and the noise environment of road users limit their effectiveness. Factors that affect the audibility of train horns and crossing bells are discussed. Train and crossing bells are generally effective for persons in their immediate environment, but are relatively ineffective for occupants of closed vehicles. Because of the lack of reliability of the detectability of auditory warning signals provided to motorists and others approaching rail-highway crossings, they should visually scan the tracks to ascertain the presence of a train.

    SAFETY: Perception and Understanding of Highway Signs [Lecture]

    Simulator Evaluation of Road Signs and Signals BIBA 903-906
      R. Wade Allen; Zareh Parseghian; Theodore J. Rosenthal
    This paper describes a accuracy versus speed paradigm for evaluating signing and traffic signal conditions using low cost simulation technology. Two research examples are reviewed. One study involved the use of an interactive driving simulator that included the presentation of high resolution signs over the apparent viewing range from 500 to 50 feet. Drivers had to control vehicle speed and lane position while identifying the meaning of symbol signs as rapidly as possible. Subjects were scored in terms of correctness and the distance at which signs were identified. A second study involved a computer controlled presentation of static signalled intersection scenes, including supplemental signs, to subjects who were required to make decisions about permissive movements. Subjects were required to make decisions about permissive movements as rapidly as possible, and were scored by the computer on correctness and response time. Results in both studies showed that both response speed and correctness degrade with the complexity of signal and sign treatments.
    Determining Legibility Distance for Highway Signs: Is the Within-Subjects Variability Being Overlooked? BIBA 907-911
      Frances A. Greene; Rodger J. Koppa; Ronald D. Zellner; Jerome J. Congleton
    Laboratory studies of warning symbol signs have been shown to underestimate legibility distances by up to a factor of two when compared with field studies. However, this research suggests it is more than simply experimental setting contributing to disparity in research findings. Using a group of old and young drivers, six symbol signs were investigated in both settings. With six trials per sign, legibility distances, defined as the distance at which the sign is correctly identified from a menu, were collected.
       Large within subject variability was discovered in both age groups. This variability led to alternative ways of defining the dependent variable equivalent to designs of past studies examining legibility distances of the same signs. Different results arose out of the subsets created. The consideration is not just should a field-based versus laboratory-based methodology be used. An argument is posed that recommended distances at which signs are placed must be determined from a "worst-case" scenario. This premise requires a reexamination of our research methodologies for determining placement of highway signs.
    Using the "Blur Tolerance" Technique to Predict and Optimize the Legibility Distance of Symbol Highway Signs BIBA 912-915
      Frank Schieber
    This study investigates the hypothesis that variations in symbol sign legibility distance can be accounted for on the basis of a sign's dependence upon high spatial frequency contours to convey critical information. Using digital image processing techniques, highway signs were blurred to remove all high spatial frequency information. A blur recognition threshold was established for each experimental sign by sequentially "deblurring" it until the observer could report the critical details defining its recognition criteria. Correlational analyses were then conducted to determine if legibility distance (collected in a previous study) could be predicted from the blur recognition threshold data. A significant correlation was observed between blur recognition threshold and sign legibility distance (r = -0.734, N=12, p < 0.001). That is, symbol signs with high levels of "blur tolerance" could be recognized at significantly greater viewing distances. These results support the application of new computer-assisted "recursive-blur" design techniques to optimize the effectiveness of symbol highway signs and related visual stimuli (see Schieber, Kline and Dewar, 1994).
    The Effect of Related and Unrelated Memory Loads on the Prerecognition Visual Processing of Traffic Signs BIBA 916-919
      Lloyd L. Avant; Alice A. Thieman; Michael W. O'Boyle
    Prerecognition visual processing of traffic signs was evaluated while subjects maintained one of four different types of memory load: low imageability nouns, traffic sign words, random shapes, or traffic sign shapes. Recall was uniformly high (mean=92%) and did not differ among groups. There was a highly significant interaction among groups (different memory loads), sign messages (Stop, Right, Left, Slow), and sign formats (symbol vs. word). Holding random shapes in memory eliminated prerecognition processing differences among sign messages for symbol format signs. However, for all other memory loads, differences among sign messages were significant. Tests across the memory load conditions for each format of each sign message showed that, for the Stop symbol, the Right symbol, and the Right word signs, the various memory load conditions produced no significant differences. For all other sign messages in both symbol and word format, there were significant differences among memory loads. In summary, these data show that the action message presented in traffic signs is being unconsciously processed within the first few milliseconds of visual processing, and that these operations involve unconscious activation of memory processes that store the meanings of various signs.

    SAFETY: Decision Making in Technical Environments [Symposium]

    Decision Making in Technical Environments BIBA 920-921
      Barry Strauch
    This symposium focussed on decision making in such technical environments as nuclear power plants, chemical plants, surgical operating rooms and airline cockpits. Despite their differences these share a potential of severe consequences that could result from faulty decisions by operators. Although highly skilled and trained to control the often complex machinery, operators are occasionally presented with situations that can lead to faulty decision making.
       Decision making is similar across various environments. The operator is presented with a situation calling for a decision. He or she must then identify the various alternatives available, weigh each, and select the one most likely to minimize cost and maximize benefit.
    The Organizational Context for Decision Making in High-Hazard Industries BIBA 922-925
      John S. Carroll
    The study of decision making in high-hazard industries has recently shifted from the traditional focus on control room operators to other actors in the sociotechnical system, including designers and management. I discuss the importance of hierarchy and specialization in organizational decision making, drawing on research in organizational psychology, sociology, and related disciplines.
    Quick Decisions in Two Diverse Settings BIBA 926-930
      Joseph P. Balkey
    This paper compares quick decisions during high hazard times in the operating rooms of a chemical unit and a hospital, finds commonalities and differences, and suggests changes to improve decisions within each industry. People within both industries decide quickly because they reduce options by screening information, receive numerous feedback from instruments within their respective environments to monitor the process, and training helps them overcome inconsistent symptoms. However, both industries can improve. As examples, the health industry internally monitors performance better than the chemical industry. The chemical industry operates with fewer people by using written rules and verbal communications whereas the health industry operates with more of a hierarchy with very little verbal or written communications.
    Developing Effective Flight Crew Decision Makers BIBA 931-934
      Elizabeth A. Lyall
    The need for including decision making training objectives in the development of flight crew training programs is evident given the dynamic and complex environment in which the pilots perform. The experience of one airline in developing training for their flight crews has demonstrated that these types of objectives can be effectively integrated into their line oriented flight training (LOFT) program. The primary goal of LOFT is to create a realistic environment in which the pilots perform as if they are in line operations. Performance in this environment sets the stage for an effective training debriefing. An essential part of creating this environment is allowing for realistic decisions to be made that are followed by the appropriate outcomes and consequences. Pilots have consistently rated these training events as effective and helpful. Besides these student critiques, the effectiveness of the training is also evaluated using check pilot observations of crew performance in line operations.
    Temporal Factors in Aviation Decision Making BIBA 935-939
      Judith Orasanu; Barry Strauch
    Time pressure has been found to reduce the quality of decision making by restricting consideration of information and options and by inducing strategy shifts. Time pressure is usually considered an external variable manipulated by the experimenter. In this paper we distinguish between externally-induced time pressures and crew-generated time pressures, and examine how crews can mitigate or exacerbate external pressures. The roles of both types of time pressures in crew decision making are examined in three air transport accidents and in crew performance in full-mission simulated flight. Implications for crew training are discussed.

    SPECIAL SESSIONS: Alternative Format

    Games for Explaining Human Factors: Come and Participate BIB --
      Ronald G. Shapiro; Megan L. Brown; Jody L. S. Wilson; Royce M. White; Mark J. Sugg; Roxann Adams

    SPECIAL SESSIONS: Demonstrations

    A Cost-Benefit Analysis Tool (CBAT) for the Management of User Interface Design and Evaluation Tasks BIBA 940
      Charles Bowen
    A user interface cost benefit analysis is an examination of the estimated costs associated with user interface design and evaluation tasks against the potential benefits to the software product. The Cost Benefit Analysis Tool (CBAT) is a human factors management tool written in SuperCard. The tool is based on research by Mayhew (1993) and Mantei and Teorey (1988). CBAT supplements previous research efforts by the addition of more user interface tasks during typical software development lifecycle phases. The main cost benefit form is completed by means of electronic worksheets which are used to estimate both the costs and the benefits used in the overall analysis. Both cost and benefit estimates can be tailored to individual organizational needs. The tool provides the capability to print out data used to perform the cost benefit analysis and on-line help is provided for all worksheets. Future enhancements to CBAT are suggested in the following areas: additional user interface cost tasks and benefit estimates, the saving of analysis data into individual files, the inclusion of a graphic scheduling capability, and improvements to the software performance and flexibility.
    Demonstration of Power: Performance and Objective Workload Evaluation Research BIBA 941
      Mark D. Rodgers; Carol A. Manning; Charles S. Kerr
    The Federal Aviation Administration (FAA) is developing a method to determine whether future air traffic control systems will provide the benefits to the National Airspace System (NAS) that were proposed when they were conceived. The purpose of this project was to develop a set of objective measures to characterize the productivity of an individual air traffic controller. Software was developed to compute measures of airspace characteristics, controller activities, and air traffic situational characteristics. This software, the Performance and Objective Workload Evaluation Research (POWER) program, computes a set of numerical measures based on routinely collected air traffic control data. The POWER program was written to interface with the Situation Assessment Through Re-creation of Incidents (SATORI) system, originally developed to re-create operational incidents (Rodgers & Duke, 1993). An engineering validation was conducted and a psychometric assessment is underway to evaluate the reliability, validity, and utility of the measures and a subset will be chosen to characterize controller taskload and performance. POWER will then be used to measure controller performance and taskload on ATC sectors to be transitioned to future systems. These baseline taskload and performance measures will be compared to taskload and performance measures obtained from future ATC systems after system implementation. POWER will also be used to evaluate alternative future systems display configurations at the Civil Aeromedical Institute (CAMI) Air Traffic Control Future Systems Simulation Laboratory.
    Test Planning, Analysis and Evaluation System (Test PAES): A Process and Tool to Evaluate Cockpit Design during Flight Test BIBA 942
      Scott L. Smith; Bradley Purvis; Stu Turner
    The basic Test PAES components include a Structured Crewstation Evaluation Process, a set of Structured Test Procedures (STPs) providing the "best engineering practices" for performing a particular type of evaluation, databases providing reference and process guidance information, the computer based tools to support the steps in the process.
    Cog-C: A Tool for Estimating Cognitive Complexity and the Need for Cognitive Task Analysis BIBA 943
      Michael J. DeVries; Sallie E. Gordon
    Task analysis is a methodology used during many different phases of system development. However, because many tasks involve complex cognitive processing, there is an increasing need for designers to perform cognitive task analysis. It is generally agreed that cognitive task analysis tends to be costly in terms of time and effort, and many designers ask how they would know when cognitive task analysis should be performed. This demonstration features a computer-based decision aid, Cog-C, to help designers answer this question. The soft care tool is based on the assumption that cognitive complexity is a major factor in determining when cognitive task analysis must be performed. The tool therefore helps the user determine the relative level of cognitive complexity for a set of tasks. It does this by (1) guiding the user in developing a task/subtask hierarchy, (2) guiding the subject matter expert in estimating the "amount" of various categories of knowledge required for subtask performance (e.g., concepts, rules, patterns, and strategies), and then (3) providing output including the number of steps required for task completion, standardized subscores showing the relative amounts of each knowledge category, an overall cognitive complexity score, and a general recommendation as to whether the task is a potential candidate for cognitive task analysis.
    Workload Assessment Monitor (WAM) BIBA 944
      Glenn F. Wilson
    The Workload Assessment Monitor (WAM) is a PC based system which has two functions: 1) It collects and provides on-line reduction of psychophysiological data, and 2) provides on-line classification of operator state (workload, fatigue, etc.). Each of the physiological measures can provide unique information concerning operator state and by combining these measures there is a pattern that has utility beyond the individual measures.
    An Improved Version of the Multi-Attribute Task Battery (MATB) and Data Reduction Program (MATPROC) BIBA 945
      Scott H. Mills; Kirby Gilliland
    This demonstration will present both an improved version of the Multi-Attribute Task Battery (MATB; Comstock & Arnegard, 1991) that provides markedly better sound capability while eliminating the need for a second microcomputer, and an associated user-friendly data reduction and processing package called MATPROC.
    Distributed and Dangerous Training Challenges BIBA 946
      Franklin L. Moses
    The conduct of training using actual equipment in field settings can be costly and even dangerous. This particularly is true for military missions such as air-ground (i.e., Close air Support) training. however, it also is true for training to fight fires, respond to disasters, control riots, and so on. This special session provides a demonstration and discussion about using simulation and simulators to supplement field practice. It points-the-way for realistic distributed training of dangerous missions.
    Introducing ACT (Activity Catalog Tool): A Software Instrument for Task and Behavioral Analyses BIB --
      Anthony D. Andre; Leon D. Segal
    RadView: A Preliminary Approach to Visualizing Occupational Radiation BIBA 947
      Douglas G. Hoecker; Timothy M. Lloyd; Harry Plantinga
    Visualizing three-dimensional distributions of radiation intensity, summed from multiple sources at a given work location inside a nuclear power plant, is a topic of interest to several prospective classes of worker in these plants. RadView is a concept for visualizing radiation that can be adapted to different users' applications once the core problem has been solved: how to effectively display this normally invisible phenomenon, while superposing the displayed data on the visually-cluttered environment that is typical of many work scenes. This demonstration presents the results of a preliminary feasibility study. At this stage, the results help more to clarify the problem than to propose a technical solution.
    Computer-Aided Systems Human Engineering: Performance Visualization System (CASHE:PVS) An Interactive, Hypermedia Design Tool BIBA 948
      Donald L. Monk
    The Performance Visualization Subsystem (PVS) of the Computer Aided Systems Human Engineering (CASHE) system is an interactive, hypermedia, ergonomics database developed for use by human-system designers, educators and researchers. CASHE:PVS version 1.0 allows users to access ergonomics data and models, stored on CD-ROM, as text, graphics, animations, and audio. The PVS hyperlinked data bases contain the complete implementations of the Boff & Lincoln (1988) Engineering Data Compendium and MIL-STD-1472D. Two specialized visualization tools, DataDigitizer and Perception & Performance Prototypes (P{cubed}), assist the user in understanding and applying this ergonomics data.
    TravTek Driver Interface and Camera Car Data Collection Capabilities BIBA 949
      Rebecca N. Fleischman; Thomas A. Dingus
    TravTek was the development and operational field test of an Intelligent Vehicle Highway System (IVHS). Drivers were provided with navigation and route guidance information, a local services and attractions directory, real-time traffic information and emergency services. Goals of the drivers/system interface design were ease of learning and use as well as minimizing the potential for distraction and visual and cognitive demands on the driver. A Camera Car Study collected detailed eye-glance and driving performance while participants navigated using TravTek route guidance visual and voice displays, paper map, and textual direction list. A data collection vehicle was developed for the study. This demonstration complements a paper by Dingus, et al, also presented al this meeting.

    SPECIAL SESSIONS: Posters

    Integrating Human Factors and Occupational Medicine: A Comprehensive Ergonomic Program BIBA 950
      Laurie Wolf
    To meet the challenge of rising health care costs and the transitions currently underway in United States health care systems, medical and rehabilitation professionals are seeking cost effective strategies to assist employers in preventing and managing injuries. Human Factors can make substantial contributions to the classic three component model of preventive medicine which includes primary, secondary and tertiary components. Injury prevention strategies include ergonomics, medical injury treatment, and rehabilitation services to provide a comprehensive occupational health program.
       Regulatory and reimbursement environments that guide the concepts of this integrative approach and specific program components will be addressed in the poster session. Future regulations and development of cost effective service packages will assist in changing corporate attitudes to better understand the long term value of ergonomics and the benefits of an integrated approach to managing costly work injuries.
    How Do People Answer Questions about the Inner Workings of Machines after Studying Illustrated Texts? BIBA 950
      William B. Baggett; Arthur C. Graesser
    We investigated the procedures subjects used to answer questions about machines after learning about the inner workings of the machines from texts and pictures. One goal of our research is to understand the role played by texts and illustrations when subjects construct mental models of machines. A second goal is to understand the mechanisms of question answering.
       Subjects studied illustrated texts describing causal event chains that unfold during the operation of everyday machines. After studying each illustrated text, subjects provided written answers to deep reasoning questions about the machine described. Each question referred to a particular event from the causal event chain. Four types of questions were asked: why did event X occur?, how did X occur?, what are the consequences of X occurring?, and what if X didn't occur?.
       In our analysis of the answer protocols, we tested two predictions of the QUEST model of question answering (Graesser & Franklin, 1990). First, QUEST's are search procedures specify the type of causal relationships between the queried event and answers to questions. Second, QUEST's intersection principle predicts that the presentation of information in multiple formats has a greater than linear impact on answer quality. That is, QUEST predicts that information coming from both the text and the picture occurs in answer protocols more than twice as often as its frequency in the total pool of information. Preliminary analyses support the are search procedures but not the intersection principle.
    Providing Team Feedback for Existing Training Events BIBA 951
      Maureen L. Bergondy; Jennifer E. Fowlkes
    The growing recognition of the importance of teamwork to overall performance in complex systems has led practitioners in many training domains to augment individual training approaches with team training approaches. Although team training is recognized as important, the state-of-the-art in team training technologies is immature
       This poster presents a methodology for incorporating team performance observation and feedback into training exercises that exist within an established training curriculum. Optimally, the augmentation of individual training with team training would focus on all components of training (information, demonstration, practice and feedback), however, implementation can be cost and time prohibitive. The approach described in this poster is focused on implementing observation and feedback into existing practice exercises, but it can be used to support a full-scale integrative approach.
       There are several reasons that the approach described in this poster may be useful. First, teamwork is critical for the successful accomplishment of many tasks. Second, existing practice exercises may incorporate many affects of teamwork implicitly. The training strategy proposed here would make such aspects of task performance explicit. Moreover, it would lead to standardization of team feedback, an essential component of successful training. Third, while it is important to provide "hands on" training, either live or with simulation, the cost and time investments are substantial, thus this training time should be optimized. The approach described here would entail augmenting this already existing training at a minimal cost, but at the same time incorporating into that training an essential component -- teamwork.
    Training versus Intuition: A Study of the Effects of Training on the Use of Ergonomic Chairs BIBA 951
      Sharon Broome; Lesia L. Crumpton
    In recent years, both researchers and manufacturers of ergonomically designed equipment have made large strides in convincing companies to allocate funding to purchase ergonomically designed equipment. However, to obtain the full benefit of providing ergonomically designed equipment, employees must be knowledgeable of how to properly use this equipment. Knowledge of how to properly use this equipment requires training on the mechanics of operation as well as information on how to properly position the equipment to capitalize on the advantages of its use. Comprehensive training programs that address both the mechanics of operation as well as proper positioning may increase the employees utilization of this equipment. Thus, the goal of this research was to evaluate the effects of comprehensive training on the utilization of ergonomically designed equipment.
       Twenty-seven participants were evaluated before and after receiving comprehensive training on the use of ergonomically designed chairs to determine the number and type of adjustment features utilized as well as their accuracy in positioning the chair. During the session, participants received both verbal and written instruction on the mechanics of operating each adjustment option available on the chair and information on how each adjustment feature should be used to maximize its intended benefit. Results of the study show that the incorporation of comprehensive training increased the number and type of adjustments made to the chair. Specifically, ten percent of the adjustment options were utilized before training and sixty percent after training. Also, comprehensive training affected the time spent making adjustments and accuracy of positioning of the chair. Fifty-four percent of the adjustments made before training resulted in correct positioning of the body while seventy-eight percent of adjustments made after training resulted in correct positioning of the body. Findings of this study, demonstrate that comprehensive training had a positive effect on the utilization of ergonomically designed chairs.
    Are Stereoscopic Displays Beneficial in Virtual Environments? BIBA 952
      Jennifer A. Ehrlich; Michael J. Singer
    An expensive component of Virtual Reality (VR) technology for training dismounted soldiers in distributed interactive systems is the use of stereoscopic visual displays. In VR systems, display screens are mounted into a helmet, situating them directly in front of the user's eyes. If the same view is sent to each screen, the result is a monoscopic display. However, different views can also be presented to each eye, reproducing some aspects of binocular disparity and stereoscopic vision. When a head-tracking system is employed, the scene shifts in conjunction with the user's head movements. This experiment compared the relative performance effectiveness of stereoscopic and monoscopic displays within Virtual Environments, as well as head-tracked and non-head-tracked interaction with the display. The tasks involved tracking, self movement, object manipulation, and distance estimation covering a range of psychomotor activities that are similar to tasks performed by armed forces personnel. Analysis indicates the stereoscopic display that simulates one's normal binocular disparity did not result in performance gains over monoscopic displays. Therefore, for these types of tasks, stereoscopic displays may not be necessary. Head-tracking also did not significantly improve user performance. The potential to scan fully one's environment will be discussed as an indispensable element in VR-based distributed interactive systems.
    The Use of Instructional Systems Design and Performance Analysis to Design and Evaluate Office Ergonomics Training BIBA 952
      Richard W. Goggins; Michelle M. Robertson
    A considerable amount of effort and monetary resources are invested in training programs in private industry, but little of this effort is directed towards evaluation of training effectiveness. This poster describes a field study evaluating the effectiveness of a training program in office ergonomics developed for a large aerospace firm using the principles of Instructional Systems Design (ISD). A four level analysis including measures of reaction, learning, behavior and results was applied to determine the effectiveness and potential benefits of the training. Forty employees whose primary function was computer work participated, 20 receiving training and 20 serving as a control group. Measures taken prior to training included an observational analysis of posture and work habits, postural discomfort surveys and a pencil-and-paper pre-test. Measures taken after training included a reaction survey, a post-test, and observational analysis and discomfort surveys identical to the pre-training measures. Reaction and learning results were both positive, while observation revealed that the experimental group was able to significantly reduce time spent in awkward postures. In addition, a significant decrease in overall discomfort was found in the experimental group. Data on absenteeism and productivity proved inconclusive, due to the small sample size. The control group did not change significantly during this period. Further analysis revealed that the trainees were somewhat hampered by the work environment in making desired changes, but overall, training helped them to overcome many of the deficiencies in their workstation arrangement and posture. A return-on-investment analysis revealed that the training program resulted in an acceptable payback period for the group trained. This study lends support both to ISD methods as an effective means of developing training and to performance analysis as a means of evaluating training effectiveness in a way that demonstrates the value of ergonomics training.
    Training Implications of Differential Elaboration in Learning to Troubleshoot Devices BIBA 953
      Christopher R. Hale
    Explanations can facilitate symptom-fault rule acquisition when learning to troubleshoot devices. For explanations to be useful in designing pedagogy, however, we need to understand what learning components benefit most from explanations.
       In this experiment, subjects first learned about a device and then learned eight rules characterizing common malfunctions. During rule learning, subjects received no explanations, explanations that elaborated item-specific information or relational plus item-specific elaborations. Subjects relearned the rules after a 24 or 48 hour delay. Cued recall accuracy decreased from 24 to 48 hour retest delays for subjects receiving no explanations and item-specific elaborations. Subjects receiving relational elaboration exhibited slight increases in recall accuracy over these delays.
       The results suggest that rule learning consists of two components, search and recognition, and that explanatory elaboration differentially affects these components. Recognition is crucial early in learning, so elaboration should focus on increasing associative strength. As associative strength decays, search becomes crucial. Elaboration which improves search efficiency will allow subjects to maintain performance across these delays. Training strategies should reflect these differential elaboration effects.
    The Effects of Physical Distribution of Team Members on Team Cohesiveness and Performance BIBA 953
      Carolyn M. Inzana; Ruth P. Willis; Steven J. Kass
    The advent of recent innovations in computer technology has enabled individuals who are distributed over great distances to work together as a team. The current study examined the effects of physical distribution of team members on their perceived cohesiveness and their overall task performance. Twelve three-member teams performed a radar simulation task in either a face-to-face or distributed training environment. The results indicate that face-to-face teams were more cohesive than teams whose members were physically separated from one another. Additionally, teams who performed in face-to-face environments outperformed those who were in distributed environments. The implications of these finding for distributed work teams are currently being explored.
    The Relationship among Eye Movements, Head Movements, and Manual Responses in a Simulated Air Traffic Control Task BIBA 954
      Donna J. Boyer; John A. Stern
    This study investigated the nature of eye and head movements during a two hour simulated air traffic control task. Changes in response related eye and head movements both within a session as well as across days of task performance were of interest. Ten subjects performed the task which consisted of 44 infrequently occurring events for which manual responses were required. There were four types of events; Unidentified Aircraft (UAC), Loss of Altitude (LOA), Conflict (two aircraft at the same altitude flying toward each other), No Conflict (two aircraft at the same altitude flying away from each other). The two hour session was divided into 3 blocks representing approximately equal lengths of time. The dependent measures were eye and head movement latencies with respect to manual responses. There were no significant differences between head and eye movement latencies made to the target as a function of response requirements. The relationship between the initiation of head movements and the initiation of eye movements is a stable individual characteristic; it is consistent between days as well as within the session. Timing of return saccades (returning gaze to the CRT) was task dependent, events requiring two manual responses showed different return saccade patterns than events requiring only one response. The return saccade associated with the first response occurred prior to making the manual response, while the return saccade associated with the second response occurred following the manual response. The following conclusions were drawn. Gaze control is affected not only by perceptual components but cognitive components such as depth of information abstraction required by the task. We also suggest that head movements are a likely indicator of processing demand with more head movements an indicator of more effortful processing.
    The Effects of Delayed Sensory Feedback on Object Recognition Performance: Uncoupling Perception and Action BIBA 954
      Bart J. Brickman; John M. Flach
    This work examines the role of action in perceptuo-motor performance. Two conflicting hypotheses emerge from the literature concerning this question. The reafference theory suggests that action is crucial to performance. Alternatively, the information theory suggests that action is important only when, and if, it provides an observer with differential access to information. These theories were assessed in an experiment that examined the effects of feedback delay on the performance of active and passive subjects using a dynamic occlusion methodology, similar to that reported by Flach, Allen, Brickman, and Hutton (1992). In general, feedback delay has been shown to be a cause of performance decrements in a wide variety of human controlled systems; furthermore, the degree that operators become impaired typically increases with the length of delay. The notion that performance is degraded in conditions of delayed feedback due to a "decorrelation" between control inputs and resulting sensory feedback is developed. Two alternative hypotheses concerning this "decorrelation" induced decrement were assessed. First, an information generation hypothesis suggests that the decorrelation induced by delays between control inputs and resulting feedback impairs the ability of active subjects to generate usable information, which causes degraded performance. Alternatively, the information pick-up hypothesis suggests that the decorrelation between control inputs and resulting feedback due to delay impairs the ability of observers to efficiently pick-up information that is available to them, which results in degraded performance. These hypotheses were evaluated by examining yoked pairs of active and passive observers. The data failed to reveal any statistically significant performance differences between modes. The results showed that the subjects' mode of interaction was clearly not important for performance in this task. Thus, the prediction of the reafference theory was not supported.
    Fitts' Law and the "Force" of Newton: A Match Made in Physics BIBA 955
      Mark A. Guisinger; John M. Flach; Amy B. Robison
    Two paradigms have found conflicting results in speed-accuracy tradeoffs for arm movements. In the first paradigm (Fitts, 1954), movement time is the dependent measure while accuracy (width) and distance (amplitude) are the independent variables. This paradigm shows a log-linear relationship between the dependent measure and independent variables. In the second paradigm (Woodworth, 1899; Schmidt et al., 1978) accuracy is the dependent measure while movement time and distance are the independent variables. This paradigm shows a linear relationship between the dependent measure and independent variables. Both of these paradigms use the same variables (i.e., accuracy, distance, and movement time) but they suggest two different mechanisms for control depending on the task being performed.
       These two paradigms focus on speed (velocity) as the key kinematic variable (i.e., speed-accuracy tradeoff). The present model uses acceleration as the key kinematic variable (i.e., force-accuracy tradeoff). Using the acceleration model the results of Fitts' (1954) experiment can be predicted, using parameters derived from the Schmidt paradigm and the widths and distances from Fitts' (1954). The prediction from our model only accounts for the movements that "captured" the target on the first submovement. A Monte Carlo simulation was created to model multiple submovements. This simulation predicts movement times and the number of submovements to "capture" a target.
    The Role of Action in Perception BIBA 955
      A. B. Robison; R. J. B. Hutton; J. M. Flach
    Mode (active vs passive) has often been thought to play an important role for situation awareness. Active pilots are considered to be more in tune with the situation and better able to respond to emergency events. However, it is very difficult to manipulate mode independently from information. An alternative hypothesis is that it is not mode per se, but information that is the critical factor that determines performance (e.g. Welch, 1986). This experiment used a dynamic occlusion paradigm to unconfound mode and information. The recognition and identification of computer generated three dimensional wire frame objects from dynamic occlusion was used to investigate the role of action in perception.
       This experiment manipulated the way in which the information was generated. Recognition performance of subjects who manipulated the objects themselves was compared with performance when the object motions were computer-generated and stereotypical. It was expected that the exploratory actions and subsequent information generated by an active subject would provide better recognition than the information from the stereotypical motions of the objects. The results showed that earlier performance was better with the stereotypical motions than the active exploratory motions. Microanalysis of movement strategies suggested that subjects begin to constrain their actions in order to provide a higher signal to noise ratio, aiding recognition performance. Further, computer-generated random object motions, and a condition in which the active subjects were constrained in their actions by the input device were investigated. The random computer-generated motions were expected to provide noisy information and poor recognition performance, similar to an inexperienced active subject. The constrained active subjects were expected to produce more useful information similar to that produced by the consistent computer-generated object motions. The results were consistent with these expectations.
       The results suggest that the recognition performance in this task was not a function of observer mode (active versus passive) per se, but was a function of the signal to noise ratio of the information provided by the object manipulations.
    Distortions in Memory for Line Graphs BIBA 956
      H. Scott Kinslow; Andrea M. Lonon; C. Melody Carswell
    Graph users and designers often argue that graphs aid memory for quantitative information. But what kind of memory exists for graphs? Is memory for graphs, like that for maps, characterized by systematic distortions? In the present study, subjects studied simple time series (line graphs) for 20 seconds before drawing them from memory on blank axes. Stimulus graphs contained either one or three angles. Although angle sizes were matched across graphs, the angles could actually represent two types of trend changes: monotonic rate changes or actual changes in direction (from increasing to decreasing, or vice versa). The number of angles in the stimulus had no reliable effect on the accuracy of the angle sizes recalled by subjects, but the type of trend represented by the angle did. Rate changes showed little systematic distortion, while direction changes were consistently exaggerated (i.e., the angle was underestimated). Thus, the meaning of the graph, rather than its physical structure, was associated with obtained errors. These distortions should not be viewed as a failure of memory, but as an attempt by the information processing system to simplify complex patterns.
    Design Characteristics for Foveal and Peripheral Tasks in Multi-Task Visual Displays BIBA 956
      Virginia A. Lang; Michael Keith; Andrew Kavie
    The complexity and size of visual displays is ever increasing with the advent of new technologies, the information highway, etc. The way these displays are designed must take into account not only aesthetic qualities but also the constraints of the human visual system and the demands of the displayed tasks. The present experiment takes into consideration both these constraints by examining performance for a foveal and a peripheral continuous monitoring task which is presented in either color or monochrome. The results indicate that multi-task displays which require high task demands and both foveal and peripheral vision can be performed concurrently with no performance tradeoffs when they are designed properly. Namely, peripheral displays should be designed in monochrome and foveal displays in color for high task demand situations, but low task demand situations do not require this design consideration. Future research is needed to determine if these findings are relevant for other types of visual tasks, such as discrete tasks.
    Modeling Vibrotactile Sensation: Using Factors Commonly Associated with CTS BIBA 957
      Lesia L. Crumpton; Jerome J. Congleton
    Vibrometry threshold testing has proven to be a viable screening method for detecting the presence of Carpal Tunnel Syndrome (CTS). By identifying which of these hypothesized CTS risk factors appear to be closely related to vibrometry testing values, inferences may be drawn that will assist in evaluating the relationship between these factors and the presence of CTS. Therefore, the purpose of this research was twofold: (1) to identify the best subset of variables that would accurately describe vibrometry testing values using factors commonly associated with CTS; and (2) to generate a model to explore the relationship between this subset of variables and vibrometry threshold testing values.
       Data on vibrometry thresholds, anthropometric measurements, and risk factors associated with CTS were collected using 17 participants (34 hands). Information collected on personal and occupational risk factors and anthropometric measurements served as independent variables in this study; while, vibrometry testing results were used as the dependent variable in this analysis. The All Possible Subsets Algorithm was employed to identify the best subset of risk factors closely related to vibration sensitivity thresholds. Multiple regression techniques were used to generate a model to examine the relationship between the best subset of risk factors and vibration threshold values.
       Results of this evaluation indicate that gender, weight, personal index rating (developed using medical history, medications taken, and personal habits), wrist circumference, time spent weekly in repetitive leisure activities, and number of hours worked daily is the best subset of variables closely related to vibration threshold values. The Multiple Regression model developed using this subset of variables revealed that a significant regression relationship does exist between these variables and vibration threshold values (Prob.>F = .0001 and R² = .66). Significant variables in the regression model include gender, number of hours worked daily, time spent weekly in repetitive leisure activities, and personal index rating.
    OSHA Ergonomic Regulations: Expected Costs and Anticipated Actions by Industry BIBA 957
      Joseph R. Davis
    Multi-attribute decision analysis is presented as a method for anticipating actions by industry in response to Occupational Safety and Health Administration (OSHA) proposed ergonomic regulations. This paper begins by describing the proposed regulations and the expected costs for industry compliance or expected penalties for non-compliance. Next, multi-attribute decision models and methods are briefly explained in sufficient detail to allow understanding how decision analysis can be applied to ergonomics. Then, a summary is given of findings from a literature search concerning industry attitudes and two basic action strategies of (1) regulatory compliance or (2) intentional avoidance. Within each of these two strategies, there are specific action alternatives ranging from extreme compliance by fully implementing ergonomics to extreme avoidance by doing nothing (and risking OSHA fines) with intermediate alternatives of performing perfunctory analysis, using temporary workers, automating production, moving production out of U.S.A., segmenting the company, or exiting the industry.
       An influence diagram is presented to visually show how the overall decision involves OSHA's history of inspections, industry's action alternatives, and possible outcomes for attributes of cost, quality, and intangibles. Based on the literature findings, preferential weightings for rational decision makers are defined for attributes associated with the two most important tangible objectives (cost minimization and quality maximization) and other intangible goals ("highly" flexible production, "good" public image, "high" employee morale). Then, these weightings are used in a multi-attribute decision table that shows expected costs and anticipated actions by industry in response to the proposed ergonomic regulations. Results indicate that rational decision makers in low-profit small companies will opt for an avoidance strategy. These results were confirmed by limited validation in personal interviews with small business owners. In conclusion, combinations of new positive motivators (tax credits, free education, videotapes, etc.) along with existing fear motivators (fines, etc.) are suggested for OSHA to induce small companies to opt for compliance.
    The Effect of Hand Posture on Force Exertions during Sewing Task BIBA 958
      Adrienne Drohomirecky; Kentaro Kotani
    The sewing industry exhibits a high incidence of cumulative trauma disorders, specifically carpal tunnel syndrome. The workload of sewing machine operators consists of risk factors such as high force exertion, repetition, awkward posture, and lack of recovery time. A typical sewing task uses a pinch grip, usually accompanied by a deviated wrist, to exert precise control on a material while directing it under the stitching needle. Since sewing tasks are associate with high frequency and forceful exertions, a posture allowing sewers to complete sewing tasks with minimal force exertion is optimal. The effects of wrist posture, pinch type, and pulling angle on force exertions required to complete a sewing task were investigated in this study. The force exertions were based on the percentage of maximum voluntary contraction and the percentage of maximum contact force required to complete a sewing task. Three pinch types were tested in combination with three wrist postures and two pulling angles. The experimental conditions included lateral, chuck, and palm pinches, neutral, flexed, and extended wrist postures, and 0° and 30° pulling angles. Subjects were instructed to exert minimum force while simulating a sewing task on a wool material. Electromyography (EMG) measured muscle activity and Force Sensitive Resistors measured contact force distribution between fingers. An on-site validation was also completed. This study is significant due to its applications to the sewing industry. Based on normalized EMG and normalized contact force, optimal posture was determined to be neutral wrist, lateral pinch, and 0° of pulling angle. The results of the study may be used to establish proper guidelines for work requirements and job design in industrial sewing to reduce cumulative trauma disorder risk.
    An Electromyographic Study of Maximum Torques and Upper Extremity Muscle Activity in Simulated Screwdriver Tasks BIBA 958
      Daniel J. Habes; Katharyn A. Grant
    Muscle force exertion during work, a risk factor for the development of upper extremity musculoskeletal disorders, is difficult to estimate based solely on visual observation of the working posture. The purpose of this study was to develop quantitative muscle tension data associated with exertions in certain postures, that could be used by job analysts to rate the muscular effort required to perform manual work tasks in similar positions. Fifteen subjects were instructed to perform a series of manual screwdriver tasks, similar to those observed in industry. Muscle activity in six upper extremity muscles was measured using surface electromyography (EMG). A repeated-measures design was used to examine changes in EMG associated with changes in the following task parameters: (1) torque level, (2) handle height, (3) distance of the hand from the body, (4) handle diameter, and (5) tool orientation. The results indicated that subjects' torque capability was maximized when participants used the larger handle in the vertical orientation. There were no consistent relationships between EMG in any muscle group and torque strength, indicating that increased muscle activity did not necessarily translate into greater torque output. Significant changes in EMG were observed, however, with changes in handle position. This study supports the premise that workstations and tools should be configured to allow workers to adopt postures that maximize their capabilities while minimizing muscular effort. Furthermore, this study provides data that can be used to grade exertions associated with work in varying postures, and to justify workstation modifications intended to reduce biomechanical stress on the upper extremities.
    Vision Screeners: A Comparative Evaluation for Display Screen Users BIBA 959
      T. Horberry; A. Gale
    Recently introduced European legislation makes provision for regular VDU users to be offered vision screening by their employers. Various screening devices exist and this research investigated how accurately currently available systems compared in their measures of suitable visual function, both to each other and to an Optometric eye examination. Four screeners were stand alone systems (Titmus, Keystone, Optec, and Ergovision), and two were entirely software based (City and Vutest). Over 200 subjects were screened under controlled conditions. For the software systems data were obtained additionally at the users' own workstations. Usability measures were also recorded using video and questionnaires.
       The overall pass rates obtained by the Optometrist were found to be significantly correlated with the overall results of three of the screeners at the intermediate range. Whilst no overall statistical difference resulted between the pass rates for the six vision screeners. Additional comparison with the Optometric data demonstrated a significant correlation between the stereopsis tests performed by five of the screeners and the Frisby Stereo Acuity test. All the screeners that tested acuity at distance had a significant association with the results obtained through the related Optometric tests. In terms of Ergonomic measures then significant differences in the subjects' ratings of the ease of use for the screeners were found. Reliability analysis demonstrated that only two systems produced statistically significant results. The workplace comparison for the two software systems demonstrated that both screeners had a significant association between their overall results obtained in the laboratory and those at the users' normal workstations.
       The various advantages of the software based and the stand alone systems in comparison to the standard optometric eye examinations are detailed.
    The Role of the Keyboard Design in Performance, Musculoskeletal Discomfort, and Fatigue BIBA 959
      Christopher S. Pan; Naomi G. Swanson; Traci L. Galinsky; Libby L. Steward
    The present study compared an alternative keyboard (Keyboard X) with the standard keyboard on measures of performance, discomfort and fatigue. Twenty female clerical workers were recruited from a temporary employment agency for participation in the study. Subjects performed a text-entry task for three consecutive days in a simulated VDT office environment. All subjects worked on a standard keyboard on the first workday, then were assigned randomly in groups of one or two to Keyboard X or the standard keyboard for testing during the two following workdays. Ten subjects were assigned to each of the two conditions. Self-ratings of musculoskeletal discomfort and fatigue were assessed at the beginning and end of each day and before and after each mid-morning, lunch and mid-afternoon break. Performance data (keystrokes/hour and errors/hour) were collected on a continuous basis throughout each workday. The results indicated that overall performance in terms of keystrokes/hour was significantly lower for Keyboard X. Performance dropped significantly from baseline levels when the subjects began typing on keyboard X, but returned to baseline levels by the end of the study. The results also indicated a significant increase in fatigue and musculoskeletal discomfort over each workday for both keyboard conditions.
    A Case Study of Factors Influencing the Rejection of Scissor Lifts for Box Palletizing BIBA 960
      Carol Stuart-Buttle
    Scissor lifts are a common recommendation in industry to reduce stresses of manual materials handling (MMH). This case study investigated why some lifts installed by a meat processing plant for a conveyor to pallet MMH task were not being used by the employees despite the apparent improvement in the workplace. The Lumbar Motion Monitor (LMM) was used to assess tri-axial trunk motions and the velocities and accelerations associated with the job. The software of the LMM uses a low back disorder (LBD) risk prediction model developed by Marras et al. (1993). Three palletizing conditions were evaluated: to the floor, at a scissor lift and at a modified scissor lift. Near and far box positions at three levels of the floor palletizing task and three box positions at the scissor lift were assessed. Analysis was a combination of observations, the probability of being a member of a "high risk" group for LBD as defined by the Marras prediction model and the NIOSH (1991) lifting index. Results showed the original scissor lift workplace had a higher probability of LBD risk than floor palletizing due to several subtle barriers and workstation function problems. These barriers and problems increased maximum sagittal flexion and maximum lateral velocity during lifting. However, the NIOSH lifting index suggested an improvement of the scissor lift over the floor palletizing task as the guideline is not sensitive to dynamic characteristics. After modification, the scissor lift workstation had a risk comparable to the middle level of floor palletizing and the changes reduced the higher risk associated with installation problems and low level MMH of the floor palletizing. The case study also indicated potential further studies of trunk motions with alternative assistive devices and different starting heights in MMH.
    Jigs and Fixtures: Design Review Guidelines for the Ergonomics Engineer BIBA 960
      Maryanne Townes
    Traditionally, the ergonomics engineer has not played an active role in the design of jigs and fixtures used to manually assemble parts. This is problematic, since tool redesigns are often infeasible due to cost and production constraints. Therefore, less effective measures must be used to control ergonomic hazards, since the hazard(s) cannot be engineered out during the design process. References, guidelines and checklists used by many tool designers do not include ergonomic engineering principles and methods. In addition, many ergonomics engineers may not be schooled in the elements of jig and fixture design, which may reduce their effectiveness as part of the tool design team. The purpose of this paper is to provide the ergonomics engineer with an overview of the design process, and proffer guidelines to assist in the review of jig and fixture designs.
    Optimal Grip Force Work/Rest Patterns Developed from a Model to Predict Localized Muscle Fatigue during Maximal and Submaximal Work BIBA 961
      David D. Wood; Donald L. Fisher; Robert O. Andres
    Twenty college age women performed one hour of isometric gripping and relaxing using 3 equivalent patterns, low force long duration, medium force medium duration, and high force short duration. The amount of physiological work was held constant during all conditions. The smallest grip force decrement was seen in the medium force condition.
       Previous researchers have reported that changes in force during an isometric contraction held to exhaustion or a series of brief MVCs can be modeled using negative exponential functions of time. Here, several models were developed which predicted the change in peak and ending grip force during both maximal and submaximal contractions. Parameter values were found for each model which minimized the sum of the squared difference between the predicted and observed grip force. The final model explained 94.2% of the variance in the group data. The parameter values for the final model were used to predict the work/rest pattern which has the smallest grip force decrement. The total physiological work was held constant in each pattern by calculating the product of the duration (in seconds) and the exertion level (in percentage of MVC). A pattern of 26% MVC for 23 seconds and resting for 37 seconds yielded the smallest and therefore best grip force decrement. This finding may have significant implications in the ergonomics debate on the importance of contraction force and frequency and the development of cumulative trauma disorders.
    Preliminary FAA Human Factors Design Standard for Airway Facilities (HFDS-AF) BIBA 961
      Joseph A. Birt
    This poster session addresses an on-going R&D project, its preliminary product (a draft standard), its current status, and its future potential.
       In keeping with the 1994 HFES meeting theme, the session and product illustrate a design standard project that focuses upon fundamental human factors contributions to system engineering and system development for the sponsoring agency. For instance, the preliminary document is oriented to remote maintenance control as well as to hands-on maintenance. It has: some unique topics for the FAA (e.g., user documentation, and system security), a comprehensive human-computer interface section, and "human-factored" features and format. The resultant document captures years of research and applications experience.
       The initial edition is intended for acquisition of new or off-the-shelf FAA systems and equipment that are to be operated, managed, or maintained by the Airway Facilities organization. Such a standard is one tool for applying human factors and is not a substitute for professional Human Factors efforts on the government and contractor sides of system procurement.
       The session addresses "where the document stands" in the review process. The session treats the future potentials for: updating the document with emerging technologies, identifying research to fill gaps, developing an electronic version, and developing possible editions that address Air Traffic and Flight Standards concerns.
    The Effects of Speech Intelligibility in a Simulated Industrial Process Control Task BIBA 962
      David G. Payne; Darrin M. Conant
    Payne, Peters, Birkmire, Bonto, Anastasi, and Wenger (in press, Human Factors) used a dual task paradigm and varied speech intelligibility levels in an auditory task while subjects performed concurrent visual tasks. They found significant cross-modal interference effects but only in tasks that involve the central processes of working memory and decision making. The present research assessed cross-modal interference in a more complex decision making environment. We used the NASA Multi-attribute Task battery, which consists of three visual tasks (choice reaction time, compensatory tracking, complex decision-making) and one auditory communications task. Subjects performed in both single and dual task conditions, with the dual task conditions always including an auditory task in which we varied the intelligibility level (as measured by the Modified Rhymes Test) of the speech signal. Replicating Payne et al. (in press) there was no effect of speech intelligibility on performance in the compensatory tracking task, which does not load heavily on working memory or decision making, but there were effects of speech intelligibility in the reaction time and decision making tasks. These results demonstrate that the same pattern of selective interference effects reported by Payne et al. (in press) can be obtained with different visual tasks and under conditions that more closely mimic the task demands of many real-world situations. Finally, this study represents an important conceptual link between our earlier laboratory research (Payne et al., in press) and the simulation research of Whitaker, Peters, and Garinther (1989).
    Understanding the Visual Impact of VDT Viewing BIBA 962
      Cosmo Salibello; Jon Torrey
    The use of visual displays (VDT -- Video Display Terminal) in modern information processing environments is associated with the onset of computer eyestrain complaints, such as blurred vision, double vision, color fringes, dry eyes, burning eyes, frequent changes in eyeglass prescriptions and pain in the area of the neck and shoulders.
       Available research supports the contention that the image formed on a pixel-based display (VDT) has a light amplitude distribution that is Gaussian in nature; i.e., such an image is brightest at its center and less bright toward its edges. Thus, the Gaussian image is inherently out of focus since its borders are not sharp and distinct, since it lacks good figure-ground contrast and since the high spatial frequency information required for accurate focus is absent.
       Additional research suggests that human accommodation reacts to viewing Gaussian images in much the same way as it would in a featureless field. Under such conditions, focus is driven toward a resting point of accommodation (RPA). The eye is forced to make repeated efforts to regain focus in a vain attempt to maintain a clear picture. Estimates that from 20,000 to 30,000 shifts in focus can occur in an 8-hour span of VDT viewing lend credence to the volume and severity of worker complaints.
       A technology exists which duplicates, for an eye doctor, the accommodative demand of a typical visual display. Doctors using the technology can diagnose and treat the computer eyestrain complaints of their patients under the same conditions that brought on the symptoms.
       This poster explores the visual ergonomics of typical VDT working environments, the nature of Gaussian imagery as used in visual displays, the effects of existing office conditions on VDT workers and the suggested strategies both for improving those conditions and for seeking professional eyecare services when appropriate.
    The "Purple Haze" of Nonsignificant Results BIBA 963
      Rebecca W. Miller; Keith M. Rettig; Mark W. Scerbo
    The present study was designed to evaluate several methods for conveying statistical significance in graphs. One objective of the study was to evaluate two methods for depicting variance. Specifically, traditional error bars derived from the standard error (SE) were compared to an alternative format using a pooled estimate of variance. The "pooled" error bars have an advantage in that judgments of statistical significance between means become a simple matter of inspecting the figure for regions where error bars do or do not overlap. Several styles of error bars were created (bidirectional, displayed on both sides of the means; and unidirectional, displayed on the same side or opposite side of the means). In addition, "cloud" displays were created in which the variance about each mean was represented by a different colored (red or blue) rectangle. Thus, a purple region appeared where the two rectangles overlapped. Two different sets of clouds were evaluated, one depicting means and one without. A tabular display in which the means, SEs, and sample sizes provided in numerical format was also studied. The error bar styles were combined factorially with two methods of deriving SE (traditional and pooled). Multiple instances of each type of figure were created at 4 different probability levels, two levels each depicting both significant and nonsignificant relationships. Subjects were asked to judge whether the figures or tables represented statistically significant relationships. The results showed that the traditional method produced more than 3 times as many errors as the pooled method. Moreover, these differences were exacerbated for judgments of relationships centering around the .05 probability level. Both sets of cloud displays produced performance levels that were equivalent to those of the bidirectional and unidirectional opposite displays. By contrast, the tabular display produced the most errors and longest response times. These results highlight the advantage of using the pooled estimate of variance for conveying statistical significance in graphs. Further, the novel "cloud" error bar style was shown to be an effective method of depicting variance. As multimedia technology matures, clouds may prove a viable alternative to traditional error bars.
    Visual-Task Performance Can Benefit from Secondary-Task Demands BIBA 963
      David A. Washburn; R. Thompson Putney
    An experiment was conducted to determine whether attention-switching times would predict visual performance in dual-task environments. Thirteen subjects were tested on a battery of five tasks, where the tasks were presented in random order either alone or in two-task combinations. Response times were not significantly longer for the lexical decision task, the mental rotation task, and the relative numerousness task when paired with a short-term memory task that required continuous vocal rehearsal (mean response time 1106 versus 1026 in the dual-task and single-task conditions, respectively). More surprisingly, however, subjects responded significantly faster on the lexical decision, mental rotation, and relative numerousness tasks when the tasks were embedded within the pursuit tracking primary task than when they were presented alone, F(1, 144) = 9.44, p < .01. For example, subjects made lexical decisions significantly faster when they had to stop tracking, switch attention to an array of letters, and decide whether the letters constituted a word than when the same task was presented but no tracking was required. Across tasks, response times averaged 1057 msec when tasks were performed individually, but only 909 msec when subjects performed the tasks while pursuit tracking. These data are consistent with previous reports of trial-initiation or procedural difficulty effects. Increasing the difficulty of initiating a trial can elicit a shift in attention that will result in significant improvements in performance. These findings imply that training may be most efficient under conditions of high procedural difficulty, and that vigilance-type tasks might produce better performance if they included a demanding, continuous-response component to elicit and maintain the focus of attention. [Supported by ARI grant DAAL03-92-G-0382, NASA grant NAG2-438, and Georgia State University]
    Task Re-Categorization for Reducing the Attentional Demands in the Cockpit BIBA 964
      Daniel P. McDonald; Richard D. Gilson
    With the increasing sophistication of aircraft technology, there becomes a growing demand on attentional resources within the cockpit. Parallel processing or "chunking" is considered to be one solution for reducing attentional workload demands. Some flight tasks may be effectively re-categorized for situations when the differences between tasks of a selected group need not be emphasized, (e.g., poor weather conditions and restricted air space being singularly displayed as "airspace to avoid") which can allow for a reduced workload demand. With a re-categorization strategy the pilot may maintain optimal flight performance, and with reduced attentional demands may better notice unexpected stimuli.
       The purpose of this experiment was first, to demonstrate the effectiveness of "task re-categorization", wherein a human operator can simultaneously process more than one stimuli, and second, to determine if this strategy will allow operators to more easily notice unexpected stimuli while performing their task(s). A modified version of a Neisser and Becklin (1975) experiment was performed using a superimposed video containing two event sequences and other unexpected stimuli. Subjects in the (separate) group were forced to attentionally separate the event types, whereas the (combined) group was forced to attentionally combine the two event types, while maintaining a choice response condition for both. Mean responses times were faster and response accuracy was greater for the combined group. Following the experiment the combined group also recalled more of the unexpected stimuli than did the separate group.
    Designing a Better Weapon System with a Human-in-the-Loop Simulator BIBA 964
      Laura Schmid; JoAnna Weyer
    This poster describes the human factors evaluation tool Texas Instruments developed to study the concepts of the Pre-Planned Product Improvement (P{cubed}I) for the Joint StandOff Weapon (JSOW). The improvement program will add a pilot-controlled seeker to the existing glide bomb launched from a variety of aircraft. Our evaluation tool is a simulator that helps the systems engineers study the impact on the pilots. The engineers explore various design concepts and scenarios by measuring system performance using actual pilots in the loop and collecting feedback from the pilots.
       Design concepts explored with the tool include the format of the imagery (the core of the design), human-in-the-loop input delays due to transmission and processing of data between the JSOW and the aircraft, filter effects for antijamming purposes, and datalink effects on the imagery when a signal has dropped below a threshold point. The evaluations will help engineers tune their design prior to implementation. This early design analysis saves the contractor and customer valuable time and money. We needed a practical method for human factors studies with a complex system. The human-in-the-loop simulator was our cost effective solution.
    A Preliminary Study of Aircraft Accidents and Incidents Attributed to Complacency BIB --
      Jennifer J. Vardaman; Stephen J. H. Veronneau; Bruce C. Wilcox
    Development of a Form Fill-In Package: A First Step Towards a Paperless Environment for Graduate Program Administration BIBA 965
      Sameer Bhagwat; Tushar Phondge; Anand Gramopadhye; Joel Greenstein
    The competitiveness of business in all branches of the economy depends on office sector productivity. When all the information is on paper, inevitably corrections have to be made to produce the final clean copy. This process is time consuming. This problem was addressed in the Graduate School at Clemson University by developing a computer-based form fill-in package as an alternative to filling out the forms using a typewriter. This paper presents the initial survey and the detailed task analysis that were performed to establish the need for developing a form fill-in package. Other alternatives, such as redesigning the form for better entry using a typewriter, and using a word processor to fill out the forms were also considered. It was found that the computer-based form fill-in package was the best alternative for automating the process. This paper also discusses the behavioral goals that were established for the software package and the results of user testing. Comparisons between the original system and the computer-based system are also made. Future modifications in the form fill-in package, such as the provision of electronic signatures to allow electronic mailing of forms, are also discussed.
    The Effect of Static and Dynamic Whole-Hand Gestures on the Perception of Grasped-Size in a Virtual Environment BIBA 965
      J. K. Caird; P. A. Hancock
    Virtual environments (VE) afford a user sufficient capability to meaningfully interact with computer created, three-dimensional representations. Interaction with virtual environments through whole-hand input devices is a frontier of human-computer interaction research. The emphasis of this research was the haptic (active touch) and visual perception of hand input to virtual representations. An experiment is reported that tested whether gestures rather than direct grasping of virtual cylinders affected the accuracy of perceived size. When objects could be grasped with a closed fist gesture and manipulated, the perception of object size was significantly more accurate than when they could not. The results provide insight into how people actively seek clear perceptual information when limited haptic information is available. The results are framed using the theoretical concept of affordance (Gibson, 1977). A discussion of the usability of a specific whole-hand input device is also given.
    Correlates of Route Traversal Performance in a Virtual Environment BIBA 966
      John P. Gildea; John H. Bailey
    Virtual environments (VE) are the manifestation of efforts to use vision, audition, touch, and olfaction to seamlessly interface with computer generated environments. Despite their departure from traditional computer interfaces, some VE interfaces resemble more traditional interfaces, such as a video game joystick. When VE interface devices are similar to those used by arcade and home video games or personal computers, an individual's experience with these devices may be expected to correlate with their performance in a VE. Because information of a more quantitative nature (e.g., years of computer experience) used in pervious research did not correlate highly with performance, we developed three self-report subjective measures for possible use as covariates; 1) enjoyment of video games; 2) skill at playing video games; and 3) confidence in using computers. These three questions were administered as part of a research effort to investigate spatial knowledge acquisition in a VE. Several variables were used to measure performance during VE route rehearsal, as well as in a test phase conducted at a traditional computer workstation. Among these variables were wrong turns, journey time, distance traveled, collisions with objectives, time colliding with objects, time in rooms, and time on floor 1, on floor 2, and on floor 3. During VE rehearsals, collisions and collision time were significantly related to enjoyment, and time on floor 2 was significantly related to skill at playing video games. During the test phase, time in stairs and time on floor 2 were significantly related to enjoyment, and journey time was significantly related to skill at playing video games. A principle components analysis (PCA) on the performance variables revealed two primary components that may represent route knowledge and psychomotor abilities.
    Difficulties in Detecting a Missing Target BIBA 966
      William Elliot Inman
    Cognitive Psychology has investigated how people search for a particular piece of information (Sternberg 1976, Neisser 1968). But little has been done to assess the cognitive processes involved in a search for missing information. This experiment asked subjects to make judgments about sets of numbers that were Correct (and complete), Wrong (including an error of commission), or Missing (including an error of omission). Sets were based on one of four prime numbers (2, 3, 5, or 7). Fifty percent of the sets were Correct (e.g., 2, 4, 6, 8). 25% were Wrong (e.g., 2, 5, 6, 8), and 25% were Missing (e.g., 2, *, 6, 8, 10). Set size (four or five numbers) and the position of an error within Wrong and Missing sets (2nd, 3rd, or 4th) were also varied. Numbers within sets were presented to subjects in random order on computer. At a "?????" cue, subjects registered their judgment for each set by pressing one of three keys on the keyboard. Statistical analyses revealed that judgments of Missing required more time and were less often accurate than judgments of Correct or Wrong. A two stage model is proposed. First, people search for salient errors. Second, if no error is found, people then search the set again for Missing information. Longer reaction times are the result of the time required for a second search. Inaccurate judgments are the result of a failure to instigate or complete the second search.
       These results have implications for the creation of error debugging messages provided by statistical analysis and programming software. Often, error messages do not identify the type of error made: Wrong versus Missing. Programmers are presented with an error message suggesting that there is an error of commission when, in fact, there is an error of omission. Such error messages may inhibit the second search for missing information and thus do more harm than good.
    Incorporating Task Complexity into Hierarchical Menu Design BIBA 967
      Julie A. Jacko; Gavriel Salvendy
    This research establishes a relationship between a hierarchical menu's depth dimension and the perceived complexity of a menu retrieval task. While it is recognized that task complexity impacts both a user's mental load and performance in man-machine systems, what has not been explicitly explored in past research is the interrelationship between a menu's depth and the complexity of the menu retrieval task. In an effort to validate the relationship between menu depth and perceived task complexity, twelve subjects were asked to utilize six different menus with varying structure. The structures of the menus spanned the extremes of depth and breadth as identified by previous researchers. The dependent variables were time to respond and accuracy. The independent variables in the study were depth and breadth of the hierarchy. Subsequent to experimentation, the subjects were asked to complete a questionnaire which was designed to gauge the users' perceptions of the task complexity associated with the different menu structures. The results in this study regarding the influence of depth on time to respond and errors were consistent with previous studies. The results indicated that the greater the depth, the longer the time to respond, and the greater the number of errors. Furthermore, as depth increased, the task complexity of the menu structures was perceived by the users to increase significantly. A link between these phenomena and an existing framework for task complexity is proposed along with an assertion that the cognitive component influencing users' perceptions of task complexity was short-term memory load.
    Effects of Discrepant Information on Decision Making BIBA 967
      Tonya M. Beers; Gregory M. Corso
    People often encounter, and must make choices regarding, inconsistent information. This study investigated the effects of discrepant information sources on task performance and decision making. Eight undergraduate students were required to evaluate a traffic situation, with traffic signals for both through and turning traffic, and to determine whether or not it was legal and safe to make a turn (in the direction of the signal) without stopping. Discrepancy is introduced into traffic situations when people are required to process the different colors (and their meanings) found on a traffic light. In some cases, the decision was very easy; in others, it was expected that, with color discrepancy between the two signals, decision time would be longer. Results showed, however, that while, on average, accuracy was not as good in the discrepant conditions, reaction time did not differ much between discrepant and non-discrepant conditions. The results of this study suggest that designers should consider the fact that inclusion of redundant displays in a system may result in greater human error if discrepant information must be interpreted. Feedback should be made as consistent as possible; and, when inconsistencies exist, redundant displays should be equally viable choices for accurately displaying information.
    Reversal of the Cry-Wolf Effect: An Investigation of Two Methods to Increase Alarm Response Rates BIBA 968
      James P. Bliss; Mariea D. Washington; Bryan S. Fuller
    In complex task environments such as airplane cockpits, false alarms have been shown to lead to a "cry-wolf" effect, with potentially disastrous implications for primary task and alarm response performances. This research examined the feasibility of two methods to reverse the cry-wolf effect: a hearsay method, where subjects were given information by a confederate about the frequency of false alarms in an alarm set, and an urgency method, where the urgency of a set of alarms was increased. Three groups of 20 undergraduate students performed two consecutive sessions of a psychomotor task. During each session, subjects were presented ten visual/aural alarms on a 50% reliability schedule (50% true alarms, 50% false alarms). As in a typical false alarm situation, there were no visual or aural differences between true and false alarms, but verbal feedback was given about response appropriateness. In Session 1, all subjects were presented 10 moderate-urgency alarms (as recorded from a B-757/767 simulator) to establish the cry-wolf effect. After Session 1, subjects in the Hearsay group were told that the next session would include alarms that were 75% true (in reality there was no reliability change). They were then presented a second session of alarms on a 50% reliability schedule. Subjects in the Urgency group were also presented alarms that were 50% true in the second session, but the urgency level of the Session 2 alarms was greater. Finally, for Control subjects there was no difference between Session 1 and Session 2 alarms. Response frequency data were analyzed by using a series of t-tests to compare Session 1 and 2 response rates for each group. Results indicated that only the hearsay strategy increased subject responding. Hearsay subjects responded to significantly more alarms in Session 2 (p < .05); however, subjects in the Urgency and Control groups showed no change in alarm response frequency.
    An Investigation of Work-Schedule Factors in Mainline Human-Factors-Caused Train Accidents BIB --
      R. Todd Brown; Douglas Scheffler
    A Fuzzy Model of Vehicular Guidance BIBA 968
      Robert C. Hale; Nong Ye
    As more people populate the nation's roadways, vehicle safety advancements are needed. The authors are investigating this issue by proposing a fuzzy model of vehicular guidance. Three main tasks are associated with vehicular guidance including maintaining path, regulating velocity, and reacting to signs and obstacles (Kramer and Rohr, 1982). This poster demonstration introduces the initial efforts to identify the factors that describe a driver's reaction to obstacles.
       In developing this conceptual model, a systematic approach is being employed. This five step approach includes a problem definition and statement of phenomena, scope of the problem, input and output definitions, input/output relationships, and a plan to validate the fuzzy model.
       Seven inputs have been identified including the driver's perception of car speed, distance between obstacle and car, rate of change between car and obstacle, position of obstacle relative to car, vertical size of obstacle, horizontal size of obstacle, and the visual evoked potential of the driver. In addition, the authors plan to evaluate three outputs including force applied to the break pedal, angle of the gas pedal, and angle of the steering wheel.
       Each input and output are described by fuzzy subsets and membership functions, and input/output relationships are defined by fuzzy compositions. This poster demonstration will provide a small example showing the capacity of these fuzzy utilities.
    Color Coding of Aircraft Emergency Exit Lighting: Stereotypes among Subjects from Four Language Groups BIBA 969
      Florian G. Jentsch
    An important issue facing aviation is the provision of safety information that is understood by people from diverse cultures. One prerequisite for the successful evacuation of an aircraft is the localization of exits by occupants. Transport category aircraft have an emergency floor lighting system which usually follows this color scheme: White lights lead to the exits which are indicated by red lights. This convention is compatible with the scheme used for exits signs in aircraft and in many buildings in the U.S., but it is different from the convention used in most European buildings. There, emergency exits are marked by green signs. Because of this difference, it was expected that European subjects would generalize their everyday experiences of the color scheme in buildings to aviation and respond that lights indicating aircraft exits are green as well. A sample of 150 college students from seven universities in four areas (English-U.K., French, German, English-U.S.) completed a questionnaire about aviation safety features. Significant differences (p < .001) between European and U.S. Ss were found both with respect to their knowledge about the current scheme as well as regarding the color Ss thought should be used in emergency floor lighting. European Ss across countries selected green significantly more often than red. Conversely, U.S. Ss selected red significantly more often than green. Also, very few Ss (<15%) from any language group selected one of the other three choices given in the questionnaire (i.e., white, yellow, blue). This effect may have potentially dangerous consequences, especially when considering that red is also often used to indicate "stop" or "danger." Future designs should therefore attempt to convey information "culture-free," for example through directional lights. In the meantime, a stronger reference in passenger information to the current scheme might be a useful alternative that should be tested.
    Promoting Safety Behaviors through Stories, Simulations, and Reflection BIBA 969
      Kevin C. Moore; Henry P. Cole; Robert H. McKnight; Larry R. Piercy
    After an accident has occurred, individuals who witnessed the scene revisit the action by talking it over with others or by replaying it in their minds. During this replay they often put themselves in the role of the person who was in danger. However, more often then not, they make better decisions leading to happier endings. Role playing of "If it were me in that situation, I would have..." is reflective behavior and an effective learning tool. After the fact reasoning and justification help guide our living and understanding of our own and others conduct (Acker, 1992; Sarbin, 1986). One rationale for using simulations in safety instruction lies in the natural creation of these vivid stories, and the internalization of the actions and the decisions involved.
       Making stories (simulations) salient to the learner is paramount for the transfer of knowledge. To create an effective simulation several variables must be addressed from the point of view of the learner. The designer should begin by compiling an inventory of the intended populations' demographic characteristics. Other important aspects include a description of the tasks and the conditions under which the task will be performed, accurately depicting the language used by the learner during the task, selection of key decision points that require the learner to identify potential hazards, and selection of actions that eliminate the hazards and prevent injury events.
       In addition to being exposed to simulation design methodology, individuals will have the opportunity to view two short interactive simulations. This presentation stresses the simplicity and low cost measures for simulation design in safety instruction.
    Nighttime Curvature Perception of Drivers as a Function of the Number of Curve Delineation Devices BIBA 970
      Helmut T. Zwahlen
    This exploratory study investigated the monocular (4 subjects) and binocular (7 subjects) curvature perception accuracy of young drivers under curve approach and nighttime conditions in a 1:50 scaled laboratory set up. The experiment consisted of a sequential comparison of a 90° curve segment with a standard radius equipped with 12 equally spaced 1:50 scaled retro-reflective yellow/black miniature (H 12.2mm, W 9.1mm) chevrons with a 90° test curve segment which could have 2, 3, 4, or 8 equally spaced miniature chevron signs along a curve radius of 95%, 97.5%, 100%, 102.5% or 105% of the standard curve radius. For each presentation the standard curve was presented first to the subjects (black road environment, low-beam illumination) for 2 seconds, then the subjects rotated 90° and observed the test curve (one of five curve radii, with either 2, 3, 4, or 8 equally spaced chevrons) for 2 seconds. A forced choice response (smaller, larger than the standard curve radius) was required from the subjects. All experimental conditions (5 radii, 4 chevron levels, 5 replications for each subject) were randomized for each subject. The curve approach viewing distance from the subject's eyes to the beginning of the curve was 4.57m (15 ft) which represents 228.6m or 750 ft in the real world, while the curve radius of the standard curve was .9144m or 3 ft which represents a curve radius of 45.6m or 150 ft in the real world. The overall averages for the percentage of the number of correct responses were calculated for all conditions and plotted against the number of chevrons. Based upon the results we may tentatively state that the overall average number of correct responses increases from 2 to 3 chevrons and then remains about the same for 4 and 8 chevrons. Further, the values for binocular viewing are somewhat higher for 4 and 8 chevrons. Therefore, we may tentatively conclude that for the above conditions 4 equally spaced chevrons within a total visual field of about 6° provide adequate curvature estimation cues for unfamiliar drivers approaching a curve at night.
    Designing a Graphical User Interface for Business Telephone Users BIBA 970
      Pamela A. Savage
    This poster session presents a case study of research undertaken during the design of a next generation small business multimedia desktop communicator. A simulation of the user interface (developed using SuperCard on a Macintosh personal computer) was used as stimuli in the early phases of customer testing (e.g., focus group research and Phase I of usability testing). A prototype consisting of a conventional telephone connected to a pen-based graphics tablet was used as a testbed for customer testing later in the development cycle (e.g., Phase 2 of usability testing). Specific examples of redesign based on customer feedback and needs are shown. In addition, a high-level description is provided of the iterative design process in which a variety of methodologies were employed, such as: customer observation, focus group research, expert review, questionnaires, customer site visits, and usability testing. The process began with a conceptual design phase driven by customer needs (e.g., affordance, tailorability, and media negotiation). Equipped with these needs, two multi-disciplinary teams designed two GUIs. As expected, two very different GUI designs resulted. A multi-line interface contained a communications tool palette that provided feature access, while a single-line interface was designed using a notebook metaphor with tabs for feature access. Although, the notebook user interface was developed to meet the needs of single-line customers, it was later modified to accommodate multiple lines. The designs were simulated according to "test drive scenarios" which were representative of typical user actions and were tested with customers during focus groups. During the focus group research, each GUI was demonstrated via a scenario which included an assortment of call-handling, note-taking, and directory functions, and gathered both quantitative (e.g., rating and ranking measures) and qualitative data (e.g., subjective reactions and comments). Based on customer feedback, the notebook GUI was selected for further development and underwent a series of usability tests and iterative redesigns.
    Effects of Telecommunication Media on Interpersonal Self-Disclosure and Deceptive Behaviors BIBA 971
      A. Rodney Wellens; Eva Szeli; Marci Gittes
    Several authors have proposed ways of conceptualizing electronic communication media in terms of the number of verbal and nonverbal communication channels supported by the technology (e.g., Heilbronn & Libby, 1973; Korzenny, 1978; Lengel & Daft, 1984; Wellens, 1986). It has been suggested that informationally "lean" media, such as computer messaging, may have a disinhibiting effect upon users brought about by feelings of increased anonymity and reduced social constraints (e.g., Kiesler, et al., 1984). Two interpersonal communication behaviors that might be affected by increased feelings of anonymity and remoteness are self-disclosure and deception.
       Two experiments were conducted that independently examined the effects of communication media upon self-disclosure and deceptive behaviors. Experiment I compared the amount and depth of self-disclosure revealed by 48 college women administered a standardized clinical interview in either a face-to-face, telephone or computer messaging condition. Results indicated that subjects engaged in a greater volume of self-disclosure in the face-to-face condition than in the computer messaging condition, with an intermediate amount disclosed in the telephone condition. The judged depth of self-disclosure did not vary between media conditions. Experiment II compared the number and degree of self-reported intentional exaggerations offered by 80 college women who were asked to deceive a "therapist in training" within either a face-to-face, two-way television, telephone or computer messaging setting. Results indicated that subjects' willingness to engage in deceptive behaviors did not vary much between communication conditions. Only for extreme, "bold face" lies was there any evidence of differences between communication conditions. Subjects were more willing to present themselves in an exaggerated positive light when engaged in face-to-face interaction and more willing to portray themselves in an exaggerated negative manner while being interviewed over a computer messaging system.
    Tactile Path Information in User Interface Navigation BIBA 971
      D. V. Keyson
    Current means of showing a path for movement in graphical user interfaces tend to rely on visual indicators which can be distracting if presented during a task that requires visual attention. In the current study, the use of tactile and combined visual-tactile directional cues as a means for communicating a direction of movement was explored. Tactile, visual, and combined visual-tactile cues were contrasted in their abilities to indicate a direction of movement in a single and dual-task game. In the single task, participants were instructed to rotate a trackball in the indicated direction. In the dual-task a visual search task was performed in addition to the movement task. The tactile cue, compared to the visual cue, reduced movement response times, whereas movement accuracy was sometimes lower. The combined visual-tactile cue produced response times equal to the tactile cue and greater movement accuracy in all conditions. Performance gains using tactile cue information under visual load were attributed to a higher degree of sensory compatibility between a tactile directional cue and a motoric response as opposed to a visual cue and a motoric response. Benefits of game feedback and a simulated task in studying navigational performance using multi-sensory cue information are discussed.
    The Effect of Type of Instruction and Help Systems on Learning the Macintosh Operating System BIBA 972
      John K. Muller; Mark A. Wise; Jennifer L. Dyck
    The value of personal and office computer systems is measured by their contribution to productivity, and justified by their ability to reduce costs. This premise is confounded by a visit to any bookstore, which reveals hundreds of texts devoted to providing needed guidance for wayward users. Universities and businesses alike spend millions of dollars hiring computer support personnel to aid in performing such basic tasks as installing commercial software and formatting floppy disks. The prevailing assumption in current interface design, is that the graphical-user-interface based operating systems provide an "intuitive" means by which novice users can effectively and efficiently perform basic functions. But, what is really required of an interface for it to be truly and completely intuitive?
       The present study examined the effect that on-line help and different levels of instruction have on novice user performance within the Apple Macintosh operating system. Naive computer users were assigned to one of six instructional conditions in a 3 x 2 (type of instruction x on-line help system) between-subjects design. Subjects received either 1) no instruction, 2) a manual covering basic mouse skills, or 3) a manual covering basic mouse and file manipulation skills. Balloon Help (on-line help) was either turned on or off in each condition. Subjects were then asked to solve a series of 14 tasks which included mouse/icon manipulation, file manipulation, and transfer tasks. The transfer tasks required the subject to relate previously learned skills to novel situations.
       The benefits gained from this research will aid in future interface design by revealing the benefits and limitations of on-line assistance and the impact of level of instruction on novice user performance. The scope of these results, however, is not limited to personal computers, but may be applicable to other situations in which users are introduced to novel computer systems (e.g., air traffic control workstations).
    On-Line Methods and Procedures in a Client/Server Environment: A Success Story BIBA 972
      Kenneth R. Ohnemus; Diana F. Mallin
    Regardless of the usability or sophistication of a graphical user interface (GUI), some jobs still require methods and procedures (M&P) to complete the necessary duties. Many tasks in a telephone central office use M&Ps to avoid service outages as well as other potentially disastrous events for customers. Since the ultimate goal was to assist users in performing their jobs, the challenge was to provide a mechanism for deploying user-centric M&Ps to users at telephone central offices in over 600 locations. The application supported by the M&Ps is a telephone central office inventory system which assists users in maintaining an accurate inventory of telephone equipment and connectivity, and facilitates asset management by central office personnel. The M&Ps had to be on-line while providing a look and feel consistent with the application. Users also needed to be able to access both the application and M&Ps simultaneously. To eliminate the need for paper copies, supply easier access, and create a more user-friendly environment, hypertext-linked M&Ps utilizing a commercial UNIX software package were developed. This paper discusses issues surrounding the design and implementation of on-line M&Ps in a distributed client/server environment and why this effort was a success.
    Perception of Alert Messages for Computer Displays BIBA 973
      Dick Steinberg; Ruth C. B. Goulet; Kacheshwar Pathak
    Research is being conducted to develop human engineered display designs which optimize the man-to-machine interfaces for the U.S. Army's Ground Based Radar (GBR) real-time display developed by the Raytheon Company. The human engineered displays will help to minimize personnel reaction time, fatigue, and human errors. The research is motivated by the large quantity of information available to personnel monitoring a typical mission. Of particular concern are the alert messages. These operator alerts could contain information which pertains to personnel safety and mission success. Potentially, an operator may have two dozen alert messages sent simultaneously. These alert messages must be monitored and comprehended so the crucial decisions can be made by personnel within a short time. The goal of the research performed was to determine methods that minimize the time required for mission personnel to react to an event displayed in the alert message area while maintaining cognizance of all other pertinent data. Three display methods were developed and laboratory tested. The results identified an improved format that reduces operator response times by 30% as compared to the traditional alpha-numeric text format. Additional testing is necessary to validate the results for actual GBR display and scenarios. Results from future testing may provide the opportunity to enhance current display designs for follow-on contracts of the GBR.
    Sauk Prairie Memorial Hospital: POE -- Way-Finding Evaluation, Design, and Implementation BIBA 973
      Stephen C. Boelter
    Sauk Prairie Memorial Hospital is a regional hospital serving rural communities in southwestern Wisconsin. In late spring of 1993, the hospital was completing a major addition and the administration determined the need to resolve newly discovered way-finding problems. A case study POE was utilized to systematically evaluate the facility, develop design criteria for signage and other wayfinding aids, and to facilitate implementation of the system. The project was divided into three stages: 1) a thorough facility evaluation with development of design recommendations and guidelines, 2) research sign systems which would meet the design criteria and solicit bids from local vendors, 3) implementation of the system design with the selected sign vendor. In addition to the practical evaluation objectives, the project included demonstrating how to adapt and apply current way-finding theory to design practice, educate the constituencies within the hospital about way-finding and coordinate the project.
       As part of the final design outcome, a custom sign system was developed for both the campus exterior and facility interior. The new system provided clear and understandable way-finding information, flexibility for physical change, expandability, a quality image and compatibility with a commercial system. Upon conclusion of the project and during the post-installation evaluation, the client acknowledged that the issue of way-finding should have be addressed as part of the planning process.
    Comprehension and Compliance to Elevator Service Signs BIBA 974
      John W. Brelsford; Michael S. Wogalter; Paul B. Begley; Lori F. Scancorelli; Jay H. Williams; Stephanie A. Terry
    This research examined comprehension of and compliance to four different elevator service signs. The purpose of the signs was to dissuade people from tying up the elevator when they are only going up one floor or down two floors. Three of the four signs were described by Chapanis (1964): an original sign and two others that he offered as possibly being better. The fourth sign was an enhanced version that used human factors principals derived from research studies on the topic since the Chapanis (1964) article. The enhancements involved the use of color, signal word, signal icon, pictorials, and concise, clear wording. In Experiment 1, participants rated the understandability of each sign and their willingness to obey the signs' instructions. Both questions showed the same pattern of results. The original sign was rated lowest and the enhanced sign was rated highest; the two other signs received intermediate ratings. In Experiment 2, the signs were placed on each floor of six buildings in conspicuous locations near the elevator call buttons. People's use of the elevators during the posting of each sign as well as during five no-sign (control) periods was measured. The experimenter rode the elevators and counted the total number of passengers using the elevators as well as the number who rode up only one floor or down less than two floors (noncompliers). Results showed that only the new enhanced sign had an effect on compliance compared to the other three signs and the control conditions. These results suggest that the sign principles developed from research in recent years are likely to assist in promoting comprehension and behavioral change.
    A Measure of Subjective Perceptions of Stress: A Validation and Reliability Study BIBA 974
      Leslie Beth Herbert; Michael J. Paley; Kristen M. Haggis; Donald I. Tepas
    Previous research suggests that the subjective perception of stress varies with one's experience and the proximity of the stressful event. Current researchers found an interaction between ratings of tension and experience for firefighters responding to emergency calls. These researchers employed a measure of perceived tension in both a primary study of 21 firefighters and a follow-up study of 17 of these men. The firefighters responded to the question, "How tense did you feel during the following situations on your last call?" For each of the five proximities of an emergency call, waiting for a call at the station, on the way to the call, at the call location, on the way back from the call, and back at the station after the call; firefighters responded on a four point Likert-type scale, ranging from "Not at all" to "Extremely." In addition to the question being presented on daily work logs, the question was included on the background survey, asking how one "usually feels" at each of the call proximities. This question was also employed on a background survey and daily log forms used in a primary study and a follow up study of eight male marine search and rescue crew members. Using the same question format, the crew members were also asked to rate their perceived physical tiredness. This study examined the psychometric characteristics of the data gathered from both worker groups, and the appropriateness of using the measure on a one time background survey as an accurate representation of actual subjective perception of tension. Correlations between the two separate administrations of the background survey for the measure of perceived tension were statistically significant (p < .05). Test-retest reliability of the rating of perceived tension for both the firefighters, = 967, and the crew members, r = .943, was obtained. Concurrent validity (p < .05) was found between the ratings on the background survey and on the daily work logs for both the firefighters, .987, and the crew members, r = .914. Examining the crew members' responses to the measures of perceived tension and physical tiredness, divergent validity was found for the ratings on both the background survey and the daily work logs. The results found by this study demonstrate that the one-time general background survey estimate of perceived stress used in these two studies is valid, easy to administer, and an accurate measure of daily workday log stress estimates.
    Work Design, Smoking Behavior, and the Productivity and Well-Being of VDT Operators: The Results of a Field Study BIBA 975
      George V. Kissel; Robert A. Henning
    A workplace intervention of scheduling frequent breaks from computer work was used at two work sites (N=91) to determine if smokers (N=33) and non-smokers respond differently to changes in organizational design. The experimental conditions were: 1) control, 2) additional rest breaks only, and 3) additional rest breaks with stretching exercises. Operators at Site 1 were assigned to one of the three conditions, while at Site 2, operators received the three conditions in the order listed above. The breaks or breaks and exercises occurred every 15-min. Operators completed surveys daily to assess mood, perceived work effort, and social interaction. Measures of productivity and smoking behavior were also collected. Smokers reported feeling less relaxed, less vigorous and more fatigued than non-smokers at Site 1. At Site 2, smokers reported feeling less relaxed but less active and more vigorous than non-smokers and smokers reported more social interaction and were more productive than non-smokers. With added breaks, smokers at Site 1 reported decreased fatigue, while at Site 2, both smokers and non-smokers reported increased relaxation. Therefore, smokers and non-smokers were found to differ in their responses to organizational design. In addition, the workplace intervention of added rest breaks benefited smokers more than non-smokers. These results suggest that smokers differ from non-smokers in their reactions to organizational factors in the workplace and that smokers may use smoking behavior as a means to adjust to organizational demands. Therefore, smoking behavior merits consideration as an important mitigating factor when the effects of organizational interventions on worker productivity and well-being are being considered or evaluated. Given the current trend towards smoke-free workplaces, the impact of factors like those identified in this field study should be considered when efforts are made to promote smoking cessation programs in the workplace.
    Software Use and Disuse: From Shelfware to Improved Software Utilization BIBA 975
      Laura Silver; Irwin Marin
    All too often, software that has been purchased is not used. A term has emerged to describe this phenomenon, "shelfware." Our research investigates shelfware, seeking to understand how interacting behavioral factors affect software use. We also are applying the results to develop a system for improving software utilization. In our approach, the behavioral factors (independent variables) are categorized into individual, group, and organization user levels. The dependent variable chosen as indicative of utilization is the perceived amount of software use. From an initial sample of responses obtained from our questionnaire on software utilization, a set of optimal decision rules (variable interactions) are induced using a machine learning program with an ID3 learning algorithm. These rules are then converted to an expert system we call a Shelfware Advisor Front End (SAFE). Initial results indicate that the amount of software utilization varies with interactions among individual, group, and organizational factors in the following decreasing rank order: the motivation of the individuals using the software, group cohesion, and software cost. The conditional rules induced from the model suggest which variables to adjust to increase software utilization. To experience the approach firsthand, at the poster session we can work with interested parties using SAFE for their particular cases of software. Their data readily can be entered, and the person can leave with information on how to decrease the occurrence(s) of shelfware and increase software use. After returning to their organizations, they can experiment with the information suggested by SAFE. If consent is gained, their data may be used to extend this ongoing research. By using this approach, we hope both to further the understanding of the variables that impact software use, as well as to put this information to use in decreasing the occurrences of shelfware and thereby improving software utilization.
    Taking a Byte Out of Crime BIBA 976
      Cathrine E. Snyder; Victoria K. Turley; Michele Terranova; Monty Grubb
    The human operator is an integral part of fingerprint processing at the Federal Bureau of Investigation. The automated system currently being procured by the Federal Bureau of Investigation will provide support to the operator, but human operation and judgment will not be replaced. The Federal Bureau of Investigation, therefore, asked us to build a prototype and conduct a human factors usability evaluation. The prototype included Fingerprint Processing Workstations with the capability to support operations on scanned fingerprint cards in a process that is analogous to current paper processing.
       The evaluation was to have an additional purpose of promoting user acceptance and introducing a new work environment to the fingerprint processing operators. As a result, our evaluation included 11% of the fingerprint processing workforce; the subject population was heterogeneous with regard to age and experience. We used an orientation, two- and three-person team sessions, individual hands-on sessions, performance assessments, workload ratings, post-evaluation questionnaire, and focus groups to perform the study. The users' prior involvement in the design process encouraged both specific and general comments that were valuable inputs to the final system design. The usability questionnaire showed a high level of user acceptance. We believe the evaluation also successfully introduced a new way of processing fingerprint cards to a large section of the Federal Bureau of Investigation's workforce.
    Individual Differences in Graph Appreciation: Are Preference and Performance Related? BIBA 976
      Andrea M. Lonon; H. Scott Kinslow; C. Melody Carswell
    Numerous researchers have indicated that graphical formats can often enhance memory and understanding of quantitative information beyond what is conveyed by tabular or other alphanumeric formats. However, there seem to be many "everyday users" who would disagree with the experts, instead endorsing statements such as "Graphs are usually a waste of space," or "When I encounter a graph I usually skip it." The question of interest here is whether preference for graphs is actually related to one's ability to use them as memory aids.
       Subjects who scored in the highest and lowest quartiles of a graph preference survey were asked to perform a graph memory task. The 18 high preference subjects were more accurate than the 18 low preference subjects when recalling line graphs with more than one departure from linearity. High preference subjects reported using more strategies on average than low preference subjects, including memorization of individual data points and creation of global mental images.
    Individual Differences in Pilots' Perception of Simulated Night Vision Goggle Scene Imagery and Superimposed Flight Symbology BIBA 977
      D. Michael McAnulty; David B. Hamilton; Dennis K. Leedom
    During night vision goggle (NVG) helicopter flight, it is difficult and potentially unsafe for aviators to divert their attention from the external scene to scan cockpit instruments. To address this problem, systems (generically called ANVIS-HUD) have been developed that superimpose flight symbology on the imagery in one of the NVG intensifier tubes. However, previous research indicates that superimposed symbology may distract the pilot's attention from obstacle detection, recognition, and avoidance, and may interfere with proper scanning patterns. Flight scenarios that simulated NVG imagery only (IO), symbology only (SO), and the imagery and symbology combined (IS) were presented to 36 helicopter aviators in three sessions. The aviators monitored and responded to predefined targets (scene features and events and/or symbology out-of-tolerance states for four instruments). Both the percentage of correct detections and aviator reaction time were measured. The aviators detected and responded rapidly to a high percentage of IO and SO targets, but performance was better for symbology than for scene targets. Performance was significantly degraded in the IS condition, but the decrement was small compared to the increased information available. The aviators divided their attention effectively between the types of information. There were significant individual differences in ANVIS-HUD performance, however. Older, more experienced aviators performed better with the scene information; younger, less experienced aviators performed better with the symbology information. Presenting the symbology to the aviator's dominant eye generally produced better performance, but there were conflicting results. Most aviators developed a scanning strategy, but the pattern varied.
    The Relationship of Psychomotor Skill to Perceived Proficiency on a Self-Reported Biographical Inventory BIBA 977
      T. Nontasak; D. R., Jr. Street
    This study examined the potential usefulness of perceived psychomotor skill as an adjunct to landing craft air cushion (LCAC) crew selection. Currently, LCAC vehicle crew members are selected based on their performance on a test battery involving cognitive and psychomotor skill. The ongoing assessment of these entry requirements is essential for selecting the most qualified candidates. Training performance has also been demonstrated to relate to various personality and background inventory (BI) items. The BI consists of items that reflect an individual's perceived level of psychomotor skill. We determined if perceived psychomotor skills are related to actual psychomotor abilities. One hundred and ninety subjects were administered a computer-based psychomotor selection test battery that included experimental personality and BI items. We compared the responses regarding perceived skill to performance on the psychomotor tests. Our correlation analysis results revealed a significant positive relationship between perceived and actual psychomotor skills (p <.05). The results suggest that certain BI items may be useful in initial screening of potential LCAC crew trainee candidates. Implications for selection and training are discussed.
    Development and Evaluation of Automobile Symbols: The Focus Group Approach BIBA 978
      Scott A. Macbeth; William F. Moroney
    Sayer and Green (1988), using the production method, had individuals produce symbols that were compared to automotive symbols from ISO Standard 2575. Under the production method, each individual is provided with a brief scenario describing a control or display, then, the subject draws a symbol representing the symbols name. The current study also used the production method but utilized focus groups (N=12) rather than individuals to develop the symbols. This study is divided into two distinct phases: 1) a production phase, wherein 56 symbol designers worked in groups of 4-7 to develop symbols and 2) an evaluation phase, wherein, 70 new evaluators ranked the symbols for meaningfulness. All participants were college students. Data from the evaluation phase suggested possible replacements for three ISO symbols: Brake Failure, Air Vents All Outlets, and Windshield Wiper Intermittent. Several other symbols were identified as troublesome, specifically, Air Conditioner, Unleaded Fuel Only, Hazard Warning, Choke, and Master Lighting Switch. These were judged to be inadequate in both the ISO form and the forms produced by the focus groups.
    Effect of Task Difficulty on Fault Diagnosis Strategy BIBA 978
      Anpin Chin; Ram R. Bishu
    The application of appropriate strategies is the key in fault diagnosis. Most studies on fault diagnosis have paid attention to complexity simplification, and various approaches have been studied to assist humans in overcoming difficulties. Interface design and training are two main directions to improving human performance in fault diagnosis. A number of visual aiding tools have been designed to decrease the effect of perceptual complexity. With a set of topographic search tasks, this study evaluates the efficacy of two typical strategies, linear tracing strategy and structure decomposing strategy, on different levels of task difficulty.
       Fifteen subjects participated in a factorial experiment. The highlighting aiding tool designed from the Linear-Tracing (LT) strategy was preferred by the subjects because of the advantage of saving time, but this tool resulted in lower accuracy in decision-making. Although all the subjects were instructed to pay attention to the accuracy rather than the speed, the problem of changing the focus from accuracy to speed usually occurred while using the highlighting tool. The blocking tool designed from the Structure Decomposing strategy (SD) provided higher accuracy but took time. The physiological decline caused by aging showed no effect on the accuracy but resulted in speed loss. Another finding showed that the effect of perceptual complexity might not be fully determined by the size of the test problem, but by the average amount of pertinent input/output signals of feasible components in such a topography search task.
    Decision-Making Performance and Decision Aid Usage under Controllable and Uncontrollable Stress BIBA 979
      Robert D. Peters; Deborah A. Boehm-Davis; Joanne B. Fertig
    Previous animal research has shown that stress that is perceived to be out of the organism's control causes different physiologic changes and leads to more detrimental effects on performance than stress that is perceived to be under at least partial control. With humans, differential physiologic changes and greater decrements in certain cognitive tasks have resulted from exposure to uncontrollable versus controllable stress. The current study examined the effects of controllability of stress on decision-making performance. Unpredictable bursts of loud noise were used as the stressor. It was hypothesized that uncontrollable stress would have greater negative effects on mood and decision-making performance than controllable stress. In addition, it was hypothesized that the stress groups would make differential use of a computerized decision aid and that availability of the aid would lead to better decision-making performance than a no aid group. A 3 (uncontrollable, controllable, or no stress) by 2 (decision aid or no aid) mixed factorial design with repeated measures of decision-making performance was used to test these hypotheses. Results confirmed the experimental hypotheses regarding stress. Significant differences were found between the controllable and uncontrollable stress groups with regard to 1) mood changes (i.e., uncontrollable stress subjects showed increases in fear and anger and a decrease in happiness, while mood for the controllable stress subjects remained relatively stable) and 2) the decision-making process (i.e., uncontrollable stress subjects were less systematic and consequently less accurate than controllable stress subjects). Contrary to expectations, type of stress did not differentially affect decision aid usage. Having a decision aid available, however, did positively affect mood and decision-making performance.
    Human Factor Criteria and Metrics Applied to Bioverification Technologies BIB --
      Christine L. Rainey; Sheila K. Humphreys; Len Natkin
    Increasing the Correct Connection of Automobile Battery Jumper Cables with an Enhanced Warning BIBA 979
      Michael S. Wogalter; Michael J. Kalsher; Barbara L. Glover; Amy B. Magurno; Jake T. Fisher; Daniel L. Dunham
    Every year people are injured while improperly "jump starting" automobiles using battery booster cables. A common scenario leading to injury occurs when people attach both negative leads to the battery terminals instead of properly grounding the negative lead of the "dead" battery to that vehicle's engine block. An incorrect configuration can cause the "dead" battery to explode, discharging strong sulfuric acid. Two experiments examined the effectiveness of pictorial-based tag warnings illustrating the proper connection of jumper cables and the hazards associated with improper connection. Experiment 1 used pictures of cars with open hoods. When the warning was present, participants were significantly more likely to draw the correct connection sequence than when the warning was absent. Experiment 2 used actual booster cables to connect two realistic appearing mock batteries in adjacently parked vehicles. When an enhanced warning was present on the cables, participants were significantly more likely to accurately connect the batteries compared to an unenhanced (a current manufacturer's) warning or no warning. The present research shows that well-designed pictorial warnings can modify inaccurate beliefs and behavior concerning proper jumper cable connection.
    Postural Stability and its Relation to Simulator Sickness BIBA 980
      Eugenia M. Kolasinski; Sherrie A. Jones; Robert S. Kennedy; Richard D. Gilson
    The use of simulators as training devices has become widespread, especially in the military for pilot training. Although simulators provide an effective means of training pilots at a much reduced cost compared to actual flight, "simulator sickness" may result. Postural instability is one symptom of this sickness and a well-documented effect of simulator exposure. Postural stability is often measured before and after simulator exposure to determine decrements due to exposure. However, it does not appear that this measure is typically used as a predictor of sickness. The analysis reported in this poster attempts to determine if there is a relationship between postural stability and simulator sickness. Pre-exposure postural stability data and post-exposure simulator sickness data were collected from Navy pilots in conjunction with a training session in a helicopter simulator. These data were analyzed for relationships between postural stability and sickness. It was hypothesized that individuals who are less posturally stable will be more likely to experience simulator sickness or will experience more severe sickness. On the other hand, individuals who are more posturally stable will be less likely to experience simulator sickness or will experience less severe sickness. Several analytical techniques were attempted: Pearson correlation, Spearman correlation, Gamma and Lambda categorical methods, ANOVA, and Trend Analysis. Each technique evaluates a different aspect of the data and no one technique best represents the findings. Although clear, strong relationships have not been found, there is evidence suggesting that postural stability is associated with simulator sickness as hypothesized in this study. Further research with a more diverse population and different types of simulators, including virtual environments, is clearly warranted.
    Object Recognition, Size Estimation, and Distance Estimation in Real-World and Virtual Environments BIBA 980
      Donald R. Lampton; James P. Bliss; Bruce W. Knerr
    Technologies to immerse users in three-dimensional computer generated spaces, virtual environments (VEs), have many potential training applications. Practical training applications are limited by the costs and performance shortfalls, such as display resolution, of current immersive VE technology. Research is needed to determine technology requirements for training applications and, as important, effective methods of using the technology. To support this research we developed a battery of perceptual and psychomotor tasks to measure human performance in VEs. This paper describes the performance of twenty-four naive VE users on three of the visual tasks from the battery; object recognition, size estimation, and distance estimation. Subjects were required to recognize an object (a human figure) at a distance of 40 feet, estimate its size (height), and estimate the range to the figure as it moved toward them. Range to the figure was estimated at actual distances of 40, 30, 20, 10, 5, and 2.5 feet. A Virtual Research Flight Helmet driven by dual 486/50 mhz PCs provided a stereoscopic visual display of the virtual tasks. A separate group of 36 subjects performed the tasks in a real-world setting roughly analogous to the VE. All subjects recognized the object at a distance of 40 feet, the maximum distance used. Significant differences (p <.001) were found between VE and real-world estimates of size and distance. In the VE, subjects tended to underestimate the size of the figure and overestimate its distance. Real-world estimates were more accurate than VE but differed significantly from perfect performance.
    An Assessment of Training Methods for Complex Mechanical Assembly Skills BIBA 981
      Peter J. Linnerooth; Phillip N. Goernert; Daniel Houlihan; Wayne C. Harris; Samuel J. Dollar; Renden S. Bruner
    Training via behavioral modeling has been widely implemented in industrial-organizational contexts. At present, however, empirical evidence regarding the characteristics of an effective model, the cost effectiveness of behavioral modeling training and the efficacy of behavioral modeling versus other training methods is lacking (Tannenbaum & Yukl, 1992).
       The present investigation addressed these concerns by briefly training and testing 80 college student subjects in a mechanical device assembly task. Videotape Self-Modeling (VSM) (Dowrick, 1983) Videotape Other-Modeling (VOM), Self-Directed Mental Rehearsal (SDMR) and No Training Control Group (NTCG) training methods were compared. VSM and VOM videotape training methods depicted correct, dexterous and rapid assembly task performance by self- or other models respectively. SDMR training involved the production of assembly task imagery via mental visualization from memory.
       Statistical analysis of assembly times indicated that the SDMR group was superior to the control group and both video modeling groups. Additionally, there was no significant difference in assembly times between the videotape and control groups. Furthermore, the advantage in assembly times by the SDMR group was maintained over a four month follow-up. It is hypothesized that the use of mental rehearsal allowed subjects to capitalize upon skills gained during successive assembly attempts. Owing to the cumbersome nature of their production, VSM and VOM training models could not be updated to depict changes in skills throughout assembly training, negating the potential advantages of their depiction of superior assembly performance.
    Computer-Based Interactive Courseware: A Medium for Team Training? BIBA 981
      Randall L. Oser
    Recent efforts in team training research have resulted in systematic approaches for the development of team training. These methodologics have been used effectively to develop and implement a variety of team training programs. Most of these team training programs have focused on traditional types of instructional media (e.g., lectures, roleplays, case studies, simulation), while few team training programs have utilized computer-based instructional technologies. The technological advances in computer-based interactive courseware (ICW) systems (e.g., real-time video, branching, networking, synthesized/digitized audio) have potential application in team training settings. While the use of ICW for team training is intuitively appealing, little is known about how to best develop or design computer-based ICW team training programs.
       The purpose of this poster session is to discuss the use of computer-based interactive courseware for team training. The poster will focus on application of existing team training design methods for implementation into ICW. Specifically, the poster will: (a) outline a framework for approaching the development of computer-based team training (i.e., information, demonstration, practice, and feedback), (b) describe existing ICW capabilities which are applicable for team training, (c) outline an effort to integrate team training into ICW, and (d) discuss the need for team training research related to the use of ICW. The design of ICW for team process skills training appears to be possible by applying current team training strategies.
    Shared Mental Models for Team Performance: An Anatomy BIBA 982
      J. Thomas Roth
    The construct of mental models has been developed to explain cognitive processes and phenomena in a number of domains. Effective mental models provide structures for knowledge that enable interpreting and predicting external phenomena. Recent research has established the primacy of team members' development and maintenance of mental models with certain elements shared among team members, as a critical determinant of team performance. These shared mental models should include representation of: (1) the team's structure; (2) team roles and role interdependencies; (3) patterns and modes of communication among roles and team members, (4) the team's tasks, and (5) critical characteristics of the team performance situation or context. Remaining to be developed are means and methods for developing effective shared mental models supporting team performance. This presentation sets forth a candidate structure and suggested content for shared mental models supporting team performance, and suggests a hierarchy among mental model components; thus, an anatomy. Lower-level, more general mental models provide structures for the derivation or creation of higher-level, specific mental models that facilitate performance in more specific circumstances. In some cases, higher-level, more specific mental models may be alternate instantiations of knowledge structures inherent in lower-level mental models. Data supporting the shared mental model hierarchy were derived from extensive observation of military teams training and performing in naturalistic environments, over a 2-year period.
       The shared-mental model hierarchy suggests how and when training for developing effective mental models might take place. A proposed sequence and structure for team training, to engender development of effective shared mental models supporting teamwork, is outlined. When and how elements of the proposed team training should be provided, in terms of the development of effective persistent teams with turnover in membership but stable basic team-role structures, is discussed.
    Identification of Standardized Tasks for the Assessment of Operator Workload BIBA 982
      N. Clayton Silver; Curt C. Braun; Robert S. Kennedy
    As cockpits become increasingly complex, operation and control of the aircraft requires that pilots integrate information from a wide variety of sources. Thus, the pilot's workload has increased as a function of this complexity. Traditional efforts aimed at reducing operator workload via automation have produced mixed results. An alternative to this approach involves using adaptive automation in which system and subsystems are automated as a function of operator workload. The feasibility of such automation, however, is dependent upon the reliable measurement of operator workload. Therefore, the purpose of the present study was to examine the differences in workload and performance in a cognitive task, namely channel monitoring. Nineteen undergraduate psychology students completed a complex counting task. Participants were instructed to monitor a specific number of channels and respond after counting a specific number of tones in each monitored channel. There were three monitored channels, three interstimulus intervals (0.5, 1.5, and 3.0 seconds), and number of events counted (2, 4, and 8). The participant responded to all 27 conditions. Perceived workload increased and performance decreased as the number of channels monitored increased, the interstimulus interval decreased, and the number of events counted increased. These results demonstrate that an available workload measure can provide differences in operator workload and serve as the foundation for future research. These findings and tasks can be used as a tool to assess the sensitivity of other workload measures.
    Development of "Presence" Measures for Virtual Environments BIBA 983
      Michael J. Singer; Bob G. Witmer; John H. Bailey
    A primary argument for the efficacy of Virtual Environments (VE) applications is that the user is "present" in the simulated environment. Presence is defined as the subjective experience of being in one environment (there) when one is physically in another environment (here). Presence is probably a normal awareness or attentional phenomenon based in an interaction between external or immersive factors and internal tendencies. These same factors support the transition to or experience of presence in a remote or artificial environment. Major immersive factors identified in current literature have been used as the basis for two different questionnaires. The Immersive Tendencies Questionnaire (ITQ) was developed to identify possible behavioral correlates or indicators for experiencing presence in artificial environments. The Presence Questionnaire (PQ) was developed to address both the extent of involvement in a VE experience and the effect of different factors or features of the virtual environment. Results of questionnaire administrations in conjunction with experiments into basic task performance and building interior route learning in VE indicate reasonable reliability values (Cronbach's alpha) for both the ITQ (.75) and PQ (.80). The PQ and ITQ correlate significantly (r=.31, N=132, P<.001). Several subscales were revealed in Cluster analyses on each questionnaire. Preliminary investigations with a priori subscales indicate relationships with movement/manipulation task performance. Correlations between the PQ and a standard Simulator Sickness measure reveal significant negative correlations between the overall scores and several subscales. Analyses of data from three experiments are presented in connection with revisions made to the questionnaires.
    Improving Questions on Questionnaires BIBA 983
      Sailaja Balijepalli; Arthur C. Graesser; Shane Swamer
    Designers of questions on questionnaires and forms attempt to maximize the ease of comprehending questions, and the reliability and validity of the answers. However, there are tradeoffs in trying to satisfy these goals. Questionnaire designers have in past used a brute-force, empirical approach to improving the quality of questions by pretesting questionnaires on a pilot sample of respondents. Researchers in survey methodology have often expressed the need for a satisfactory theory of question revision which is grounded in psychological research. We conducted some studies to illustrate how well QUEST, a cognitive computational model of human question answering, works in improving questions on questionnaires and forms. Study 1 measured the incidence of problematic questions in a sample of forms and questionnaires. The results suggest that QUEST can handle most of the problems. Study 2 assessed whether the revision of problematic questions improved the reliability of answers on four of the original five forms. The questions produced significantly more reliable answers when they were revised according to the QUEST model. Therefore, these initial studies indicate that QUEST holds some promise as a theoretical guide for question revision .
    Construction and Validation of a Musculoskeletal Risk Questionnaire BIBA 984
      Libby Cole; Roger Rosa
    The aim of this project is to develop a questionnaire for identifying jobs associated with the risk factors for musculoskeletal cumulative trauma, including repetition, force, and whole-body exertion. The questionnaire, designed to be minimally ambiguous and generalizable across occupations, contains familiar, everyday activities (e.g., walking, slicing a potato). The activities serve as perceptual-cognitive "anchors" against which work activities are judged to determine level of exposure to musculoskeletal stress.
       During the first phase of questionnaire development, ergonomics experts generated common activities involving repetition, force, or whole-body exertion. During the second phase, workers rated their familiarity with the activities using a scale ranging from 1 (no familiarity) to 7 (high familiarity). One-hundred fourteen activities rated as very familiar (i.e., a rating of 5 or greater) by at least 75% of the respondents were retained for the next stage. During the third phase, experts and workers classified the familiar activities into the three risk factor categories (force, repetition, and whole-body exertion). Eight whole-body activities, 25 repetitive activities, and 41 force activities were classified similarly by at least 50% of the experts and workers. During the fourth phase, workers rated the intensity of the categorized activities. Nineteen activities of varying intensities rated similarly by at least 50% of the respondents were selected as anchors for the risk factor scales. To use these scales, workers compare their job activities with the anchors to indicate the amount of repetition, force, and whole-body exertion their jobs require.
       Discriminant validity of the questionnaire is being assessed by comparing the ratings of assembly workers and manual materials handlers. It is expected that repetitive motion ratings will be greater in assembly workers and whole-body activity ratings will be greater in materials handlers.
    A Mobile Reconfigurable Crewstation -- MRCS BIBA 984
      R. J. Edwards; D. F. Streets
    The Centre for Human Sciences has designed and produced a mobile reconfigurable crewstation -- MRCS -- which is used to investigate the human factors impact of emerging military technology in an operational setting. The crewstation comprises a reinforced fibreglass shell mounted upon a wheeled utility base vehicle. This allows the system to be used on military training areas and public roads.
       The crewstation can be configured for 2 or 3 crewmen in variable positional arrangements. Each workstation is equipped with reclinable seats, all round optical vision and a conventional cathode ray tube display. This can show images from a vehicle borne mast mounted steerable sensor package which comprises a thermal imager and a daylight camera. The designated "command" crewstation is further equipped with a stabilised panoramic sight. The "co-command" workstation is equipped with a fully validated driving simulator which is takes its input from the base vehicle driving controls. An Apple Macintosh IIfx computer hosts a raster scan map installed into a SuperCard environment. Map representation is on an active matrix, colour, flat panel, LCD screen. This allows either crewman access to the system, and permits map orientation to any preferred position. A simulated global positioning system is also available.
       Crew activities are recorded onto a micro-computer by an observer in the crewstation, or onto video. Communications may also be recorded. The MRCS has been used in proving trials lasting over 24 hours, and in studies investigating the use, and elucidating any operating difficulties of "electronic maps". The system will be used in studies supporting research for the British Army's next generation of vehicles.
    Differences in Inference Generation between Young and Older Adults BIBA 985
      Eugenie L. Bertus; Arthur C. Graesser
    Understanding written directions (i.e., medical instructions or technical equipment instructions) is a common necessity in everyday life. It can be particularly important for older adults who may risk life-threatening problems if they do not properly following instructions. The research presented here examined reading comprehension differences in old and young adults. In particular, the study observed the generation of knowledge-based inferences while subjects comprehended scientific expository texts. Subjects read experimental passages one sentence at a time in a self-paced reading time paradigm. The sentence reading times were analyzed using multiple regression. The predictors variables of interest were: confirmed expectation scores and inference volume scores. The regression accounted for a significant percent of the data (r² = .69). For both young and old adults, the confirmed expectation score was significant. A sentence was read faster if it confirmed an expectation set up by a prior sentence. These results suggest that older adults are more likely to be guided by the overall expectations generated in the text.
    Age Differences in Memory and Metamemory in a Lateral Orientation Task BIBA 985
      Regina C. Colonia-Willner; Anderson D. Smith
    Elderly adults have been demonstrated to gain remarkably less knowledge about new environments than do younger adults. Because the implications of memory for lateral orientation are important for both the ability to acquire new spatial information and to retain longstanding cognitive maps, research on how different age groups learn to navigate a strange environment and retain orientation clues may be of particular interest. In order to be strategic in using what has been acquired, though, one needs more than memory. One needs to know one's own memory processes. This memory monitoring skills are called metamemory.
       Following a task designed to explore the role of visual detail and propositional content in determining age-related differences in memory for lateral orientation, metamemory questions were proposed to the subjects -- postdictions -- to examine metacognitive processes. Simple vs. complex and concrete vs. abstract slides of black-and-white drawings were used. Sensitivity of memory was tested in twelve conditions where half the target pictures was reversed at retrieval. A total of 240 volunteers (ages 17 to 26 and 65 to 78) served as subjects. The metamemory questions generated data in three different metacognitive provinces: 1) the "Positive Feeling" data (subjects who reported that they enjoyed the experiment); 2) the "Complex Easier" and the "Simple Easier" data; 3) the "Visuo-Spatial/Abstract" and the "Labeling/Abstract" data. The A' results of the metamemory subsamples are analyzed to investigate 1) whether correct responses can influence metacognitive performance; 2) whether perception of "complex-easier" as opposed to "simple-easier" and actual performance correlate; and 3) whether individuals using "visual" as opposed to "labeling" strategies exhibit significantly different results.
    Comparison of Older and Younger Driver Responses to Emergency Driving Events BIBA 986
      Esther Kloeppel; Robert D. Peters; Christina James; Jean E. Fox; Elizabeth Alicandri
    This study investigated the responses of older and younger drivers during performance of emergency maneuvers in an interactive driving simulator. Thirty-six drivers, equally distributed among three age groups (20-29; 35-44; 65-74) participated in the 20 mile simulated drive, during which they encountered four emergency events. Two baseline segments were also collected. The emergency events were situations where other vehicles performed unexpected maneuvers: pulling out in front of the subject's car from a side street, and turning left in front of the subject's car at an intersection. Information on driver performance variables, overall avoidance, and emergency avoidance response time was collected. Older drivers were not different from younger or middle-aged drivers in avoidance response time, speed, deviation from the speed limit, brake pedal force, and overall avoidance. Age differences were found in lateral placement at intersections. Older drivers drove further to the right of the lane center than younger and middle-aged drivers. It is believed that this is a result of a general conservatism of older drivers. This research narrows the scope of investigations into intersection accidents. Older drivers did not exhibit increased avoidance reaction times. In this experiment, subjects were not performing turning maneuvers. Future research should be directed, when possible, toward investigating driver behavior when making turning maneuvers across traffic.
    Postural Disequilibrium Following Adaptation to Virtual Environments: Concern for Post Simulator Activity BIBA 986
      Kevin S. Berbaum; Robert S. Kennedy; Curt C. Braun
    Virtual environment technology may increase the usefulness of simulation in military training by improving on some aspects of current technology and by permitting simulation to be applied in training domains not currently addressed. However, like its forerunner -- current simulation technology, there may be some hazards associated with its use. To assess changes in postural stability associated with adaptation to simulated environments, measures of simulator sickness and postural stability were taken from 127 military pilots completing flight training in a helicopter simulator which included full 6 degrees of motion and 60 inches of travel. Before and after each of ten simulator flights flown every weekday for two weeks, the participants completed a Simulator Sickness Questionnaire, a self report symptom checklist that may be scored for nausea, visuomotor and disorientation. In addition, before-and-after tests of walking and standing steadiness were completed. Postflight sickness was significantly greater than preflight. Postflight postural stability was significantly lower than preflight stability. Both motion-sickness and postural stability decreased across ten exposures to the simulator. That individuals adapt to the simulator is indicated the reduction in reported motion-sickness following repeated exposure. However, as they become increasing adapted to the simulated environment, they experience a reduction in postural stability upon leaving the simulator. As they become adapted to the simulator, they are less well adapted to the real world. The implications of an individual who is less able to function in the real world extend far beyond the training effectiveness of a simulator or the symptoms experienced by a trainee. Individual adapted to a virtual environment should not be expected to interact appropriately with the now non-adapted real world.
    Human-Systems Integration Standards Development Facility: Human Factoring NASA's Human Factors Document BIBA 987
      Cletis R. Booher; Jacquie Minton
    NASA-STD-3000, the Man-Systems Integration Standards (MSIS), represents NASA's ongoing effort to capture the human engineering requirements related to crew interfaces found aboard manned spacecraft. With a basic volume (Volume I) serving as the requirements base, separate volumes have been developed for specific application to evolving programs, and this family of documents is maintained as current as possible through a process of continuing evaluation and update.
       As the family grew over the years, and as the individual volumes expanded to include more data, the goal was always to maintain a logically organized and user friendly format. User feedback has been of great assistance in the pursuit of this goal.
       With document format, as well as content, being fair game for the review and update process, there have been as series of seemingly simple changes over time that have enhanced the user's ability to derive necessary data from the various volumes.
       Human factoring the MSIS has become a challenging but rewarding task.
    TAV-8B Flight Test Results for the Demonstration of an Airborne 3D Audio Cuer BIBA 987
      Marion P. Kibbe; Daniel J. Francis
    This poster presents a summary of pilot opinions on the utility of an airborne 3-Dimensional (3-D) audio cuer and noise cancellation system which was developed at Armstrong Medical Research Laboratory and demonstrated at China Lake, CA in a TAV-8B Harrier. The 3-D audio cues were heard through a customized head set and helmet, equipped with a head tracker coupled to the audio cuer. The helmet was equipped with active noise cancellation which when turned on would subtract ambient aircraft noise. 3-D audio, in both straight and level flight and while maneuvering, was used to provide: 1) direction of arrival information of three ground based targets or threats; and 2) localized separation of two communication channels. There were eleven test flights.
       Pilots reported that the 3-D audio system worked well, giving azimuth cues with approximately 15 degree accuracy, which they felt was adequate to orient attention toward a threat or target location. Elevation cues were less accurate, providing only two rough judgments: low or high. They felt that the greatest promise for future implementation would be to couple 3-D audio to a threat warning system, allowing faster but not more accurate localization of a threat. Communication channel separation was helpful; many aviators thought that 3-D localization of the communication channels would provide more utility. Active noise cancellation seemed to work, but most aviators did not feel that a major requirement existed for the system in tactical fixed wing aircraft.
    Automation and Accountability in a Low-Fidelity Flight Task BIBA 988
      Susan T. Heers; Christopher A. Marchioro; Kathleen L. Mosier; Linda J. Skitka
    Automation and automated aids are developed and implemented to enhance and supplement human capabilities. However, increased automation in operational arenas such as aircraft cockpits has led to the emergence of "automation bias" in decision makers, i.e., errors resulting from reliance on automated feedback as a heuristic replacement for vigilant information seeking and processing. A potential strategy for ameliorating automation bias is to impose accountability on the decision maker. Accountability in this context refers to the need to justify one's performance to and be evaluated by important others.
       The present study is the first step in a research program investigating the existence of automation bias and incorporating accountability demands as an ameliorative variable. Forty subjects performed various flight analogous tasks. An Automated Monitoring Aid displayed information and response directives to assist in two of the tasks, but was not 100% accurate. A supplemental system containing the correct responses was provided and could be utilized at the subjects' discretion to verify the automated directives. The primary manipulations were the degree of accountability imposed on the subjects and the reliability of the automated system. Preliminary results show that subjects verified less than 65% of the directives, indicating a bias towards following the automation. This is reflected by the lower accuracy scores of the subjects in the unreliable automation condition. Use of the supplemental system also created a performance trade-off. As the number of verification steps increased, response times lengthened and flight performance deteriorated, the latter effect exacerbated under more difficult tracking conditions. Verification behavior, however, was not differentially affected by either manipulation. While the accountability manipulations in this study do not appear to ameliorate the automation bias, more research needs to be conducted to determine task and environmental variables which interact with accountability demands.
    Visual-Motion Cueing in Altitude and Yaw Control BIBA 988
      Walter W. Johnson; Jeffery A. Schroeder
    Research conducted using the Vertical Motion Simulator at the NASA Ames Research Center examined the contributions of platform motion and visual level-of-detail (LOD) cueing to tasks that required altitude and/or yaw control in a simulated AH-64 Apache helicopter. Within the altitude control tasks the LOD manipulation caused optical density to change across altitudes by a small, moderate, or large amount; while platform motion was either present or absent. The results from these tasks showed that both constant optical density and platform motion improved altitude awareness in an altitude repositioning task, while the presence of platform motion also led to improved performance in a vertical rate control task. The yaw control tasks had pilots sit either 4.5 ft in front of the center of rotation, or at the center of rotation. The pilots were required to regulate their yaw, while the platform motion was manipulated in order to present all combinations of the resulting rotational and lateral motion components. Ratings of simulation fidelity and sensed platform motion showed that the pilots were relatively insensitive to the rotational component, but highly aware of the lateral component. Together these findings show that 1) platform motion cues are important when speed regulation is required during altitude change; 2) platform motion contributes to the perception of movement amplitude; 3) lateral, but not rotational, motion cues are essential to the perception of vehicle yaw; and 4) LOD management yielding constant optical density across altitudes improves altitude awareness.
    Using Discrete Event and Air Traffic Control Simulations in Teaching Human Factors in System Development: Part 1 -- Syllabus BIBA 989
      W. F. Moroney; J. A. Cameron
    This poster describes a project, utilized in a system development course, which provided students with an opportunity to respond to a simulated Request for Proposal (RFP) based on a real world problem. Specifically, students were asked to upgrade the workstation utilized by controllers at a Terminal Radar Approach Control (TRACON) facility. The system was simulated on a PC-based computer, and students were required to:
  • 1) be knowledgeable about the tasks performed by an air traffic controller
        operating at a Terminal Radar Approach Control (TRACON) facility,
  • 2) develop basic skills in discrete event simulation (Micro Saint) which would
        allow them to demonstrate the impact of changes which they proposed to
        upgrade the system,
  • 3) prepare a response to a simulated RFP, and
  • 4) be part of the development process. Selected literature and references were provided, and the students functioned as a team to develop the response to the RFP. The course emphasized both the process of developing the response and the contents of the product.
       If our students are to have the skills needed to work in today's information age, they must be exposed to fairly complex systems and know how to use simulations to evaluate tradeoffs effectively. It is the authors' hope that this experience has provided them with such an opportunity.
  • Using Discrete Event and Air Traffic Control Simulations in Teaching Human Factors in System Development: Part 2 -- The Student's Experience BIBA 989
      J. A. Cameron; K. D. Packingham; M. J. McCloskey; J. J. Jordan; S. Krsacok; T. M. Brothers; W. J. Rankin; W. M. Sorgenfrei
    This poster describes the experience of students enrolled in a course in Human Factors in System Development which included an introduction to the use of discrete event simulation and an opportunity to participate in the design of enhancements to a complex system. A simulated Request for Proposal outlined the parameters of the task to which students responded. This poster outlines the process used by the students in preparing and presenting their proposal and describes their recommended enhancements. It concludes with a description of some of the lessons learned through the overall class experience.
    The Relationship between Interest and Learning in an Introductory Human Factors Course BIBA 990
      Marc Resnick
    For several years, the Human Factors and Ergonomics Society has allotted a session at each annual conference to the discussion of human factors courses. Topics such as appropriate material to include, the use of projects and the discussion of textbooks have been addressed. However, a more basic principle needs to be considered. How do we promote and maintain students' interest in human factors as a field of study? Even more basic is the question of whether interest in the human factors will promote achievement, and if so, how? Previous research has investigated the relationships between interest and learning in elementary and high school students in topics from baseball and circuses to physics and math. How much of this research applies to human factors and to college level courses in general remains to be seen.
       Ten undergraduate industrial engineering students taking human factors for the first time evaluated chapters of two books over the course of a semester. They rated sections of 24 chapters on arbitrary scales for their interest in the material, the clarity of the material, and the usefulness they perceived the material to be to a practitioner of human factors.
       Interest and clarity (r = 0.57, p<0.01), interest and usefulness (r = 0.81, p<0.01), and clarity and usefulness (r = 0.36, p<0.01) were all positively correlated. Order was significantly correlated with interest (r = 0.29, p<0.01), clarity (r = 0.26, p<0.01) and perceived usefulness (r = 0.21, p<0.01). Correlations between course grades and interest (r = -0.40), clarity (r = -0.31), and perceived usefulness (-0.22) were all negative.
       Implications for the design of introductory human factors courses are discussed. The necessity for taking steps to maintain students' interest throughout the semester is reviewed. Evidence is presented for including hands-on assignments such as group projects, laboratory homework and small group discussions.
    Challenges of Operationally Testing a Test and Evaluation Tool BIBA 990
      Valerie Gawron; Scott Smith; Thomas Joseph; Aric Turner; Brad Purvis
    Two years ago, a need was identified to develop an integrated set of computer-based tools to support crewstation evaluation during flight test. In response to that need, the Test Planning, Analysis, and Evaluation System (Test PAES) was developed. During the development, almost 300 requirements were identified. In addition, the operational domain was expanded from flight test to both ground simulation and operational test in uninstrumented aircraft. The user population also expanded geographically. Finally, constraints of cost and use of current commercial-off-the-shelf (COTS) hardware and software were imposed. To meet these numerous and diverse needs, Test PAES took on numerous and diverse components including: 1) a computer system composed of COTS hardware and both COTS and custom software, 2) a set of databases, 3) two models: one to predict the effects of time delay on debrief data accuracy and the other to predict system performance in operational use, 4) a Structured Crewstation Evaluation Process (SCEP), and 5) Structured Test Procedures (STPs).
       How do you test such a complex system efficiently? How do you obtain the most useful data for basing decisions to modify the system? How do obtain user comments every time they have a comment? These questions and potential solutions were addressed in the poster.
    Adrenergic Stress and Performance on a Command and Control Task BIBA 991
      Karl F. Van Orden; Sandra L. Benoit; Glenn A. Osga
    Performance on a simulated antiair (AAW) and antisurface-ship (ASUW) warfare task was examined as a function of task-independent cold air-induced adrenergic stress. Twenty subjects, trained and experienced in modern naval warfighting systems, completed the study. The 54-minute scenario task required subjects to seek out information on a tactical display regarding the appearance of new symbols, respond to questions regarding general symbol recognition, and respond to questions concerning specific, scenario dependent, symbol characteristics. Near the conclusion of the scenario, subjects were also ordered to fire missiles at designated targets and to fire missiles at targets of their own choosing. Performance on the task was recorded while subjects were exposed to room air temperatures of either 22°C or 4°C. The cold exposure paradigm produced an adrenergic stress response -- as indicated by significantly elevated pulse and norepinephrine levels. Results showed task performance was nearly identical for both groups for the scenario as a whole. However, during a 12-minute portion of the scenario when hostilities were simulated, the ambient and cold stressed groups demonstrated divergent behavioral response patterns; stressed subjects were more liberal with missile fire, and more conservative in their responses to the scenario dependent questions. The responses to the symbol recognition questions during the hostilities portion of the scenario were equivalent for both groups. These results are indicative of a shift in the allocation of cognitive resources as a function of adrenergic stress, and have implications for the development of complex human-machine interfaces.
    Reliability of the Nerve Conduction Monitor in Repeated Measures of Median and Ulnar Nerve Latencies BIBA 991
      I. Washington; M. Shih; S. Burastero; P. Tittiranonda
    In response to the growing number of reported cases of CTS, many companies have introduced screening tools to detect CTS in its early stages. Formal nerve conduction studies are considered the "gold standard" for CTS diagnoses, and abnormality in nerve function is considered one of the earliest indicators of a developing compression neuropathy. This study evaluates the reliability of a portable nerve conduction monitor (Nervepace S-200, Neurotron Medical, New Jersey) which measures both motor and sensory nerve latencies. This instrument's portability and ease of use have made it potentially attractive to health and safety officers interested in conducting active surveillance in the field. It may provide inexpensive, fast, and reliable results in settings where formal nerve conduction testing is infeasible.
       In this study, we examined the reliability of the portable monitor in measuring median and ulnar nerve latencies. All testing was performed by one operator on 25 volunteer subjects between the ages of 20 and 35,10 male and 15 female, who had no prior symptoms of CTS. The median motor, median sensory, ulnar motor, and ulnar sensory latencies of each subject were measured at the same time each day for three consecutive days. The results showed a high degree of intrarater reliability. No significant differences were found in latencies between days among any of the 4 groups of measurements (p < 0.05), and the coefficient of variation was less than 0.2 ms. for all measurements, with the variation being highest in motor latency measurements of the ulnar nerve.
    A New Technique for Early User Evaluation of Entertainment Product Interfaces BIBA 992
      J. H. D. M. Westerink; P. J. Rankin; G. M. M. Majoor; P. S. Moore
    There is a bewildering range of alternative multi-media products which could be designed and developed to entertain the consumer. If sensitively handled, early evaluation by users can guide the completion of a balanced product design and appropriate interface. The relative incompleteness of early prototypes, however, is often an obstacle for engaging real users in a deep evaluation and therefore reduces the choice of suitable evaluation methods. In this poster, we discuss the merits and drawbacks of a known method (the 'Co-discovery' method) in which pairs of subjects explore prototypes together. We also present a new extension to the method (the so-called 'Listener' technique) in which the pair are then asked in an informal setting to describe their experience to a third subject.
       These methods were applied to a novel product concept, VideoMix, for teenagers to make their own pop-videos. The user can assemble digital video and synchronized animations to interpret stored music tracks on Compact-Disc Interactive (CDi). VideoMix aims to stimulate consumer creativity by drawing teenagers smoothly along a learning curve, from experiential pleasures in playing along with the music, into reflective pleasures of personalization and a sense of ownership through finer controls and more functionality off-line.
       We find that, when confronted by a combination of both hands-on testing of a working prototype and hands-off walk-through of the complete product concept, the users generate invaluable input to the (re-) design process, though of different kinds. Co-discovery, employing subject pairs who are friends, is very successful in this respect. Furthermore, the new Listener technique is of great help for eliciting the users' summarized impressions of the product concept in their own terminology and order of recall, and for assessing the relative value of proposed features. These methods are especially suitable for early evaluation of entertainment products, where pleasure plays perhaps an even more important role than task efficiency.

    SYSTEM DEVELOPMENT: Development Activity at the System Level [Lecture]

    The Russian Approach to Task Analysis and Ergonomics Design BIBA 993-997
      Gregory Bedny; David Meister
    Political changes in the East now make available to Western Human Factors scientists information about ergonomics developments in the former Soviet Union. This paper indicates the similarities and differences between American and Russian ergonomics viewpoints.
    Beyond the Bounds of the Human Factors Tool Kit: Computer-Human Interface Design in a Complex System BIBA 998-1002
      Douglas L. Miller; G. Jonathan Wolfman; R. Timothy Mullins; Colin Crehan
    To achieve the potential of human factors involvement in computer-human interface design, human factors engineers must transition from being isolated specialists to integrated components of the mainstream life-cycle development process. A fundamental obstacle to this transition has been the limitations associated with traditional human factors methods. The search for better methods has resulted in a recent evolution in the human factors tool kit. While this evolution has increased tool kit utility, it is not yet as robust as needed for the development of complex systems. For the past several years the human factor team at Coral Federal Systems Company has been the central focus of system design and development activities for the Tower Control Computer Complex (TCCC). The TCCC will replace most Federal Aviation Administration airport tower cab equipment with advanced workstations and software. At least five limitations have been encountered with the traditional human factors tool kit due to the complexity of the TCCC and the environment in which it will be used. This paper describes these limitations, and the alternatives that have been successfully employed to produce an operationally suitable computer-human interface as part of an integrated life-cycle effort. At the heart of these remedies were the use of a dedicated team of representative end-users and a variety of non-traditional design evaluation techniques, throughout the project life-cycle. Among these techniques was the evaluation of laboratory prototypes using a "hands-off' intellectual review process and electronically supported group-based evaluations.
    What We Do with Human Factors Data and Theory: A Flight Deck Design Philosophy for the High-Speed Civil Transport Aircraft BIBA 1003-1007
      K. Michael Dresel; David D. T. Pepitone
    This paper reports on the results and lessons learned from constructing a design philosophy for a new aircraft. The High Speed Civil Transport aircraft is the next-generation supersonic transport, planned for initial operating capability in 2005. Current objectives for the aircraft include cruise speeds of Mach 2.4, ability to take off and land in low visibility, and restricted forward vision. These objectives necessitate consideration of major changes in some of the functions currently allocated to the human flight crew. An explicit design philosophy was defined as the first step in ensuring that system development proceeded with clear emphasis on supporting the human operators in accomplishing the goals of transporting their passengers and cargo safely, comfortably, efficiently and on schedule. This paper discusses the development and details of the integrated flight deck design philosophy that will be used to guide the development of a High Speed Civil Transport flight deck. The paper describes
  • the goals, scope and benefits of the flight deck design philosophy;
  • the effect on the current system development process;
  • the method used to produce the design philosophy;
  • examples of the philosophy and guideline statements, with rationale; and
       finally, suggestions for improving the transfer of basic and applied
       research into the system design process.
  • Evaluation of Complex Human-Machine Systems Using HFE Guidelines BIBA 1008-1012
      John M. O'Hara
    The purpose of this paper is to discuss the role of human factors engineering (HFE) guidelines in the evaluation of complex human-machine systems, such as advanced nuclear power plants. Advanced control rooms will utilize human-system interface (HSI) technologies that can have significant implications for plant safety in that they will affect the ways in which plant personnel interact with the system. In order to protect public health and safety, the U.S. Nuclear Regulatory Commission reviews the HFE aspects of plant HSIs to ensure that they are designed to HFE principles and that operator performance and reliability are appropriately supported. Evaluations using HFE guidelines are an important part of the overall review methodology. The Advanced HSI Design Review Guideline (DRG) was developed to provide these review criteria. This paper will address (1) the issues associated with guideline-based evaluations, (2) DRG development and validation, and (3) the DRG review procedures.

    SYSTEM DEVELOPMENT: Steps in the System Development Process [Lecture]

    Operator Role Definition: An Initial Step in the Human Factors Engineering Design of the Advanced Neutron Source (ANS) BIBA 1013-1017
      H. E. Knee; P. F. Spelt; M. M. Houser; W. E. Hill
    The Advanced Neutron Source (ANS) is a new basic and applied research facility sponsored by the U.S. Department of Energy that is proposed for construction. It will provide neutron beams for measurements and experiments in the fields of materials science and engineering, biology, chemistry, materials analysis, and nuclear science. The facility will provide a useful neutron beam flux that is at least five times more than is available at the world's best existing facilities. It will also provide world-class facilities for isotopes production, materials irradiation testing, materials analysis, and the production of positrons. ANS will be unique in the United States in the extent to which human factors engineering (HFE) principles will be included in its design and construction. Initial HFE accomplishments include the development of a functional analysis, an operating philosophy, and a program plan. In fiscal year 1994, HFE activities are focusing on the role of the ANS control room reactor operator (RO). An operator-centered control room model was used in conjunction with information gathered from existing ANS system design descriptions and other literature to define RO responsibilities. From this list, a survey instrument was developed and administered to ANS design engineers, operations management personnel at Oak Ridge National Laboratory's High Flux Isotope Reactor (HFIR), and HFIR ROs to detail the nature of the RO position. Initial results indicated that the RO should function as a high-level system supervisor with considerable monitoring, verification, and communication responsibilities. The relatively high level of control automation has resulted in a reshaping of the RO's traditional safety and investment protection roles.
    Network Analysis as a Technique to Guide the Task Analysis of ATIS/CVO BIBA 1018-1022
      John D. Lee; Mireille Raby
    Applications of Advanced Traveler Information Systems (ATIS) and Commercial Vehicle Operations (CVO) include technology ranging from simple sensors and alarms to complex combinations of databases and displays. This range of technology will impose a variety of task demands on drivers, and these demands need to be cataloged. Without a means of focusing the task analysis describing these systems, a complete description of all possible interactions among the potential functions of ATIS/CVO systems would be intractable. To address this problem, we have adopted network analysis techniques from sociological and anthropological studies of social groups as a tool to examine complex systems and to guide a task analysis. Network analysis provides a quantitative analysis of information flows that link system functions that can focus a task analysis on important functions and critical interactions between these functions. This paper describes measures of centrality and clusters and how these measures can guide any complex task analysis in the same way it focused the task analysis of ATIS/CVO systems.
    Estimating Cognitive Complexity and the Need for Cognitive Task Analysis BIBA 1023-1027
      Michael J. DeVries; Sallie E. Gordon
    Because an increasing number of systems are being developed to support complex cognitive functioning, task analysis is commonly being augmented wills cognitive task analysis, which identifies cognitive processes, knowledge, and mental models relevant to task performance. Cognitive task analysis tends to be lengthy and time-consuming, so designers frequently ask how they might know if it is actually necessary for a specific project. In this paper, we assume that much of the need for cognitive task analysis depends on the inherent "cognitive complexity of the task. We present a model of cognitive complexity, and show how it was used to develop a computer-based tool for estimating relative cognitive complexity for a set of tasks. The tool, Cog-C, elicits task and subtask hierarchies, then guides the user in making relatively simple estimates on a number of scales. The tool calculates and displays the relative cognitive complexity scores for each task, along with subscores of cognitive complexity for different types of knowledge. Usability and reliability were evaluated in multiple domains, showing that the tool is relatively easy to use, reliable, and well-accepted.
    An Integrative Bargaining Paradigm for Investigating Multidisciplinary Design Trade-Offs BIBA 1028-1032
      Clifford E. Brown; Jonathan A. Selvaraj; Brian S. Zaff; Michael D. McNeese; Randall D. Whitaker
    In design teams, decision making entails negotiation among parties pursuing common goals with potentially divergent interests and objectives (Bucciarelli, 1988). In multidisciplinary design teams, these parties negotiate from perspectives further biased by their respective backgrounds, expertise, and roles. System design can be improved if we better understand how technical data are communicated and assimilated, how mutually advantageous tradeoffs are discovered, and how the managing of design tradeoffs can best be supported.
       As part of our larger research effort in Collaborative Design Technology, we are examining the processes by which integrative design tradeoffs are realized, in preparation for enhancing these processes through data visualization and communication tools facilitating mutual understanding and decision making. This initial report describes our work to date in creating and validating an experimental paradigm to serve as a testbed for subsequent studies of multidisciplinary design practice. This paper describes the paradigm and the initial attempts to demonstrate its ecological validity. This initial validation effort involved a comparison of novices and experts in the field of design and their performance on the design decision making task. We found that experts performed better than novices on the design task, which provided initial validation support for the experimental paradigm.

    SYSTEM DEVELOPMENT: Tools in the System Development Process [Lecture]

    Using a Distributed Interactive Simulation Environment to Investigate Machine Interface and Training Requirements BIBA 1033-1037
      Beverly J. Winsch; Nancy K. Atwood; Kathleen A. Quinkert
    In anticipation of changes brought about by increasingly powerful technologies and systems, the Combat Vehicle Command and Control (CVCC) program evaluated the use of an automated command and control (C2) system developed for the Abrams series tank. The system included a prototype C2 device with map display, navigation and digital messaging capabilities, an automated target acquisition system, and digital workstations in a Tactical Operations Center. Results yield a number of recommendations and represent the culmination of a five-year research program which has successfully utilized an iterative approach to investigate training and system design requirements for the prototype CVCC system. Data are discussed within the context of the need for design and training efforts aimed at alleviating the growing information management requirements faced by users of emerging C2 technologies.
    Automated Data Analysis BIBA 1038-1042
      Edward M. Connelly
    Automated Data Analysis (ADA) is a powerful, flexible analysis tool for systematically building performance determinants. ADA is designed to assist analysts in assessing the linkage from system processes to outcomes. It allows for or accounts for the varying conditions during performances and permits formation of complex analyses which extract relevant information from multiple variables while ignoring irrelevant data from the data base. In particular, ADA permits:
  • 1. Focus on selected portions of the performance space,
  • 2. Characterization of the system processes and examination of the relationship
        between system processes and outcomes,
  • 3. Building analysis variables which can be made simple or as complex as
        necessary as knowledge of the importance of system processes to outcomes is
        obtained, and
  • 4. Use of powerful restructured analysis tools.
  • Usability Tools and Ease of Use: Technological Misfits? BIBA 1043-1047
      Timothy J. Pavlick; Denise Benel; Richard Horst; Susan Eaton; Steven Gregory
    Often times usability professionals employ methods and tools that, in themselves, are not very usable. As technology proliferates through a more diverse user population, usability methods must be adapted to accommodate more general usage by this growing and diverse population. This paper describes a usability testing center life-cycle plan containing design, implementation and integration which incorporates usability into the design of the center equipment and its test conduct procedures document. The IRS's vision of system developers forming 'visiting test teams' which rotate through the center, necessitated that the center be generally accessible. A touch screen equipment interface allowed the presentation of a common interaction style for all audio/video recording equipment in the usability center. Likewise, an integrated behavioral logging/video editing facility enabled quick production of video highlights tapes. The contractor also provided three day usability engineering and testing training courses for visiting test teams as a means of making the process more accessible. Following training, preliminary tests were conducted in the Usability Center which allowed an opportunity to usability test the center and procedures. Based on reports that documented the results of these preliminary tests, significant changes to the test conduct procedures, center and equipment layout made the method more accessible to visiting test teams. The conclusion of the effort resulted in a certification plan for the government to take over full usability testing operations at the lab.
       The end result was acceptance and understanding, by system developers, of the usability process and its value to system design.
    Operator-Centered Control of a Semi-Autonomous Industrial Telerobot BIBA 1048-1051
      Philip F. Spelt; Sammy L. Jones
    This paper presents work done by Oak Ridge National Laboratory and Remotec, Inc., to develop a new operator-centered control system for Remotec's Andros telerobot. Andros robots are presently used by numerous electric utilities, the armed forces, and numerous law enforcement agencies to perform tasks which are hazardous for human operators. This project has automated task components and enhanced the video graphics display of the robot's position in the environment to significantly reduce operator workload. The procedure of automating a telerobot requires the addition of computer power to the robot, along with a variety of sensors and encoders to provide information about the robot's performance in and relationship to its environment. The resulting vehicle serves as a platform for research on strategies to integrate automated tasks with those performed by a human operator. The addition of these capabilities will greatly enhance the safety and efficiency of performance in hazardous environments.

    SYSTEM DEVELOPMENT: Panel

    Human Factors Certification of Systems BIBA 1052-1056
      John A. Wise; V. David Hopkin; Paul Stager; Kelly Harwood
    There is growing interest in the regulatory organizations (e.g., FAA, ICAO) to establish human factors based certification procedures for aviation technologies. This panel will discuss some of the issues debated during an international workshop on human factors certification of aviation technologies.

    SYSTEM DEVELOPMENT: Developing Human Factors Guidelines for Intelligent Vehicle/Highway Systems: Third Progress Report [Symposium]

    Developing Human Factors Guidelines for Intelligent Vehicle/Highway Systems: Third Progress Report BIB --
      Barry H. Kantowitz; Thomas J. Triggs
    Predicting Driver Behavior Using Advanced Traveler Information Systems BIBA 1057-1061
      William A. Wheeler; John D. Lee; Mireille Raby; Rhonda A. Kinghorn; Alvah C., Jr. Bittner; Marvin C. McCallum
    As a part of the Intelligent Vehicle Highway System (IVHS), Advanced Traveler Information Systems (ATIS) will offer tomorrow's drivers significantly expanded capabilities for getting where they want to go safely and efficiently. Vehicle-based navigation systems combined with information on highway conditions and services have the potential for improving driver performance. Though ATIS may offer considerable advantages, the system design must be consistent with the primary tasks of controlling and operating the vehicle. This paper describes an attempt to identify the likely interaction between what a driver must do to operate the vehicle safely while at the same time using the various ATIS systems. As such, it is an attempt to visualize what driving with these advanced systems will be like and to translate that vision into standard human factors task analytic techniques. Though a broad range of ATIS systems and functions were addressed in this project, this paper will address the macro-level task analyses that resulted from the examination of 165 tasks related to ATIS use.
    Driver Acceptance of Unreliable Route Guidance Information BIBA 1062-1066
      Richard J. Hanowski; Susan C. Kantowitz; Barry H. Kantowitz
    Human factors research can be used to design safe and efficient Advanced Traveler Information Systems (ATIS) that are easy to use (Kantowitz, Becker, & Barlow, 1993). This research used the Battelle Route Guidance Simulator (RGS) to examine two important issues related to driver behavior and acceptance of ATIS technology: (1) the effect of route familiarity on ATIS use and acceptance and (2) the level of information accuracy needed for an ATIS to be accepted and considered useful. The RGS included two 486 computers that provided drivers with real-time information and traffic reports. Drivers used a touch screen to select routes on one computer monitor and watched the results of their selection (i.e., real-time video of the traffic) on a second computer monitor. Drivers could use the system to obtain information about the traffic conditions on any link before traversing a route. In this experiment, subjects were exposed to four experimental conditions involving manipulation of the driver's familiarity with the route and the reliability of the traffic information obtained from the RGS (i.e., 100%, 71%, and 43% accuracy). The driver's goal was to reach the destination as quickly as possible by avoiding heavy traffic. The results indicated that drivers were able to benefit from system information when it was reliable, but not when it was unreliable. Trust ratings for the 43% accuracy group were significantly higher at the beginning of the four trials than at the end. Also, drivers were more apt to rely on the ATIS and accept information given in an unfamiliar traffic network versus a familiar one.
    Identification of Desired System Features in an Advanced Traveler Information System BIBA 1067-1071
      Rhonda A. Kinghorn; Alvah C., Jr. Bittner; Barry H. Kantowitz
    This study evaluated consumer acceptance of variations of a currently available Advanced Traveler Information System (ATIS). We also wanted to determine if video demonstration was sufficient for observers to grasp the fundamentals of ATIS devices. A total of 109 licensed drivers viewed two videotaped demonstrations of TravTek, and then completed questionnaires. Principal Factor Analyses resulted in patterns of desired features and other composite variables used in regression analyses. Basic map features (e.g., vehicle position/location, outline of route) and voice features were the two dominant feature patterns. Other composite variables included understanding of the system capabilities, trust, self-confidence, tolerance of system errors, demonstration fidelity and attention. Results of the regression analyses indicated that different variables were significant predictors of each pattern of desired features.
    The Effects of Sensory Modality and Information Priority on In-Vehicle Signing and Information Systems BIBA 1072-1076
      Michael A. Mollenhauer; Jaesik Lee; Ken Cho; Melissa C. Hulse; Thomas A. Dingus
    During this study, subjects drove an interactive driving simulator and were presented road sign information from a visual dash-mounted LCD display or from digitized auditory voice. information priority was also manipulated in that subjects received all sign information typically present in the roadway environment, or only "filtered" high priority regulatory and notification information. The effects of display type and filtering on information recall, driver performance, and driver preferences were measured. The results indicate that presenting information in an auditory mode results in a higher level of road sign information recall, but also decreases the subjects' driving performance when compared to a visual display. Subjects were also able to recall more road sign information and drove at a higher level of performance during the filtered conditions. Subjects rated auditory information as more distracting than visual information.

    TEST AND EVALUATION: Panel

    T&E: Where are We Now and Where are We Going? BIBA 1077-1079
      David Meister; Thomas P. Enderwick; Alvah C., Jr. Bittner; James C. Geddie; William F. Moroney; Frederick A. Muckler
    The purpose of this panel discussion is to examine the present status of test and evaluation (T&E) methodology, to consider the problems it faces, and to project its future in the 21st century. The discussion will attempt to raise the consciousness of measurement specialists to an awareness of the factors that affect their work.

    TEST AND EVALUATION: Test and Evaluation Methodology [Lecture]

    The Critical Incident Technique: Forty Years Later BIBA 1080-1084
      Lawrence G. Shattuck; David D. Woods
    This year marks the both anniversary of the publication of John Ranagan's paper entitled "The Critical Incident Technique" in Psychological Bulletin. In the years since its publication, much has happened in the field of human factors. The critical incident technique is still a common tool among human factors practitioners, though it has often been modified. With a new generation of practitioners in training, it is important to consider what they are learning about this tool. A survey conducted with future human factors professionals highlights some misconceptions concerning the critical incident technique. These misconceptions seem to originate from two sources: the treatment of the technique in human factors textbooks and handbooks, and the use of the technique in current research and application work. Some modern variations of the critical incident technique are discussed. The critical incident technique is viewed as an instance in which human factors specialists are shaping their tools to meet their needs. A set of principles is proposed to guide today's human factors practitioners in conducting cognitively oriented field research.
    Test and Evaluation in a Field Environment: Lessons Learned from a Successful Program of Research BIB --
      Michelle R. Sams; Richard E. Christ
    Strategy-to-Task: Human Factors Operational Test and Evaluation at the Task-Level BIBA 1085-1089
      Terence S. Andre; Samuel G. Charlton
    Human factors operational test and evaluation (OT&E) at the function/characteristic level has not always provided an appropriate balance of addressing both the needs of the system user and the decision-maker. System users are primarily concerned with the characteristics and capabilities of their system. Acquisition decision-makers, on the other hand, are more concerned with force structure and how potential military systems fit within the national military strategy. Human factors OT&E has traditionally considered the user of the system by testing human factors at a characteristic, rather than mission or operational task-level. In order to address the needs of the decision-maker, OT&E has adopted a strategy-to-task formulation that can have the undesirable side effect of decreasing the visibility of human factors test results. Because human factors measures are considered at the system function/characteristic level, significant human performance/human-machine interface issues are not always visible at the level of higher task elements and missions. Systems which require significant human-in-control or human-in-the-loop operability may lend themselves to consideration at the task-level. Testing human factors at the task-level within the strategy-to-task framework provides both the decision-maker and user with the necessary information to buy and properly operate the system.
    Understanding Perceived Image Quality: New Applications for Verbal Protocol Methodology BIBA 1090-1093
      Gerhard P. Deffner; Mashiho Yuasa
    The present study investigated the effectiveness of 'cued retrospective verbalization' used in combination with eye-movement recordings as memory aids to study image quality evaluation of display products. That is, recordings of eye movements during image quality judgments were played back as cues for recalling and verbalizing thoughts which occurred during prior image evaluation. The study also examined the consistency of subjects' image quality evaluation responses over time and whether retrospective verbalization influenced and altered subsequent preference. The experiment consisted of two sessions separated by a 60-hour interval. Subjects in the experimental condition performed retrospective verbalization in both sessions whereas subjects in the control condition did so only in the second session. Judgments of perceptual quality were stable over time, and there was no indication that cued retrospective verbalization influenced subsequent perceptual evaluation. Moreover, subjects in the experimental condition stated the same critical image characteristics for the reasons of their preferences if they had not changed their choices between the two sessions. When there were changes in preference over time, their verbalization protocol indicated clear shifts of attention from one critical image characteristic to another. Cued retrospective verbalization appears to be an effective tool to examine the processes of display image quality.

    TEST AND EVALUATION: Test and Evaluation of Product Design [Lecture]

    Quantitative Evaluation of Four Computer Keyboards: Wrist Posture and Typing Performance BIBA 1094-1098
      C. Chen; S. Burastero; P. Tittiranonda; K. Hollerbach; M. Shih; R. Denhoy
    The present study focuses on an ergonomic evaluation of 4 computer keyboards, based on a quantitative analysis of wrist posture and typing performance and on subjective analyses of operator comfort during typing. The objectives of this study are (1) to quantify differences in the wrist posture and in typing performance when the four different keyboards are used, and (2) to analyze the subjective preferences of the subjects for alternative keyboards compared to the standard flat keyboard.
    An Experimental Field Test of Automotive Headway Maintenance/Collision Warning Visual Displays BIBA 1099-1103
      Daniel V. McGehee; Thomas A. Dingus; Avraham D. Horowitz
    Motor vehicle crashes resulting from one vehicle striking the rear-end of another are one of the most common types of crashes involving two or more vehicles. The National Safety Council reported (Accident Facts, 1992) that there were approximately 11.3 million motor vehicle crashes in 1991 of which 2.7 million were rear-end crashes (about 23.8% of the total). These crashes accounted for 33% of all collisions involving two or more vehicles. To address the rear-end crash problem, a color LCD display designed to indicate safe following distances was mounted in the instrument panel of an 1990 Olds Toronado Trofeo. The vehicle was also equipped with a laser range finder, forward view video camera, eye view camera, video multiplexer, and computer-controlled video cassette recorder. One hundred and eight drivers from three age groups participated in this field experiment in one of three display symbology conditions. Drivers were not explicitly instructed on how to use the headway displays. Data analyses indicated that (1) the drivers easily understood the displays, (2) those drivers who initially maintained unsafe headways increased their following distance when one of the display symbologies was used, (3) during events where changes in relative velocity (braking) took place, all three symbologies increased the overall headways, and (4) drivers preferred and understood, even better, displays that included graded headway/warning information.
    Signal Word and Color Specifications for Product Warnings: An Isoperformance Application BIBA 1104-1108
      Curt C. Braun; Lori Sansing; Robert S. Kennedy; N. Clayton Silver
    Recent work in the area of color and warnings has indicated that the level of hazard communicated by signal words varies as a function of the color in which they are printed. These findings suggest that signal word and color combinations create a continuum of perceived hazard. Although individual experiments advance the understanding of how color affects the perception of product hazard, explicit data do not always provide label designers and product manufacturers with the tools necessary to successfully apply them. To explore how color and signal words can be jointly used, the present effort applied the isoperformance technique to the problem of specifying signal word and color combinations. This technique identifies combinations of variables that produce equal (iso) levels of warning (performance). Using perceived hazard data from 30 participants, signal words and colors were systematically grouped into categories that conveyed equal levels of hazard. How the isoperformance technique might serve as a tool for label designers is described.
    Evaluation of Half-Mask Gas Collection for Metabolic Measurement in the Workplace BIBA 1109-1112
      Richard S. Farley; Phillip A. Bishop; Paul Ray
    Metabolic measurement is often important in job design and research. Metabolic measurement via indirect spirometry has traditionally used a mouth-piece nose-clip headgear (MNH) method of collecting expired gases. MNH is often uncomfortable, prohibits oral communication, may fail because of loss of nose-clip, and may interfere with or alter some work tasks because of the awkwardness of the headgear. Gas collection masks (MASK) offer the possibility of partially or totally alleviating the problems of MNH but increase the possibility of erroneous measure due to leakage. This study compared the measurement of performance and subject's subjective ratings of a typical MNH vs. MASK (Hans Rudolph model 8900 series) over a range of work loads from light to maximal in 20 well-trained subjects (M=12, F=8). Ventilation (Ve), oxygen uptake (VOW), and respiratory exchange ratio (RER) were collected with a Sensormedics MMC-1, and heart rate (HR) with a Polar Heartwatch. Comfort was assessed with a Likert questionnaire. Results from the final work rate are shown below:
        Variable
        Ve
       
       
       VO2
       
        RER
       
        HR
       
       
       
       
        (L/min)
       (ml/kg/min)
       
       
        (bpm)
        MNH
       
        105,217
       
       52.2
       
       1.13
       
       187
        MASK
       
        115,281
       
       56.6
       
       1.09
       
       187 Results show this MASK produced values as high as the MNH with greater subject comfort at both peak and sub-maximal work. The observed Ve in the MASK condition suggests that leakage was not problematic and there could be a tendency for the MNH to actually impede the measurement. Based on previous preliminary studies and these data, careful use of the MASK is recommended over MNH for work place metabolic or other ventilatory measurements.

    TEST AND EVALUATION: Test and Evaluation Involving Higher-Order Functions [Lecture]

    Information Categorization for BMD Command and Control BIBA 1113-1117
      Beverly G. Knapp; Bruce G. Coury; Annette R. Ensing; Scott L. Godfrey
    Information requirements were investigated for operators in two BMD C2 (Ballistic Missile Defense Command and Control) Centers -- command level and operations level. The objective was to elicit their underlying perceptions and cognitive structure of incoming data items, to address information display and decision support design requirements. Experts judged information items using an unconstrained sorting task, and data were analyzed using multivariate scaling methods. Findings revealed differences between the two groups in clustering of and associations between information items; e.g., more complex associations were found for the command level center. Also, the utility of the scaling algorithms for the large data sets used was assessed.
    Driver Performance Results from the TravTek IVHS Camera Car Evaluation Study BIBA 1118-1122
      Thomas A. Dingus; Melissa C. Hulse; Daniel V. McGehee; Raj Manakkal; Rebecca N. Fleischman
    The TravTek system constitutes a major Intelligent Vehicle-Highway System (IVHS) Advanced Traveler Information System (ATIS) demonstration project. The system provided in-vehicle information via color touchscreen CRT, steering wheel buttons, and synthesized voice. The TravTek driver interface was developed with the intent of providing navigation, service and attractions, and roadway incident and traffic information to the driver. The design of the TravTek interface had as its primary objectives: (1) more effective driver navigation providing the benefit of saving time, (2) easy access to valuable and convenient location information to alleviate stress and increase driving enjoyment, (3) maintenance of safe driving performance during system use and safety improvement facilitated by information for avoiding hazards and for emergency response, and (4) improvement of roadway efficiency to alleviate congestion.
       This paper provides detailed data regarding driver performance and behavioral interactions with four TravTek navigation configurations and two conventional methods of navigation: a paper map and a textual direction list. The results indicate that turn-by-turn information, regardless of its method of presentation, results in effective driving and navigation performance. A moving map display with no supplemental information required high visual attention relative to the other conditions. The other TravTek conditions resulted in lower workload superior navigation performance than the paper map control condition.
    Test and Evaluation of Knowledge Transfer between Task Domains BIBA 1123-1127
      Nong Ye; Donald J. Brinkman
    To examine the transfer of knowledge representation between task domains, working engineers and pharmacists were asked to provide pairwise dissimilarity ratings of concepts in a test domain. The test domain was independent of engineering and pharmacy and was relatively new to both subject groups. A quantitative technique based on the multidimensional scaling and directional statistics was used to test and evaluate group differences in knowledge representation. The results indicated that there was no statistically significant difference between two groups in knowledge representation of the test domain. That is, knowledge representation was domain-specific and was not transferable between task domains.
    Ecological Validity in Laboratory Based Usability Evaluations BIBA 1128-1130
      Patrick W. Jordan; D. Bruce Thomas
    An interview based survey, looking at the suitability of a laboratory facility for usability testing, raised the issue of 'ecological validity'. Ecological validity refers to the extent to which the test environment mirrors the environment in which a product would be used in 'real life'.
       Ten ergonomists, nine of whom used the laboratory, were interviewed. Opinions of the laboratory were generally positive; indeed there was a consensus as to the high value of the facility in the ergonomists' work. However, only one of those asked felt that the laboratory provided as ecologically valid testing environment. Initially, this result seems surprising; if the conditions don't provide ecological validity, this would appear to be a limitation on the laboratory's value.
       This paper considers the concept of ecological validity; discussing the contexts in which it is important and those in which it is not a priority. The extent to which it can be achieved in the laboratory and how it could be achieved are also considered. Generally, the ergonomists were fairly pessimistic about the prospects for this.
       All, bar one, of the ergonomists interviewed also conducted studies outside of the laboratory. The part played by ecological validity in deciding to evaluate in the field is discussed; at what point does the issue become important enough to force evaluations outside of the laboratory?
       Five non-ergonomists from the same organization were also interviewed. Interestingly, they seemed to give ecological validity a higher priority than ergonomists, in terms of influencing the overall value of an evaluation. They were also more inclined to expect ecological validity to be achievable in the laboratory -- by using furniture to create an appropriate range of scenarios. This suggests that, even if scenario creating has no real effect on ecological validity (this was the opinion of many or the ergonomists), it may bring 'propaganda' benefits in terms of influencing commissioners' attitudes towards ergonomists' work.

    TEST AND EVALUATION: Test and Evaluation of Performance-Shaping Functions [Lecture]

    Software Usability Testing: Do User Self-Consciousness and the Laboratory Environment Make Any Difference? BIBA 1131-1134
      Richard T. Barker; David W. Biers
    The purpose of the study was to evaluate the effect of laboratory environment, user self-consciousness, and user experience on the user's subjective evaluation of software usability. The study employed a 2 X 2 X 2 factorial between-subjects design with 2 levels of Laboratory Environment (Cameras and Mirror vs. No Cameras and Mirrors), 2 levels of User Self-Consciousness (Low vs. High), and 2 levels of User Experience (Novice, Experienced). The users were asked to learn, then use, and finally subjectively evaluate a restricted subset of common word processing features over three hours of participation. Day 1 was a training day and Day 2 was a test day. Results indicated that high self-conscious and novice users make more word processing errors. However, they were no more likely to make those errors in the presence of cameras and a mirror. More importantly, the evidence for any effect of the independent variables on subjective evaluation was sparse -- limited to the interaction of self-consciousness and laboratory environment on just three of twelve factors. Moreover, the patter of these interactions indicated that self-consciousness and the laboratory environment did not influence subjective evaluation in any predictable manner. Despite some methodological shortcomings, the conclusion was drawn that these variables do not have a major impact on subjective evaluation of software usability.
    Modeling Operational Criteria for Evaluating Speech Communication BIBA 1135-1139
      Leslie A. Whitaker; Leslie J. Peters
    Evaluation of systems can be conducted best under controlled circumstances which approximate operational conditions. In the present paper, we have examined this thesis as it applies to the test and evaluation of multi-person systems. Our specific focus has been the development of a conceptual model of speech communication requirements and the study of the impact of degraded speech intelligibility on performance of these multi-person systems. To obtain the control necessary to evaluate performance using speech intelligibility, an electronic circuit was developed and employed in a series of simulated tests of operational tasks. The application of these research findings to the task of system test and evaluation is described in the present paper.
    Behavioral Control Characteristics of Performance under Feedback Delay BIBA 1140-1144
      Christopher M. Keran; Thomas J. Smith; Eric J. Koehler; Peter K. Mathison
    As part of a larger research project dealing with technological change, the U.S. Bureau of Mines has initiated a study of the nature and sources of variability in human performance during teleoperation. One important source of performance degradation during remote work is delay in sensory feedback from the remote site to the operator during task execution, caused by such factors as transmission and signal processing lags or inertia/momentum properties of large equipment. To investigate the properties of the behavioral control system under delayed feedback conditions, the Bureau has implemented a pursuit tracking task during which delay is imposed as a continuously varying sinusoidal forcing function. Using 11 subjects, the dynamic characteristics of tracking performance were assessed across a range of variable delay frequencies. Specifically, sinusoidal variations in visual feedback delay between 0 and 400 msec were imposed during a tracking task at frequencies between .05 and 2 Hz. The results show that RMS error, gain (fundamental FFT peak for the cursor signal/fundamental FFT peak for the target signal), and phase (phase angle difference between the target signal and the cursor signal) of the tracking control system are independent of variable feedback delay imposed across a forty-fold range of frequencies. One important implication of these findings is that operators may have limited ability to adapt to feedback delay conditions that may be present during teleoperation of large mobile mining equipment.
    Workload Context Effect: An Elusive Phenomenon BIBA 1145-1149
      Douglas S. Fischer; William F. Moroney; David W. Biers
    The effect of workload context on subsequent performance and workload ratings has crucial implications regarding workload transition. However few studies have examined workload context effects; and those that have, report contradictory results. This study attempts to determine if the failure to find evidence of workload context effects might be attributable to methodological factors such as task duration, task difficulty, and experimental design. Twelve subjects "flew" three sessions of three trials on a computer-based flight simulator, and rated the workload after each trial. A pre-post experimental design presented the first and third trials at a medium level of difficulty while the second (experimental) trial was of low, medium, or high difficulty. Crosswinds of 2, 12, and 22 knots created the levels of low, medium, and high task difficulty. Analyses of the performance and workload data did not reveal significant differences in Trial 3 as a function of prior task difficulty presented in Trial 2. The inability to find workload context effects in the present study suggests that previous inconsistent findings can not be attributed to differences in task duration and experimental design. Rather, it appears that contradictory results may be attributable to differences in the range of task difficulty employed, the workload measurement tool, or both.

    TRAINING: Panel

    Improved Training Methods: Research to Applications BIBA 1150-1153
      Franklin L. Moses; Ed Salas; Janis A. Cannon-Bowers; Ray S. Perez; E. M. Roth; R. J. Mumaw; Angelo Mirabella; Marvin S. Cohen; Gary Klein
    How to train people to make good decisions, solve problems, and so on depends, as does all training, on some form of practice and feedback. The question for behavioral research often is how to improve on these basic requirements. Six panelists describe and discuss their research and experience with the relationship among training and factors such as group dynamics, stress, mental models, and naturalistic requirements. This session includes interaction among the panel and the audience.

    TRAINING: Training on the Technological Frontier [Lecture]

    Side Effects and Aftereffects of Immersion in Virtual Environments BIBA 1154-1157
      Donald R. Lampton; Eugenia M. Kolasinski; Bruce W. Knerr; James P. Bliss; John H. Bailey; Bob G. Witmer
    Immersive Virtual Environment (VE) technology, also known as virtual reality, is being touted as an important new medium for education and training. Other potential applications involve communications, medicine, architecture, astronomy, data handling, teleoperation, and entertainment. A threat to the successful application of this technology is that some users of VE systems suffer unwanted side effects and aftereffects similar to, but not limited to, symptoms of motion sickness. These effects may degrade training effectiveness and jeopardize user safety and well-being. This paper describes the incidence and severity of symptoms we recorded during four different experiments which examined VE training applications. The experiments involved a variety of tasks, simulated environments, and VE systems. We administered a 28 item questionnaire that addressed symptoms related to nausea, eye strain, and dizziness. Significant variation was observed across individuals. In each experiment some users, between 4 and 16%, experienced discomfort to the extent that we terminated their participation. Most users enjoyed the VE experiment but reported some level of discomfort. Our findings indicated that sickness resulting from VE immersion is a potentially serious problem which may not be completely eradicated by improvements in equipment. This paper describes the patterns of effects we observed, discusses the challenges of measuring effects, and outlines future research.
    Learning and Transfer of Spatial Knowledge in a Virtual Environment BIBA 1158-1162
      John H. Bailey; Bob G. Witmer
    Two experiments were conducted to investigate route and configurational knowledge acquisition in a virtual environment (VE). The results indicate that route knowledge can be acquired in a VE and that it transfers to the real world. Furthermore, although it was not explicitly trained, participants acquired some configurational knowledge. Higher levels of interactive exposure to the VE resulted in better route knowledge than less interactive exposure. There was some evidence that more reported presence was correlated with better performance on spatial knowledge tests, while more reported simulator sickness was correlated with worse performance. Finally, performance during VE rehearsals was a strong, consistent correlate of performance on spatial knowledge tests.
    Fidelity and Interactivity in Navigational Training: A Comparison of Three Methods BIBA 1163-1167
      Henry P. Williams; Christopher D. Wickens; Scott Hutchinson
    In two experiments we vary the degree of fidelity of a navigational training environment, and determine how this variance effects the degree of route knowledge and survey knowledge demonstrated in a navigational transfer task. Subjects learn about the geography of a computer-generated world by either (a) studying a 2D map of the world, (b) passively viewing a 3D course through the world on an IRIS display, or (c) actively flying through an IRIS-based simulation of the world. Groups (b) and (c) were yoked. All groups then transferred to a flight along the studied route plan in a high fidelity Evans and Sutherland visual simulation system. In Experiment 1, 60 pilots were assigned to the three training conditions, and the subjects in the two IRIS flight groups were subdivided into categories of low and high visual fidelity of the IRIS world. In this experiment, the active flight training condition yielded most accurate navigation performance (route knowledge), with the map-study group being nearly as proficient, and the passive group much lower. The map-study group had the highest recall of the geography (survey knowledge), but this knowledge did not provide them with any benefit in solving an unexpected navigational problem (functional survey knowledge). Visual fidelity had no effect on any measure of transfer performance. In Experiment 2, the speed and workload of the rehearsal flight was increased, and 10 pilots were assigned to each training group. The results revealed that the increased workload reversed the order of the two computer flight groups, with subjects in the active group who needed to rotate the map, now performing most poorly. The results of both experiments are discussed in terms of the detrimental effects of high workload to geographical learning, and to the dissociation between different kinds of geographical knowledge.
    Training with Images: Real and Representational BIBA 1168-1172
      Sandra S. Bailey
    A total of 60 subjects (24 males and 36 females) participated in a study to determine if a caricature's accentuation of critical cues results in improved recognition of handshapes used in the American Sign Language manual alphabet. Each subject was randomly assigned to one of two training groups. One group trained with photographs of the handshapes and the other group used caricatures. Once mastery of the alphabet was demonstrated, their ability to recognize the handshapes shown in four different modes (positive and negative photographs, and positive and negative caricatures) was tested. In the unrestricted condition, the duration of exposure was not artificially constrained. In the restricted testing condition, the handshapes were displayed at 320 msec, 500 msec and 700 msec. Both speed and accuracy were equally emphasized in the training and in the testing. The findings did not support the superfidelity hypothesis of caricatures. In the unrestricted condition, those trained with photographs responded significantly faster, regardless of mode, than those trained with caricatures. As predicted in the most restrictive display time (320 msec), mean response time was significantly faster with caricatures. This study has direct implications regarding the media used to train American Sign Language. The findings support the use of photographs to depict and to train novices in the ASL handshapes. Further research is needed to determine if these findings hold true as the complexity of the handshake increases.

    TRAINING: Insights into Learning and Interaction [Lecture]

    Training to Reduce the Disruptive Effects of Interruptions BIBA 1173-1177
      Stephen M. Hess; Mark C. Detweiler
    Two multi-session experiments are described in which a complex problem-solving task was interrupted at different stages of practice. In Experiment 1, subjects practiced the main problem-solving task for three sessions, with intermittent interruptions during each session. By the end of Session 3, interruptions which were similar to the main task, in terms of type of material processed and processing demands, no longer disrupted performance as they had in Sessions 1 and 2. In Experiment 2, subjects practiced the same problem-solving task for two sessions without interruptions. The same types of interruptions used in Experiment 1 were introduced in Session 3. Although the main task was well learned by the third session, the interruptions disrupted subjects' main-task accuracies dramatically. These results suggest that training tasks under uninterrupted conditions can lead to excellent performance, but may not allow subjects to develop the kinds of strategies needed to flexibly recover from interruptions when they occur.
    Training Perceptual-Rule Based Skills BIBA 1178-1182
      Neff Walker; Arthur D. Fisk; Donita Phipps; Alex Kirlik
    The results of an experiment are discussed that address how best to train perceptual-rule based skills within a domain where rules correlate to perceptual aspects of a dynamic evolving environment. Participants performed the role of football quarterback where the object of the task was to learn to identify the correct pass receiver in a simulated football system. The correct receiver could always be specified by a set of rules or subtle perceptual cues. Subjects were assigned to one of four training groups which were constructed by complete crossing of rule versus no rule learning and visual enhancement training versus no visual enhancement training. After training trials all subjects transferred to new plays in which new rules or perceptual cues were required. Transfer performance was superior for the participants who received the visually enhanced training. These results are discussed in light of theories of part-task training.
    Perceptual Learning Modules in Flight Training BIBA 1183-1187
      Philip J. Kellman; Mary K. Kaiser
    Differences between novices and experts in many piloting skills may be due to perceptual learning. Sufficient exposure to relevant stimulus variation produces more efficient information extraction, processing of higher-order patterns, and automaticity. Isolating and condensing relevant perceptual experience in part-task environments might accelerate training. Here we report initial studies of two prototype perceptual learning modules (PLMs) for flight training.
       Subjects were either experienced (500-2500 hour) civil aviators or non-pilots. In the Visual Navigation PLM, subjects received brief instruction on aeronautical chart symbology and then viewed 20-second segments of terrain (videotaped from aircraft). Each trial required a speeded, forced choice of the aircraft's location from three possible grid locations on the aeronautical chart. A separate control group received only 20 pro- and 20 post-test trials. In the Instrument Relationships PLM, subjects viewed displays of primary flight instruments and performed a speeded response classifying the flight attitude depicted. In both PLMs, subjects' speed and accuracy were measured over 9 blocks of trials.
       PLMs produced dramatic improvements in speed and accuracy for both non-pilots and pilots. Pilots initially outperformed non-pilots. Non-pilots after 1-2 hours of PLM training were as accurate and faster than pilots before training in both PLMs. The results suggest that PLMs have value for primary and recurrent training, both in aviation and other domains. Appropriately structured PLMs could condense perceptual learning processes that normally occur with extended experience. By fostering greater automaticity of pattern processing, PLMs might allow component skills to be more easily integrated in flight or other complex tasks.
    Learning Complex Visual Stimuli: Effects of Spaced Presentation and Rehearsal on Aircraft Recognition BIBA 1188-1192
      Stephen W. Jarrard; Michael S. Wogalter
    This study examined the effects of three presentation methods (one massed and two distributed) and two visual rehearsal conditions (rehearsal allowed and not allowed) on recognition of complex visual stimuli. The stimuli, photographs of military aircraft, were tested using a different view than the three views given at study. Recognition performance was measured by hit, false alarm, and discrimination indices to assess differences among the presentation and rehearsal conditions. A substantial effect of rehearsal was found. Allowing intervals for, and encouraging, post-exposure imaging increased hit and discrimination scores compared to conditions where post-exposure imaging was prevented. No significant effect of presentation method or interaction with rehearsal was noted. Exploratory analyses suggested that a study strategy involving attention to individual features to be associated with higher recognition performance. Empirical, theoretical, and applied implications of the study are discussed, and suggestions for further research are described.

    TRAINING: Investigating the Unique Contribution of Feedback in Teams: Implications For Training [Symposium]

    Investigating the Unique Contribution of Feedback in Teams: Implications For Training BIBA 1193-1194
      Renei J. Stout
    The vital role played by teams in modern society has become unquestionable. Therefore, it is crucial that viable training technologies are developed and applied to teams. One concept or training technique that has been studied extensively, though in the area of individual skills training and performance, is feedback. Within this arena, feedback has been investigated across a wide variety of task complexes, and it has been generally accepted from this body of research that feedback is beneficial. However, there is a dearth of empirical research on the effects of feedback on team performance. consequently, several questions related to the provision of feedback in teams remain. For example, it is unknown what type and what degree of conceptual training is required to "set the stage" such that trainees are able to appropriately interpret feedback and effectively incorporate the feedback into their subsequent performance. Further, it is uncertain what effect established goals, and particularly, potential conflicting goals may have on how trainees interpret and utilize feedback that they are given. In addition, it is critical to determine what the nature of feedback should be. At one end of the spectrum, one can focus on providing performance outcome feedback, such as informing team members of both individual and team level error. At the other end of this continuum, one can focus on providing feedback on the team processes that enabled the team to attain a given level of performance. Given the variety of team tasks, environments, structures, and stages of development of teams that exist, research is needed to determine how best to combine and to link outcome-type feedback and process feedback to minimize potential confusions, conflicts, and/or trade-offs in accomplishing individual level and team level performance and to optimize overall team effectiveness. The present symposium attempts to begin to address these questions, by providing both theoretical perspectives, as well as empirical findings.
    Feedback and Team Training: Exploring the Issues BIBA 1195-1199
      Elizabeth L. Blickensderfer; Janis A. Cannon-Bowers; Eduardo Salas
    As team researchers have endeavored to understand team performance and team training, feedback in the team environment has been a neglected topic. A number of issues are involved in the design and provision of feedback to teams. These include team process/outcome issues in addition to characteristics of the task, team as a whole, and the team members as individuals. The inherent problems in team feedback provide the impetus for considering new approaches to team feedback. One such approach, team self-correction, may be valuable in clarifying anticipations and explanations among team members.
    Toward Optimizing the Impact of Developmental Feedback in Team Training Simulations BIBA 1200-1203
      Kimberly A. Smith
    This paper reports evaluation data from an effort to train assertive communication in aircrews. The impact of preliminary conceptual training and simulator instructions which focused trainees on the target skill was examined on the reported usefulness of post-simulation feedback. Results indicated that trainees who received conceptual training one week prior to a simulation exercise reported the feedback they received on their performance more useful than those who had not received preliminary training. Further, trainees who were informed ahead of time which skill dimension was the focus of the simulation ruled the feedback they received on their performance less useful than those who were not given this information. Implications for the use of simulation to enhance teamwork skills and attitudes are discussed.
    Trade-Offs in a Team Tracking Task as a Function of Performance Feedback BIBA 1204-1208
      Florian G. Jentsch; Guillermo Navarro; Clint A. Bowers
    Team members often have to make decisions about which aspects of their tasks they should emphasize. One of the factors that may determine these decisions is the type of feedback. In this study, the influence of the type of concurrent performance feedback on team performance in a pursuit tracking task was investigated. Eighteen dyads performed a reciprocally interdependent team tracking task. Subjects' goal was to optimize team performance under three different conditions: One team member never received feedback while the other received either team, individual, or no feedback. The tracking error was measured. The results from this study largely confirmed the findings from previous research which had indicated a feedback by team member interaction: When provided with individual feedback, team members seemed to emphasize the perceived individual aspects of their task at the expense of the team effort. Under team feedback, the reverse occurred. In support of these findings, the current study found a significant gap in performance between the two team members under individual feedback conditions. The team member receiving individual feedback performed significantly better than their interdependent cohort. Yet, when subjects received either team or no feedback, their performance was worse than that of their team member, even if the difference failed to reach statistical significance. The results suggest that feedback can adequately focus subjects' attention towards specific aspects of their task.
    Team Skill Acquisition: Team and Individual Performance Effects of Feedback BIBA 1209-1213
      Jeanne L. Weaver; Julie M. Urban; Nalini Maniam; Clint A. Bowers
    Although team development research has, to some extent, addressed the various components of team functioning, leer studies have failed to sufficiently clarify what influences teams as they acquire the various skills that constitute complex team performance. Similarly, research regarding team structure, as of yet, has failed to fully determine how teams under various structures should be trained in order to optimize their ability to perform complex (i.e., both team and individual) tasks. Thus, there is a need to investigate differential developmental trends in teams of varying structure. Additionally, research is required to identify interventions which might optimize the developmental process. Although the effects of feedback are becoming increasingly well investigated, there has been relatively little investigation regarding the impact of feedback given on multiple task performance. The current study investigates the impact of feedback given, over time, regarding team vs. individual tasks in teams of two structures: non-hierarchical vs. product. Results supported the hypothesized differential effects of feedback type during skill acquisition under varying levels of structure.

    TRAINING: Computers in the Training of Complex Tasks [Symposium]

    Computers in the Training of Complex Tasks BIBA 1214
      Barry P. Goettl; Kevin B. Kline; J. Wesley Regian
    The increasing power and speed of desktop computers makes automated instruction more feasible than ever before. Today, PC-based automated instructional systems can be utilized for training very complex tasks from attention demanding motor skills tasks that require rapid processing of multiple sources of information to complicated procedural tasks that impose high demands on memory. With this new technology comes the need to examine the applicability of well established instructional methods in the domain of the new generation of automated instructional systems. One challenge that these new systems pose is that many of the relevant theories and pedagogies are based on research utilizing relatively simple tasks. The objective of this symposium is to examine basic research issues relevant to automated instruction and training of complex tasks.
    Massed versus Distributed Practice in Complex Skill Acquisition BIBA 1215-1219
      Wayne L. Shebilske; Kip Corrington; Jeffrey A. Jordan
    A training sequence on a complex video research task was distributed over 10 days or massed within two days. Measures of fatigue and confidence were taken. A final test battery given 1 week after acquisition consisted of retention tests, a test of resistance to interference, and a test of transfer. Trainees in the Distributed condition performed better throughout. Massed and Distributed trainees showed moderate levels of fatigue and did not differ from each other. Differences in confidence could not account for the results. Theories based on massing simple task acquisition within an hour are discussed as a framework for understanding and reducing suppression caused by massing complex tasks within days.
    Contextual Interference Effects on Acquisition and Transfer of a Complex Motor Task BIBA 1220-1224
      Barry P. Goettl
    Research in motor skill and verbal memory suggests that random sequencing of trials results in retention and transfer that is superior to blocked presentation of trials. The contextual interference effect is based largely on relatively simple motor and verbal tasks. The present study explores the generalizability of the contextual interference effect to a complex flight simulator task. Subjects (66 males and 45 females) were assigned to three groups (i.e., whole-task, part-task blocked, and part-task sequenced) and trained on a desktop flight simulator. Part-task blocked subjects practiced 13 component tasks presented in blocks (low contextual interference), and part-task sequenced subjects practiced the same component tasks presented in a sequence that was repeated several times (high contextual interference). It was predicted that part-task sequenced subjects would show superior retention and transfer compared to blocked subjects. Results indicated that whole-task subjects showed the best retention and the two part-task groups did not differ. Additionally, all three groups showed equivalent performance on the transfer task. These results suggest that the contextual interference effect may not generalize to complex tasks.
    The Type and Timing of Feedback within an Intelligent Console-Operations Tutor BIBA 1225-1228
      John D. Farquhar; J. Wesley Regian
    Feedback has remained a useful construct through a shift from a behaviorist explanation of learning to a more cognitive understanding. Research in the use of feedback in education suggests that corrective feedback, or feedback that provides the correct answer, is more effective than feedback that simply indicates an error. However, contrary to an information-processing theory of learning, these studies generally find no efficacy for feedback of a more elaborative nature such as the use of additional explanatory information. The study described in this paper investigated the type and timing of feedback within an intelligent console-operations tutor. Results indicate that when immediate feedback is employed during the acquisition of console-operation skill, elaborative feedback yields greater accuracy of the skill over the use of corrective feedback.
    The Effects of Training on Cognitive Capacity Demands for Synthetic Speech BIBA 1229-1233
      Catherine Connolly Gomez; Wayle Shebilske; J. Wesley Regian
    Previous studies have revealed that the perception and comprehension of synthetic speech may be attributed to increased processing demands in short-term memory as reflected in serial-order and preload paradigm tasks. Additionally, it has been consistently shown that the perception of synthetic speech improves with moderate amounts of training. The present study was conducted to determine if the increased perceptual effects of training for synthetic speech can be attributed to a reduction of short-term memory load. Two groups of subjects were tested with synthetic speech using the same comprehension and high cognitive processing tasks before and after training. One group was trained with synthetic speech and the other group acted as the control, receiving no training between the pretest and post-test interval. Results reveal similar increases in comprehension based on previous synthetic speech studies for the trained group. Moreover, these results suggest that training on synthetic speech promotes better allocation of attentional resources which result in improved performance on working memory capacity measures.

    TRAINING: Data-Based Training Studies [Lecture]

    The Effects of Workload and Uncertainty on Team Development BIBA 1234-1237
      Julie M. Urban; Clint A. Bowers; Ben B., Jr. Morgan; Nalini Maniam
    Two studies were performed that attempted to test and extend the team development theory of "punctuated equilibrium" proposed by Gersick (1988). In the first study, twelve five-person teams performed a resource allocation task, either under low or high workload. In the second study, twelve three-person teams performed an adapted version of this resource allocation task under either certain or uncertain task conditions. Various aspects of performance were assessed. The results of these studies support Gersick's contention that teams do go through one marked period of transition. However, this transition does not necessarily occur in the midpoint of the team's life cycle.
    Using the Backward Transfer Paradigm to Validate the AH-64 Simulator Training Research Advanced Testbed for Aviation BIBA 1238-1241
      John E., II Stewart
    The Simulator Training Research Advanced Testbed for Aviation (STRATA) is a modular research simulator which in its current configuration represents the AH-64 helicopter. The backward transfer of training paradigm was employed to determine if AH-64 piloting skills transfer to STRATA. Ten AH-64 pilots participated in the experiment. They performed a mission scenario consisting of 13 Aircrew Training Manual (ATM) tasks. No orientation or practice was allowed. Most participants rated STRATA as highly similar to the AH-64 in handling. Real-time performance ratings indicated that of 130 task events, 88.5% were performed to ATM standards. After the experiment, four independent judges rank-ordered performance on the hover task using output from STRATA's performance measurement system. Rankings showed high concordance and a high correlation with real-time ratings.
    Evaluating a Maintenance Crew Resource Management Training Program: Effects on Attitudes, Behaviors, and Performance BIBA 1242-1246
      Michelle M. Robertson; James C. Taylor; John W. Stelly; Robert Wagner
    A Crew Resource Management program for maintenance personnel has been developed by an airline company which involves several team relied concepts. Technical operations managers' pro and post-training attitudes and their follow-up attitudes (2, 6 and 12 months afterwards) concerning a variety of management and organizational factors were compared with one another as well as with maintenance performance measures. Comparisons of participants' attitudes before and after their training showed a significant improvement in attitudes indicators and these attitudes remained stable overtime. Positive trends for two of the maintenance performance indicators are demonstrated in comparing pro and post training performance measures. Increased safety and improved on-time performance was found relied to improved attitudes about participation and assertive communication. Open ended responses as well as anecdotal evidence confirms the positive changes in attitudes and behaviors.
    Development of a Real-Time Simulation with Intelligent Tutoring Capabilities BIBA 1247-1251
      Sallie E. Gordon; Bettina A. Babbitt; Herbert H. Bell; H. Barbara Sorensen
    Training programs for complex tasks are increasingly using simulations to provide transfer of training to the job environment without incurring high costs of on-the-job training. A second trend in training is toward the use of intelligent tutoring systems (ITSs) to provide individualized feedback to optimize training. Combining simulation with an ITS can be especially beneficial, but use of intelligent tutoring mechanisms such as expert systems is often difficult in a complex, real-time environment. In this paper, we describe the development of a proof-of-concept training program that combines F-16 flight simulation with an embedded real-time intelligent tutoring system. In the simulation, pilots learn the correct use of advanced fire control radar modes to locate and assess multiple enemy formations (search and sort tasks). The expert system monitors pilot behavior and verbal responses as the pilot flies the simulation. At certain critical points, if the pilot's performance has fallen outside of pre-specified parameters of "safe" behavior, the tutoring component stops the simulation and feedback is provided.

    VISUAL PERFORMANCE: Visual Search Performance [Lecture]

    The Effects of Back Angle on Target Detection BIBA 1252-1255
      R. J. Edwards; D. F. Streets; G. Bond
    It is suggested that crew members of future armoured vehicles will be reclined. However, information relating to task performance when reclined for prolonged (i.e. >30 minutes) periods is limited. This study has investigated the effects of prolonged reclining on a simple target detection task.
       Twelve subjects undertook four separate seventy minute studies in the upright (control) seated posture or reclined, with head support, at 25°, 45°, and 65° from the vertical. Subjects viewed a model scrubland scene, with light emitting diodes hidden behind the scrub, through seven vision blocks which covered a 300° field of view. During the first 10 minutes subjects responded to a sequence of 32 randomly distributed signals of 4 seconds length, with 15 seconds between each signal. They rested for 50 minutes before repeating the 10 minute test epoch. Measurements taken were number of responses, body part discomfort, sleepiness and Stress and Arousal Checklist, and were analyzed using Analysis of Variance.
       Percentage targets detected significantly reduced with back angle, with the greatest reduction being in the subjects' frontal arc. General body and neck discomfort and sleepiness significantly increased with back angle. Reported stress significantly increased and arousal decreased, but only between 0° and 65°. There was no significant time effect.
       Possible explanations for the results are that the chair and posture inhibited freedom and ease of movement, or that increased discomfort at the more acute angles may have resulted in an altered searching behaviour. Whatever the mechanism this study has shown that reclined postures may significantly impair the performance of an all-round surveillance task, particularly at the more acute back angles.
    Effects of Symbol Type, Selection Tool, and Information Density on Tactical Display Visual Search Performance BIBA 1256-1260
      William A. Nugent
    This study compared the relative effectiveness of two color-coded symbol sets and two selection tools in performing a tactical display visual search task. Performance data were obtained for 36 symbols (called target tracks), 12 in each of three warfare areas. Each target track was presented under four levels of overlap by adjoining or occluding it with non-target (distracter) symbols of the same or different color. Performance measures, which included speed, accuracy, number of selection tool uses and time-outs, were obtained for 144 trials per participant. Results showed color-filled NATO symbols yielded faster, more accurate performance than stroke-drawn NTDS symbols for all but the total overlap condition. Significant two-way interactions were also obtained between the symbol set and selection tool factors, with results showing poorer overall performance for participants in the NTDS-click tool condition. The practical applications and design implications of these findings are discussed, along with specific recommendations for improving operator performance when using color-filled NATO symbols in dense tactical track environments.
    Visual Lobe and Visual Search Performance BIBA 1261-1265
      Anand K. Gramopadhye; Rakesh Sreenivasan
    Visual lobe continues to be an important determinant of extended search performance. This study looks at the effect of various methods of training on visual lobe size. Furthermore, it relates changes in visual lobe size to improvements in visual search performance. Subjects were trained using three different methods. In method 1 subjects were trained on a visual lobe improvement task using the actual fault. In method 2 subjects were trained on an irrelevant target using the visual lobe measurement task and in method 3 subjects were provided practice on the extended visual search task. Greatest improvements in lobe size and search performance were observed for method 1.
    The Influence of Stimulus Dimensions and Training on Visual Search Performance BIBA 1266-1270
      Mary P. Czerwinski; Evan M. Feldman; Edward Cutrell
    Traditional studies of attention, training and visual search have focused on the use of separable dimensions (usually alphanumeric stimuli), and equating the number of items in consistent versus varied mapping training paradigms. However, the design of visual displays requires a heavy reliance upon configural and integral dimensions (stimuli that group). This set of studies examines the effects of configural dimensions (also using alphanumeric stimuli), as well as equating the number of training trials on specific targets between consistent versus varied mapping conditions. Predictions from extant theories of attention and visual search will be discussed where relevant. Results show that both factors have a large influence on the effects of training in visual search tasks. The influence of these variables needs to be incorporated into current theories of attention and visual search, especially as they are applied to the design of graphical user interfaces and visual displays.

    VISUAL PERFORMANCE: Matching Visual Displays to Information-Processing Capabilities [Lecture]

    Effects of Perceptual Augmentation of Visual Displays: Dissociation of Performance and Situational Awareness BIBA 1271-1274
      R. Jay Shively; Allen D. Goodman
    It is intuitive that good performance is associated with, if not caused by, good situational awareness. There are, however, some situations in which these two concepts diverge. Some examples of this dissociation, such as auto-pilots, have been identified. However, it is also possible that these concepts diverge in a much more subtle manner. This research is focused on investigating those more subtle situations. Specifically, this research addresses the effects of perceptual display enhancement based upon Ecological Task Analysis (ETA) on performance and situational awareness. A perceptually augmented display was designed based upon ETA. Globally, performance advantages were found for the group with the enhanced display. Further, the findings demonstrated a dissociation of subtask performance and operator's knowledge of the system subtask. The mechanisms involved in this dissociation are related to the characteristics of the display augmentation that led to the increased performance. The level of processing, and the presence of feedback seem to play an important mediating role. These findings have important implications for both designers and researchers.
    The Basis for the Perception and Control of Altitude: Splay & Depression Angle Components of Optical Flow BIBA 1275-1279
      Sheila A. Garness; John M. Flach; Terry Stanard; Rik Warren
    This study evaluated subjects ability to track a constant altitude as a function of the structure in the optical flow field. Optic flow was manipulated by using four different types of ground texture (splay angle, depression angle, random dot, and block textures) crossed with two global optical flow (GOF) rates (0 and 3 eyeheights/s). The subjects were asked to maintain a constant altitude while wind disturbances randomly perturbed them on vertical, lateral, and fore-aft axes. The critical independent variables were texture type and GOF rate. Texture type was a within-subjects variable while GOF rate was a between-subjects variable. The main dependent variables included RMS height error and the correlation between subjects stick activity and the three wind disturbances. For both dependent variables, an interaction was found in that the depression angle texture provided superior performance in a hover or 0 GOF rate condition. The splay angle texture provided a constant level of performance for both GOF rates, being superior to depression angle in the higher GOF rate. These results are consistent with Flach et al.'s (1992) hypothesis that the ability to pick-up altitude information from the optic flow field depends upon the amount of optical activity that is specific to changes in altitude (signal) rather than specific to changes in lateral or fore-aft position (noise). This hypothesis provides a higher order explanation for previous results on the control of altitude which had been thought to be inconsistent.
    Operator Performance in Pattern Matching as a Function of Reference Material Structure BIBA 1280-1284
      Marion P. Kibbe; Jan Stiff
    This paper examines the performance of subjects in verifying matches between aerial photographs and overlayed line drawings which vary in structure and pixel count. Both speed and accuracy improve as the pixel count increases with an asymptote at about 3500 pixels. Speed and accuracy also improve when the line drawings are made up of long lines or with lines which correspond to the cognitive content of the photographs.
    Operator Reliance on Automated Support for Target Recognition BIBA 1285-1289
      Jean MacMillan; Eileen B. Entin; Daniel Serfaty
    In machine-aided target recognition, human operators work with an automatic target recognition (ATR) system to locate targets in cluttered and degraded imagery. The operator must integrate his or her own visual judgment concerning whether a target is present in the image with the ATR's judgment, which is typically expressed numerically. We conducted a series of experiments in which subjects attempted to locate target shapes among non-targets based only on visual images and based on both visual images and supplementary numeric information such as an ATR might provide. Image quality was controlled as an independent variable through the use of distortion rates that randomly altered pixel values to degrade the image. We found that subjects maintained a constant false alarm rate as image distortion increased, at the expense of a lower hit rate. This result was found consistently in experiments where the subjects' task was to distinguish single targets from a blank background, to distinguish single targets from single non-targets, and to locate multiple targets in a multiple-object display. We also found a bias toward over reliance on image versus numeric information. As image distortion increased, subjects failed to make optimal use of supplementary numeric information and showed an unnecessary decrease in performance. The results suggest that operators may experience difficulty in working with an ATR that has a high false alarm rate, even if the ATR's hit rate is also high, and that numeric expressions of ATR judgment may be undervalued by operators in locating targets.

    VISUAL PERFORMANCE: Stereoscopic Display [Lecture]

    A Human Factors Simulation Tool for Stereoscopic Displays BIBA 1290-1294
      William F. Reinhart; Charles J. C. Lloyd
    The increasing complexity of advanced display systems places increased importance on simulation tools. Display simulation tools allow the development of displays with the desired balance of image quality and end cost while reducing the time and cost otherwise associated with iterative testing of physical prototypes. In addition, simulation offers the advantage of permitting demonstration of display systems to targeted users and decision makers early in the display design cycle, allowing for more feedback in the design process. A new display simulation tool is described in this paper with the capability of simulating advanced electronic stereoscopic displays. To illustrate the use of this tool, an evaluation of a simulated stereoscopic head-mounted display is reported. Surgeons viewed stereoscopic laparoscopic imagery and rated the acceptability of display gray scale, resolution, and field of view. Clear preferences were seen for levels of each of these parameters. Simulations such as the one described in this report play an invaluable role in defining acceptable design parameters prior to commitment to display production.
    Visualization of CAD Objects Using a Stereoscopic Display BIBA 1295-1299
      Michael E. Brown; Jennie J. Gallimore
    Subjects memorized the shape of a static 3-D object displayed on a stereoscopic CRT. In each of a series of trials that followed, single static objects were presented. The angular orientation of each trial object was one of six 36-degree increments relative to the angle of the memorized stimulus. The subject's task was to determine, as quickly and accurately as possible, whether the trial object was the same shape as the memorized object or its mirrored image. One of the two cases was always true. Disparity and interposition were manipulated in a within-subject manner during the initial memorization period and the trials that followed. Subject response time and error rate were evaluated. The experimental objective was to determine the extent to which stereopsis and hidden surface affect subjects' ability to 1) transfer to and retrieve from long-term memory spatial information about a 3-D object, and 2) visualize spatial characteristics in a quick and direct manner. Improved performance due to hidden surface is the most convincing experimental finding. The study also found a significant but limited stereopsis effect.
    Distance Perception of Stereoscopically Presented Virtual Objects Optically Superimposed on Physical Objects by a Head-Mounted See-Through Display BIBA 1300-1304
      Stephen R. Ellis; Urs J. Bucher
    The influence of physically presented background stimuli on distance judgments to optically overlaid, stereoscopic virtual images has been studied using head-mounted stereoscopic, virtual image displays. Positioning of an opaque physical object either at the perceived depth of the virtual image or at a position substantially in front of it, has been observed to cause the virtual image to apparently move doses to the observer. In the case of physical objects positioned substantially in front of the virtual image, subjects often perceive the opaque object as transparent. Evidence is presented that the apparent change of position caused by interposition of the physical object is not influenced by the strengthening of occlusion cues but is influenced by motion of the physical objects which would attract the subjects ocular vergence. The observed effect appears to be associated with the relative conspicuousness of the overlaid virtual image and the background. This effect may be related to Foley's models of open-loop stereoscopic pointing errors which attributed the stereoscopic distance errors to misjudgment of a reference point for interpretation of retinal disparities. Some implications for the design of see-through displays for manufacturing will also be discussed briefly.
    Relevant Cues for the Visual Perception of Depth: Is Where You See It Where It Is? BIBA 1305-1309
      R. Troy Surdick; Elizabeth T. Davis; Robert A. King; Gregory M. Corso; Alexander Shapiro; Larry Hodges; Kelly Elliot
    We tested seven visual depth cues (relative brightness, relative size, relative height, linear perspective, foreshortening, texture gradient, and stereopsis) at viewing distances of one and two meters to answer two questions. First, which cues provide effective depth information (i.e., only a small change in the depth cue results in a noticeable change in perceived depth). Second, how does the effectiveness of these depth cues change as a function of the viewing distance? Six college-aged subjects were tested with each depth cue at both viewing distances. They were tested using a method of constant stimuli procedure and a modified Wheatstone stereoscopic display. Accuracies for perceptual match settings for all cues were very high (mean constant errors were near zero), and no cues were significantly more or less accurate than any others. Effectiveness of the perspective cues (linear perspective, foreshortening, and texture gradient) was superior to that of other depth cues, while effectiveness of relative brightness was vastly inferior. Moreover, stereopsis, among the more effective cues at one meter, was significantly less so at two meters. These results have theoretical implications for models of human spatial perception and practical implications for the design and development of 3D virtual environments.

    VISUAL PERFORMANCE: Panel

    Integrating Human and Machine Vision: Lessons from Automated Target Recognition Systems BIBA 1310-1311
      Jean MacMillan; Curtis Becker; Marion Kibbe; Barbara O'Kane
    The development of Automated Target Recognition (ATR) systems has been the focus of considerable interest and funding during the past decade. Such systems were originally envisioned as being almost completely autonomous and capable of detecting, locating, and classifying targets and of assigning weapons to targets with little or no human intervention. Such completely autonomous performance remains well beyond current ATR capabilities, however. Under the performance levels currently achievable for ATR systems, the human operator plays an essential role in screening ATR detections and rejecting false alarms. Operators must rapidly review the ATR's judgments, and their accuracy in confirming or rejecting those judgments is a critical determinant of the overall effectiveness of the human-machine system. The integration of human and machine visual capabilities is a key factor in effective system design.
       Research on human-ATR interaction has identified fundamental issues that must be considered in the design of any system in which human visual judgment is integrated with machine-based visual judgment. The objective of this panel is to identify and discuss critical issues for the design of an aided target recognition system, in which the human operator and the ATR work synergistically, and to assess the implications of human-ATR research results for the design of systems that integral human and machine vision.
       The panel will discuss the following issues:
  • Differences in human and ATR target-recognition accuracy. How accurate must
       an ATR be, relative to the human operator, in order to be of assistance?
       What are the implications of ATR false alarms versus missed detections? How
       do ATR performance levels relate to the confidence that human operators
       place in the ATR?
  • Differences in the process by which humans and ATR systems recognize targets,
       and the implications of those differences. Automated target recognition
       technologies need not replicate the process by which humans recognize
       targets, and human and machine vision may show different relative strengths
       and weaknesses, i.e., the ATR may be superior to the human in recognizing
       some types of targets in some circumstances but not in others. What are the
       implications of these differences for operator confidence in the ATR and for
       system design?
  • How important is the design of the visual interface in human-ATR
       interactions? What aspects of display design can affect human-ATR system
       performance? What are the implications of combining human image
       interpretation with ATR judgments that are displayed numerically, e.g., a
       numeric level of confidence associated with a target identification?
  • VISUAL PERFORMANCE: Vigilance, Memory, and Mental Workload [Lecture]

    Vigilance: How to Do It and Who Should Do It BIBA 1312-1316
      David A. Sawin; Mark W. Scerbo
    The present laboratory study examined variables that are believed to impact vigilance in operational settings. First, instructions given to subjects were manipulated in order to impact their feelings of "command pressure" (Mackie, Wylie, & Smith, 1985). It was expected that instructions emphasizing detection efficiency would lead to increased subjective workload. Further, it was expected that decreasing instructional demands by emphasizing relaxing qualities of the task would not impact performance, though lower workload ratings were expected. In addition, response bias was manipulated by telling subjects to adopt either a conservative or a liberal criterion. Secondly, subjects' boredom proneness (BP) was measured to test the predictive validity of this measure in vigilance. All subjects monitored a VDT of uniform color for 30 min. They were asked to respond to 110 ms "long flickers" embedded in a matrix of 50 ms "short flickers" occurring at a rate of 15 per min. As expected, detection-emphasis subjects reported significantly higher workload. Overall performance regarding proportion of hits and A' was not impacted significantly by command emphasis, though an Emphasis x Bias interaction revealed that relaxation-liberal subjects performed more poorly than subjects in the remaining conditions. Finally, BP scores significantly predicted overall hits and A'. Results indicate that (1) much of what individuals find unpleasant about vigilance may lie with the demands implied by the initial instructions and (2) valid personality predictors of vigilance performance exist.
    Dynamic Memory: Keeping Track of Continually Changing Information BIBA 1317-1321
      Michael Venturino; Nathan J. Romano; Sheryl L. Miller; Megan Murphy; Tara M. Coffey
    The requirement to remember continuously changing information places substantial demands on the human operator's working memory system. Previous research (Yntema & Mueser, 1960) found that in keeping track of dynamically changing information, humans' memory for changing information was better when they kept track of many different attributes of a single object than when they kept track of the identical attribute of many different objects. Due to a confound in the Yntema and Mueser experiment, the unique and combined effects of information organization and similarity-based interference cannot be determined, limiting the information about dynamic memory. This experiment represents an attempt to overcome this limitation by assessing the roles of organization and similarity-based interference in dynamic memory. The experimental task was a keeping track task in which a series of changing attribute values were presented sequentially, and subjects were required to remember the most recent update for each attribute. Three factors were manipulated in the experiment: number of "objects" (one vs many objects), type of attribute (same vs different), and memory load (2, 4, or 6 attributes to remember). Results showed that as memory load increased, keeping track performance in the many-object condition decreased to a greater extent than in the one-object condition. Also, as memory load increased, accuracy decreased at a greater rate for the same-attribute condition than for the different-attribute condition. The effect size for attribute similarity was much larger than that for number of objects. It was concluded that similarity-based interference is quite destructive to dynamic memory. It appears that the cost of attribute similarity far outweighs the benefits of organizing the continually-changing attributes. Such results have implications for structuring tasks and aiding memory in situations where operators must remember information in dynamically changing environments.
    EEG and Subjective Measures of Private Pilot Workload BIBA 1322-1325
      Glenn F. Wilson; Thomas Hankins
    Complex systems can place high levels of mental demand on human operators and methods of assessing these demands are needed. Subjective and performance metrics are typically employed while psychophysiological assessment has been used to a more limited extent. In this study, civilian pilots flew a single engine propeller aircraft on a flight profile designed to produce several levels of cognitive workload using VFR and IFR conditions. Subjective and brain wave (EEG) measures were used to assess mental workload. EEG theta band activity was sensitive to a wider range of workload levels and was more sensitive than the alpha and beta bands or the subjective reports. The alpha and beta bands reliably discriminated between ground and night segments as did the subjective data.
    Physiological and Subjective Workload Changes during a Simulated Air Traffic Control Task BIB --
      Jeffrey B. Brookins; Glenn R. Wilson

    VISUAL PERFORMANCE: Interpreting Graphic Displays [Lecture]

    Evaluation of Alternative Methods of Representing Three-Dimensional Objects on Computer Displays BIBA 1326-1330
      Eric N. Wiebe
    Due to the increased use of 3D modeling software in the design and manufacture of products, careful evaluation needs to be made as to how the 3D model is represented on the computer display. The experiment's hypothesis is that both rate in which projections of a rotating object are presented and whether the object is rendered as a line drawing or shaded will effect the mental representation of the object. The experiment factorially crossed three levels of projection presentation rate with two levels of rendering (line drawing vs. shaded). All levels of both independent variables were between subjects. The subjects' score on a mental rotations test score was used as a covariant. The subjects each viewed 40 displays representing different rotating objects and identified the objects through a forced choice pair selection. RT and error rate were measured for each selection trial. Data on a total of 72 subjects was analyzed using the ANOVA procedure. The results of the experiment showed a significant main effect of the rate of presentation variable on RT. The results also showed a significant main effect of the rendering variable on error rate. No interaction was found between the two independent variables. The results indicate varying presentation rate can be an effective tool in allowing faster interpretations of an object. It is also recommended that the display technique be carefully matched to the complexity of the object being displayed and the capabilities of the computer being used to display it.
    A Comparison of Sequential and Spatial Displays in a Complex Monitoring Task BIBA 1331-1335
      Christopher M. Konrad; Arthur F. Kramer; Stephen E. Watson
    A sequential or RAP COM display was compared to a more conventional spatial display as subjects monitored dynamically changing sets of numbers and responded to occasional target stimuli. In an effort to equate the stimulus-response compatibility of the two displays subjects responded to the targets with a chord keyboard. We examined the influence of display duration on the performance with the RAP COM and spatial formats by presenting the stimuli at three different durations, 400, 800 and 1200 msec. The influence of practice on performance with the RAP COM and spatial displays was also investigated. Targets were responded to more quickly in the RAP COM than in the spatial displays at each of the three presentation durations and across over 2000 trials of practice. Accuracy was influenced by the display presentation duration. Accuracy was higher for the RAP COM than the spatial display at the 800 msec stimulus presentation duration. Accuracy was statistically equivalent for the RAP COM and spatial display format at the 400 and 1200 msec display durations, although there was a trend for lower accuracy for the RAP COM display at the faster presentation duration. Interestingly, the lower accuracy for the RAP COM display format at the 400 msec presentation duration appears to be due to illusory conjunctions. Our results will be discussed in terms of the utility of sequential or RAP COM displays in complex, real-world settings.
    The Effects of Display Layout on Monitoring and Updating System States BIBA 1336-1340
      Stephen M. Hess; Mark C. Detweiler; R. Darin Ellis
    Information-display issues are described in the context of a complex memory-updating task similar to Yntema's (1963). Subjects performed the updating task with two types of graphic interfaces. In one condition, the interface provided invariant spatial information which could be used to support task performance; in the second condition, the same information was presented in a single spatial location. The results suggest that 1) the mapping effect (one-to-many vs. many-to-one) found by Yntema did not occur when the response sets being updated were composed of the same response alternatives; 2) the type of screen layout used to support the updating task had a large impact on both accuracy and time to respond; and 3) the type of screen layout used interacted with the task's cognitive demands, including length of lag between an update and a response, and the number of variables being monitored. These results suggest that the design of information displays requires consideration of the abstract representational requirements of the task being supported.
    Animated Mimic Displays BIBA 1341
      Kevin B. Bennett
    Animated mimic displays represent the physical structure of a system, including: 1) the important components, systems, or subsystems, 2) the physical/causal connections between components, and 3) the flow of information or resources through these connections. This type of display has the potential to improve both the effectiveness of real-time performance and the efficiency of training. A research program has been initiated to determine critical factors in the design of animated mimic displays. The initial display designs were based on examples of animated mimic displays found in the literature. Four sets of psychophysical experiments investigated 1) chromatic/luminance contrast, 2) spatial/temporal frequency, 3) contours/borders, and 4) stairstep/sinusoidal waveforms. The first two sets of experiments revealed that existing designs were not effective. A theoretical explanation was developed and is discussed briefly. The second two sets of experiments investigated alternative display designs based, in part, upon these theoretical insights. The results indicate that the alternative designs will improve the effectiveness of animated mimic displays. A set of design guidelines are provided.