HCI Bibliography Home | HCI Conferences | HCII Archive | Detailed Records | RefWorks | EndNote | Hide Abstracts
HCII Tables of Contents: 89-1a89-1b89-2a89-2b91-1a91-1b91-2a91-2b93-1a93-1b93-1c

Proceedings of the Fifth International Conference on Human-Computer Interaction

Fullname:Proceedings of the Fifth International Conference on Human-Computer Interaction jointly with the Ninth Symposium on Human Interface (Japan)
Editors:Michael J. Smith; Gavriel Salvendy
Location:Orlando, Florida
Dates:1993-Aug-08 to 1993-Aug-13
Publisher:Elsevier Science
Standard No:ISBN 0-444-89540-X ISSN 0921-2647; hcibib: HCII93
Papers:362
Pages:1042+1125
Links:www.elsevier.com
  1. HCII 1993-08-08 Volume 1
    1. III. Case Studies
    2. IV. User Issues
    3. V. Methodologies

HCII 1993-08-08 Volume 1

III. Case Studies

Human-Computer Interaction within the Criminal Justice System: A Study of the Probation Service BIBA 344-349
  P. J. Baugh; A. C. Gillies
In the following paper, the authors examine the introduction of office automation into the criminal justice system in the UK -- an area of administration which does not lend itself easily to the acceptance of new technologies. In order to facilitate the examination and, owing to the complexity of the justice system, the authors examined one aspect -- the Probation Service. This service is responsible for the supervision of offenders in the community as an alternative to imprisonment. Bearing in mind the limited use of technology in the Service, it was felt that the problem area should be analyzed using a three-layer model of Human-Computer Interaction. Within this framework, the key issue to be considered was the assessment of the net benefits to be gained through the introduction of office automation into the Probation Service over a period of time. To this end, the authors examined the problem areas using a time-phase model. In order to illustrate the issues involved the authors drew on their experience of advising a Probation Service on the problems of developing and extending its office document production facilities.
A Study of Human-Computer Interaction in the Selection of Commercial Pilots for Automated Flight Decks BIBA 350-355
  Ronald E. Clark; William F., III Herlehy
An extensive current literature review was conducted, as well as interviews and surveys with a small sample of aviation human factors scientists, aviation design engineers, and pilot selection specialists. Generally, the literature and interview/survey results indicated that the components of the "Right Stuff" necessary to pilot commercial flight decks had shifted. In addition, most agreed that there has been a shift in aircraft piloting from "human" to "human-computer interaction", requiring a new perspective, and increased study. It was recommended that appropriate aviation agencies take a leading role in determining the new selection criteria for future automated/supersonic/hypersonic flight crew selection.
The Effects of New Information Technology on Interorganisational Relations: Electronic Tax Return Lodgement in Australia BIBA 356-361
  Trevor A. Williams
This paper examines the role of information technology in supporting the efforts of a government agency to strengthen voluntary cooperation in regulatory relationships.
The Relationship Between Secretaries' Opinions of Computing and Their Computer Performance within a United Kingdom Health-Care Setting: Implications for In-House IT Support and Training BIBA 362-367
  J. Arthur; P. Brooks
The opinions of secretarial staff to computer technology and their reported level of anxiety and perceived likelihood of experiencing anxiety-causing events are investigated. Significant agreement among participants (n = 100) was found for both rated anxiety and the likelihood of the computing events. Six key opinions were identified through principal components analysis and were found to significantly predict procedural and conceptual performance with computers through multiple regression analysis. A major conclusion is the dominance of anxiety as a key opinion area and influence on performance. Recommendations for in-house support are provided and include a programme to increase confidence among the user group and promote a transfer of training from conventional secretarial skills.
Metalearning and Metaknowledge Strategies to Produce Educational Software BIBA 368-373
  J. H. Sanchez
The growing literature within the past years has emphasized the major role played by cognition in mediating the complex process of human-computer interaction. This trend is also seen in the field of educational computing where some cognitive strategies have been used in order to improve the quality of educational software. This study examines the use of concept maps as metalearning and metaknowledge tools for educational software development. It also describes a new methodology used to apply constructivist learning ideas through concept maps to design and develop educational software. As a result of the application of qualitative research methods, we propose the use of current cognitive learning theory as theoretical framework for educational software production and suggest the utilization of concept maps as powerful metalearning and metaknowledge tools to interface adequately the human-computer interaction for learning purposes.
Hypermedia for the In-House Development of Information Systems BIBA 374-379
  N. I. Beagley; R. A. Haslam; K. C. Parsons
As a consequence of the availability of high level, software development environments, self tailored software can now be developed by small organisations for whom this form of tool was previously unavailable. As the ability to produce effective software without low level programming language experience is recognised widely, the number and diversity of software applications is likely to greatly increase. By taking an iterative prototyping approach to development an in-house developer can tailor products to meet the actual requirements of the real user group.
Toward a Cognitive Ergonomics Evaluation System of Typing Chinese Characters into Computers BIBA 380-385
  Zhang Kan; Guo Sumei; Zhao Huiling
In order to set up a testing system to evaluate different approaches of typing Chinese characters into computers, a model of typing Chinese characters into computers was proposed and a set of experiments were carried out to verify the model and to determine indices and functions of information processing of typing Chinese characters into computers with either phonetic rules or graphic rules. This paper reports some of the results found by the experiments.
Inferring Pilot Intent and Error as a Basis for Electronic Crew Assistance BIBA 386-391
  T. Wittig; R. Onken
In this paper, the Pilot Intent and Error Recognition module as part of a knowledge based Cockpit Assistant System is presented. This Cockpit Assistant System mainly supports the pilot crew with regard to the monitoring and planning task and provides assistance for a number of plan execution functions for the civil flight operation under Instrument Flight Rules.
   A brief survey of the concept and the function of the main system is given. Thereafter, the structure of the Pilot Intent and Error Recognition is described in more detail. Finally, the integration of the module within the main system and some results of first low scale experiments in a flight simulator are presented.
A Case Study of Plan Recognition -- Command Sequences as Acts BIBA 392-397
  Y. Inoue; M. Nagata
A person can normally act appropriately by comprehending their surrounding situation. Correspondingly, it is believed that computers can interact appropriately with users through the computer acquiring an understanding of the users' plans.
   In this study, a sequence of the OS commands is regarded as an act in "plan recognition". That is, a computer recognizes the user's plan by identifying the sequence of the OS command lines. Here an algorithm for recognizing the plan of the user from the sequences of the real MS-DOS command lines is proposed. Moreover, a system to transform the sequence of command lines into the UNIX command lines by using this recognition is implemented. The algorithm uses the hierarchy of a goal-subgoals in the plan and rules transforming a goal into its lower goals. An experiment was carried out employing real sequences of the command lines and confirmed that plan recognition is possible within a certain range of tasks.
Dual Design of Computer-Based Air Traffic Control Systems -- Examples from European Air Traffic BIBA 398-403
  B. Harendt
The Dual Design Approach will be used to describe the complex air traffic control system as a human-machine system. The Dual Design Approach is a set of principles to ensure appropriate development of both the technical and human aspects of human-machine systems. Both the technology-based design and the working-process-based design should be used in parallel to obtain an optimum. This is exemplified by two components of the air traffic control system. Examples of specific support systems -- especially from EUROCONTROL -- are shown. Based on these support systems for the radar controllers the capacity of air traffic control systems can be increased without additional risks in case of technical failures. But fully automated systems for air traffic control -- even if it was feasible -- will not be acceptable.
Integrating QFD into Software Development: A Case Study BIBA 404-409
  Jay Lundell; David Williams
In trying to develop software that is usable and that meets customers needs, two problems arise. First, although customer input may be collected at the beginning of the project, this information may be neglected as the design and implementation of the software gets underway. Second, even when customer feedback survives throughout development, members of the project team may interpret the customer feedback in different ways due to different internal objectives, leading to less than optimal designs. This paper describes a case study in the use of QFD (Quality Function Deployment) as a method to promulgate accurate customer feedback throughout the entire design cycle of a software product.
Dedicated CAD for Apparel Design BIBA 410-415
  J. McCartney; B. K. Hinds; J. J. Zhang; W. Hamilton
The work described here represents an attempt to devise CAD techniques appropriate to the apparel industry. An integrated approach is outlined which provides a greater justification for CAD, other than the ability to produce a single image of a design concept. In addition, high level tools are described which enable 3D design specifications to be evolved without exhaustive dimensional input.
Assessing Ergonomic Furniture in The Workplace BIBA 416-420
  George J. Boggs; Robert E. Warren; Daniel R. Lowther
The selection of work center furniture is critical to the continued productivity of telecommunications workers in VDT-intensive environments. This paper describes the design three separate ergonomic studies, conducted over a period of six months, whose goal was to select ergonomic furniture in a working environment. The first study assessed user comfort and user perceptions of furniture features for each of eight adjustable worktables. The second study applied this same methodology to each of ten chairs. The third study consisted of ergonomically-oriented inspections and checklist ratings in the human factors laboratory for the eight adjustable worktables and the ten chairs. The results suggest that the three factors (user body discomfort, user perceptions of furniture features and ergonomic evaluation) are tapping different aspects of furniture usability. Each measure contributes to an informed selection decision, but should not be used in isolation. A careful consideration of the total configuration of measures may lead to better furniture selection.
Task-Oriented User Documentation Using the User Action Notation: A Case Study BIBA 421-426
  J. D. Chase; Marie Paretti; H. Rex Hartson; Deborah Hix
Good documentation is a critical component of usable systems. While documentation commonly focuses on system features and functions, it is often more effective when based on a user- and task-oriented view. It must also keep pace with rapid changes in the design of a system during an iterative development process. The User Action Notation (UAN), a user- and task-oriented notation that describes the behavior of the user and the interface during their cooperative performance of a task, helps support these requirements. We present a case study of an industrial software development project that used the UAN as part of the documentation process, and show how the UAN supported translation from an interface design to user documentation.
Do Computers Solve the Problem? Telephone vs Telefax vs Computers in Home Shopping Systems in Scandinavia BIBA 427-432
  M. A. Karlsson; M. Kaulio
On basis of interviews and observations, user requirements for home shopping systems (HSS) were formulated. The results demonstrate that HSS have to be regarded as a service and as a system. User satisfaction is mainly related to aspects concerning the systems' service components while problems re communication media were downgraded. The system input /the order/ is, however, critical. Choice and design of media is consequently vital. Telefax is recommended in favour of the computer (videotex), being easier to handle and accept by users.
Observations on Game Playing BIBA 433-437
  C. N. Quinn; M. Boeson; D. Kedzier; D. Kelmenson; R. Moser; S. Rice
Computer games can provide insight into interface design. One particular category of game, the adventure game, has elements that highlight particular aspects of user-system interaction, particularly for spatial systems. We discuss the reasons why we are developing such games and some empirical observations on navigation and action issues. These issues are then related to the use of space in interface representations.
From Risk to Disease: Capturing Knowledge by a Hypertext Approach BIBA 438-443
  G. Franco
This hypertext was created to provide users with a problem-oriented tool, dealing with the presentation of an occupational disease case report. The hypertext covers the case report containing the clustered information about the patient and the library including educational material. Information elements of the 2 domains are linked in a network structure, allowing the user to match himself with a current world situation (the patient) and to switch from it to the library information network. This feature would lead the user to the perception and the awareness of the relationship between facts (presented in the case report) and concepts (expressed in the library).
The CTP Project: Computerized Textbook of Psychology BIBA 444-449
  Keiko Hatamoto; Shigekazu Ishihara; Haruo Hayashi; Shogo Sakata; Rei Ihara; Atsushi Kawamura; Akira Yamagami
The present paper introduces the "CTP Project", that intends to develop a computer assisted instruction system in the field of psychology. This system consists of a) main instruction frame system that works as a platform for easy and integrated presentation of what we know in psychology, as text, picture, animation and simulation; b) authoring system that allows instructors to establish the knowledge-modules describing psychological facts and theories; and c) presentation managing system so that we can customize the presentation sequence and to modify the interconnections among the knowledge-modules.
Testing Acceptance and Usability of Multimedia in Man-Machine Communication: A Case Study and Guidelines BIBAK 450-455
  Kaisa Vaananen; Dave Henderson
This paper describes the evolution and application of a method of usability testing. The case study evaluates the human-computer interface of a multimedia information system. The method of evaluation can be integrated into the design process of similar highly interactive systems. The objective of the work described in this paper is to improve the chances of success of any multimedia system in casual and long term use. This is done by offering two types of guidelines: 1) For usability testing of interfaces to multimedia systems, and 2) for the design of these interfaces. These guidelines are targeted at the designers and developers themselves, i.e. no expert-level knowledge of human-computer interaction is assumed.
   ShareME, the multimedia information system, was tested for its usability among 39 target users. During the case study it was discovered that seemingly small changes to the interface can turn a partly frustrating system into a pleasant and engaging one.
Keywords: Usability test, Guidelines, Multimedia, Human-computer interaction
Using QFD Techniques in User Interface Specification: First Experiences BIBA 456-461
  Pekka Ala-Siuru
In this paper we present a framework for applying QFD (Quality Function Deployment) techniques to user interface design in embedded real-time computer systems. We also discuss value analysis as a design method, and finally we describe the experiences that we have learned in applying QFD in an industrial design case.

IV. User Issues

The Influences of Information Processing Strategies on Human-Computer Performance BIBA 463-468
  K. M. Stanney; G. Salvendy
The objective of this study was to determine if more efficient performance in hierarchical computer environments could be obtained from field-dependent (low technical aptitude) individuals by manipulating information structuring requirements. Three levels of structuring were investigated: recognizing existing structure, structure recognition with concurrent task processing, and imposing structure where no salient organization exists. Thirty-six subjects, 18 identified as field-dependent and 18 as field-independent performed 60 information search tasks under three task conditions, two structured by the experimenter and one by the subjects. The results indicated that the effects of differences in the organization of task information on computer performance time can be controlled for by providing subjects with a period of time dedicated to the acquisition of a system's structure.
Metaphor Design and Cultural Diversity in Advanced User Interfaces BIBA 469-474
  Aaron Marcus
Sophisticated user interfaces will need to communicate facts, concepts, and emotional values in multimedia environments, especially for large amounts of data and large numbers of functions in consumer products for increasingly diverse users. Using metaphors to embody complex structures and processes is one technique familiar to user interface developers. The article discusses kinds of metaphors, metaphor design scenarios, and demonstrates cultural diversity in user interface design.
Usability Evaluation from Users' Point of View: Three Complementary Measures BIBAK 475-480
  Edo M. Houwing; Marion Wiethoff; Albert G. Arnold
In the European ESPRIT project 'MUSiC', metrics, methods and standards are developed for industrial use. In this paper a validation study is reported in which metrics of cognitive workload, performance and subjective usability are tested. Subjects were studied working with a menu oriented and with a graphical object-oriented wordprocessor. The main hypothesis of the study is that the packages induce a different level of user efficiency. All measures added unique information to the usability evaluation, and the combination of the domains of measurement is most informative. Contrary to the expectations, it could not be concluded unambiguously, that user efficiency is better when individuals use a package with a graphical object-oriented interface.
Keywords: Usability, Laboratory experiment, User efficiency, Metrics, Guidelines, Mental effort, Workload
Modeling Multispecialist Decision Making BIBA 481-486
  Michael Lewis; Katia Sycara
Very often, complex decisions must be made by a group of specialists rather than a single decision maker. To make an effective decision, the combination of the group's expertise must be brought to bear on the situation. Fusing expertise where individuals have very detailed knowledge in their own areas and much weaker understanding of others introduces many difficulties: (1) agents cannot communicate their expertise in an intelligible way to non-experts because of differences in vocabulary and conceptual content, (2) the process allows for incorrect inferences, and (3) no one knows what anyone else needs to know. This impasse cannot be broken until shared mental models are developed to provide a level of agreement in evaluating alternatives needed to focus the activity of the group. This paper presents a model of decision making by teams of specialists in which agents' evaluations confound expert and naive inferences in judging alternatives. The robustness of the model in accommodating a variety of communications and modeling their impact on group process is illustrated using examples involving a turbine blade and a tricycle from the domain of cooperative design.
Decision Maker's Knowledge Level and the Selection of Decision Strategies in Using a Decision Support System BIBA 487-491
  Hyung-Min Michael Chung
The study intends to examine the effects of task complexity and knowledge level on the selection of decision making strategies in using a decision support system (DSS). A theoretical framework is depicted and a laboratory experiment at a management task domain is planned.
Use of Seamless Access Protocol to Expand the Human Interface of Next-Generation Information Systems and Appliances BIBA 492-497
  G. C. Vanderheiden
The area of appliance-level information/transaction systems exists now only in its infancy, but will, as it matures, emerge even more rapidly than did personal computers. This is in part because the costs will be lower. More importantly, however, the information systems will be designed to be very user-friendly, they will address the information needs of daily living of a much broader range of people, and they will be delivered directly into our homes as well as public environments. As a result, these technologies are likely to affect the lives of a much greater number of people with disabilities than has the use of personal computers. The importance of extending the human interface of these systems to include access by people with disabilities will therefore be more important, and may be required by law.
   To address this issue, a "seamless" approach to extending the interface of public and standard consumer product information appliances is proposed to allow their use by people with reduced physical or sensory abilities. This seamless interface protocol would allow an interface to take full advantage of the sensory and motor systems of people without limitations to maximize their ease and convenience in using the information systems (including graphics and touch windows) while providing the interface adjustability to adapt to people with temporary or permanent limitations in their physical or sensory abilities (including total blindness) to allow them to access and use the information systems as well. Key to the seamless approach is the ability to provide access to the information using the same underlying flow control (rather than a separate interface protocol for specific disabilities) and its ability to adapt to a wide variety of interface/presentation designs. Using the same underlying flow control allows users with only mild limitations to invoke only those aspects of the interface options necessary to meet their needs, and allows them to work the system in the same basic manner and to access the same information as their peers (both without and with severe disabilities). The use of a general protocol rather than a specific interface format provides manufacturers with the ability to vary the interface for their products as is necessary to innovate and to differentiate their products from others. The use of standard underlying and predictable interface/control options and conventions (the protocol) allows people with sensorimotor limitations to approach and use these very different-appearing systems without special instructions (in the same way that conventions are used in most interfaces, so that the public at large is be able to figure out new information systems and kiosks as they encounter them).
Work Possibilities for Visually Impaired: The Role of Information Technology BIBA 498-503
  E. Hjelmquist; B. Jansson
The study reports results from interviews with ten visually impaired persons, employed at different work places, in all eight. For each person there is a description of the information technology (in a wide sense) used by the employee. Examples of questions included are the effect of computerization on work tasks, social contacts at the work place, short and long-term effects of computerization, and effects in particular for visually impaired. The data will be reported as summaries of case studies, but tentative generalizations will also be made.
Ergonomic Considerations for Communication Technologies for Deaf and Hard of Hearing People BIBA 504-509
  J. E. Harkins
This paper introduces technologies used by deaf and hard of hearing people in the U.S. for communication, and suggests strategies for making products accessible to these populations.
A Document Retrieval System for the Personal Computer BIBA 510-515
  Nikki Reynolds; Michal Cutler
The problem of locating a particular file from among all those stored on an individual's personal computer disks is becoming increasingly difficult. Directory (or "folder") and file names are both too difficult for a user to remember, and not sufficiently descriptive to remind a user of a document's contents or purpose. New systems need to be developed which capture, and make effective use of, those features of a document that users find truly memorable. In this paper, we describe a prototypic retrieval system that was developed to conduct experiments with users. Through these experiments, we hope to identify memorable document features, to develop methods for exploiting them.
Combining Asynchronous and Synchronous Collaborative Systems BIBA 516-521
  Gunnar Teege; Uwe M. Borghoff
There are a variety of different CSCW applications for group collaboration. Among others, systems for asynchronous coordination as well as synchronous multiuser editors are current research topics. Each system has its advantages. Thus, combining the advantages of such systems should be even more advantageous.
   We have developed a new object-oriented, asynchronous activity support system, called TACTS, and a new synchronous multiuser editor, called IRIS. We explain how both systems can be combined using the underlying state models, and show the advantages of the combination.
The Relationship Between Task Structure and Choice of Navigational Aid in Human Computer Interface Design BIBA 522-527
  Rebecca S. Lipner; Gary W. Strong; Karen E. O. Strong
An experiment, using 163 subjects, investigated the extent to which visual momentum is supported by four computer task and navigational aid combinations. Results demonstrated a task-dependent interference effect for learning a cognitive map of the display network. Subjects who concentrated on method rather than outcome were impaired in learning the organization of the display network. This finding suggests that outcome-based tasks seem to allow the user to explore the display network and learn its organization whereas method-based tasks require attention-diverting serial visual searching for objects on a display. When a map of the display network was provided as a navigational aid, performance was improved regardless of whether the task was method-based or outcome-based. However, results showed some evidence that method-based tasks require more spatial inferencing than do outcome-based tasks even when a map is present. These findings suggest some guidelines for the design of human-computer interfaces.
Articulating the Design Process: The Effects of Verbalisation Upon Design Strategies BIBA 528-533
  Simon P. Davies
Many recent studies of the software design activity have shown that design is inherently opportunistic and that it displays few of the characteristics attributed to prescriptive models. This paper suggests that these more recent studies may be methodologically flawed, arguing that the elicitation of concurrent verbal protocols may cause opportunistic deviations from a top-down approach. Moreover, the use of verbal protocols to characterise temporally bound events, such as a design decomposition, has implications for the way in which such events are described. The paper reports a study of a software design task which shows that described behaviour elicited in the form of concurrent verbal protocols can differ significantly from observed behaviour. It is argued that this arises because of a tendency to linearise verbal descriptions of non-linear behaviour. This paper suggests that some care should be taken when using verbal protocols to characterise events which have a strong temporal dimension such as design. Finally, the implications of this work for tool building and for characterisations of design problem solving are considered.
User Participation in Systems Design -- Results of a Field Study BIBA 534-539
  A. Beck
Active user involvement is one of the most important factors for successful system development and human-computer interface design. This is widely recognized both by software engineers as well as end users. However, for most software developers it is still a difficult task to properly involve users in their projects. A field study of 10 projects was conducted to enlighten the process of user participation and to gather valuable information both from the users and the developers. Results show that there are deficits in the organization of user participation, user motivation is weak, and users and developers have insufficient qualifications. Users and developers have different requirements and different abilities which can often result in conflicting situations. As the results indicate, a better organizational support is needed of the users' management. In addition, software developers need better methods and tools for successful user involvement.
Collaboration of Line and Staff in Fully Automated VLSI Factory BIBA 540-545
  Yoshimi Gamo; Wataru Susaki; Kouichi Kurokawa; Hirokazu Miyoshi; Toshio Hori
According to the natural and historical background, Japanese top managers usually consider the progress and happiness of the workers most important. A fully automated VLSI factory was built based on this concept in 1983, where the workers' human skill was regarded essential as well as the engineers' high technology. As the results, a close collaboration organization was established among the workers, the engineers and the facilities.
Providing Plant Design Knowledge to the Operators BIBA 546-551
  Kari Kaarela; Pertti Huuskonen; Juha Jaako
The user interfaces of modern automation systems as well as plant documentation are mostly device-oriented and structured according to the process or device hierarchy. Higher level concepts, such as the goals of the plant or the tasks of the operator, are rarely visible to the operators. However, these concepts do exist during the design process, but are neither recorded nor communicated and thus remain vague. This may lead to significant differences between the operators' understanding about the process they are assigned to, and to less-than-optimum performance during some shifts.
   In our view, these concepts should be captured when and where they are initially born -- in the design process. However, tools and representations in current design practice cannot deal with such knowledge. We propose the use of Multilevel Flow Modeling in industrial plant projects to provide a common conceptual model among the various participating design disciplines and finally the end users. In this paper, we present how such a model can be utilised as a guideline for the documentation and user interface development.
A Framework for User Customization BIBA 552-557
  Haiying Wang; Mark Green
We present a domain independent object-oriented framework for supporting programming-by-example in user interfaces. It simplifies the construction of domain-specific customizations by providing programming abstractions that are common across domains. It defines three basic abstractions: primitive customization objects encapsulate the recording and replay mechanisms for individual interactive objects in a domain, the modeling component records user's interaction, identifies the user's behavior pattern and creates a simulation agent; and the simulation agents perform the tasks on behalf of the user. This paper describes the methodology as well as design and implementation issues of the proposed framework.
Evaluation of Communication Methods for User Participation in Data Modeling BIBA 558-563
  Ch. Kirsch
For the construction of data base systems the process of data modeling is becoming increasingly important. The development of a data model requires the cooperation of software engineers and employees of an organization as the potential users to analyze the specific business terms, to specialize the data objects and to create the definitions of the data elements. Questionnaires, interviews or workshops can be used as empirical methods to improve the communication between software specialists and work specialists. The purpose of this study was to investigate which of these methods provides the best means to further the collaboration. In a data modeling project these three methods were used to collect information on the definitions of the entity-types of the data model. With each method, six selected employees of the participating departments accomplished the task. The final data model was used to evaluate the results of each method. The workshop led to the best results with respect to the effort for the data analysis and the quality of the results. The acceptance of the entity types was higher when the respective definitions had a higher proportion of specifications developed in the workshop. The percentage of specifications from the document analysis without user participation also correlated with better acceptance.
User-Centered Guidance for Environmental Management BIBA 564-569
  David R. Eike; Jeffery A. Fox; Richard Dailey
Environmental management requirements facing U.S. Department of Energy (DOE) facilities are detailed, complex, and often subject to rapid change. DOE's Office of Environmental Guidance, RCRA/CERCLA Division (EH-231), is responsible for developing and issuing guidance to assist DOE facilities in interpreting and complying with Federal, State, and local requirements. Recognizing the potential for computerization of the guidance, EH-231 requested that Pacific Northwest Laboratory (PNL) develop an approach for preparing automated guidance. The approach developed by PNL, termed "user-centered guidance," combines participatory design and traditional rapid prototyping techniques to produce a new form of environmental guidance that emphasizes the user's needs. This paper describes the objectives, processes and current status of this effort.
The Effect of User Characteristics on Interface Choice BIBA 570-574
  R. C. MacGregor; H. Hasan; H. T. Liao
Many empirical studies of interface choice and effectiveness suffer because of the mechanisms used to choose sample users (referred to in this paper as subjects). Subjects described as experts are very often trained beginners. This study examines the choice of interface for problem solving. Expertise is defined in terms of a subjects ability to solve problems. Four hundred and thirty one subjects were given a series of problems to solve using an automated pseudocode program PSC. The subject's expertise in problem solving was examined at several points throughout the study and this was compared to the choice of interface (menu/hot key). The results showed that there was a significant correlation between the subject's expertise and the interface choice. However, while the above average group tended to distribute themselves equally across both interfaces, the average and below average groups oscillated from one interface to another as problems increased in difficulty.
Work Analysis -- Perspectives on and Requirements for a Methodology BIBA 575-580
  Peter H. Carstensen; Kjeld Schmidt
Today computers are used as support tools in work domains with a high degree of complexity. Because of the increased complexity and the high amount of relevant aspects and perspectives of the work, the traditional computer science analysis methodologies for analyzing and describing the work often fail modeling the relevant aspects. It is argued, that an analysis methodology that grasps all relevant aspects of the work domain is heavily required. Existing methodologies are insufficient and only fragments of a framework for work analysis exist. A framework or a methodology must be based on a theory of work rather than a theory of computer based systems to avoid automation of non-automatable practice and to support users in managing unforeseen contingencies. Central characteristics of complex work domains and problems in existing methodologies are discussed. Using this as a basis, a number of requirements for work analysis methodologies are identified.
Formalisms for Cognitive Modeling BIBA 581-586
  Neville Moray
The notion of a "mental model" or "cognitive model" is widely accepted in the engineering psychology community, especially among those who model human performance in large complex systems. Several meanings have been given to the word "model", from computer programs to diagrams. In this paper it is proposed that the notion of a model as a mapping provides a definition which is sufficiently strong to encompass most current usage. The formal language for expressing the notion of modeling is a lattice algebra of homomorphic relations.
A Framework for Measuring Cognitive Complexity of the Human-Machine Interface BIBA 587-592
  Yan M. Yufik; Thomas B. Sheridan
This paper seeks to define a framework for measuring cognitive complexity of interactive control tasks. We discuss new complexity measures vis-a-vis experimental evidence about sources of complexity and the underlying neuronal mechanisms.
The Role of Mental Models in Complex Dynamic Environments BIBA 593-598
  Craig P. Speelman; Daniel Boase-Jelinek; Kim Kirsner
The concept of a mental model has been used extensively in many areas of HCI. Unfortunately, after two decades of use, the concept still suffers from vague definitions and conceptualisations. At best the concept is used to denote anything from schemes to strategies, and at worst it is used as a synonym for knowledge itself. This report describes work on a project that involves an attempt to operationalise the concept of mental models. The aim of this project is to develop techniques for training subjects to operate a complex system. In terms of operationalising the mental model concept, the major approach is to encourage subjects to view the system in a particular way (i.e., adopt a particular mental model of the system) such that only one type of performance strategy is possible. If subjects exhibit this strategy, then they will exhibit evidence of working with a particular mental model. The particular domain examined in this project is power system control, although the aim is to develop generic training principles. This report describes how consideration of operators' mental models of a system can guide the design of system interfaces. The project has demonstrated that, although interface design may benefit from such consideration, changes to interfaces may have undesirable effects in terms of performance strategies developed to operate such interfaces. This project has also demonstrated the benefit of transforming a static task into a dynamic one, provided that operators can acquire proactive performance strategies. The relationship between task design and the acquisition of proactive control processes is discussed.
Performance, Mental Models and Background Knowledge in Mastering a Simulated Dynamic System BIBA 599-604
  Eva Carling
An experimental method was tried out, where subjects controlled a simulated process and control strategies were described in several ways, a few of them developed from statistical analysis of on-line registered data. Some significant differences in strategies were shown, and developed measures discriminated between groups of subjects with varied backgrounds.
On Psychophysiology of Cognitive Complexity BIBA 605-610
  Leo Vekker; John A. Allen; Yan Yufik
This paper discusses correlation between cognitive processes and subject's physiological conditions. Formation of visual and haptic images, picture recollection, and word association were experimentally examined. Accompanying physiological indicators, however imprecise, displayed regularities, interpreted in this paper in terms of information related energy demands (Vekker, 1976), and more recent models of cognitive complexity (Yufik, 1988).
   We preface description of experiments by some definitions, to be used in later analysis. This paper distinguishes four cognitive functions: sensation, perception, recollection, and conceptual thinking (Vekker & Allen, 1993). Although addressed distinctly in the experiments, these functions remain, in fact, intimately integrated components of cognitive performance.

V. Methodologies

Quantitative Measures for Evaluating Human-Computer Interfaces BIBA 612-617
  M. Rauterberg
There currently are 4 different views on human computer interaction in measuring interactive qualities: (1) the interaction-oriented view, (2) the user-oriented view, (3) the product-oriented view and (4) the formal view. Two different possibilities of measurement within the product-oriented view are introduced in this paper. Different types of user interfaces can be described and differentiated by the concept of "interaction points". Regarding to the interactive semantic of "functional interaction points" (FIPs), 4 different types of FIPs must be discriminated. Based on the concept of FIPs, the dimensions "[visual] feedback" and "interactive directness" can be quantified.
The Fatigue Check System for VDT Workers by Measuring CFF BIBA 618-623
  Masaharu Takeda; Yoshio Hayashi; Kaoru Suzuki
Checks on fatigue experienced by VDT operators caused by stresses require an objective index in addition to a subjective evaluation of fatigue. A fatigue index must be simple and voluntary and must allow accurate data processing and storage and display of results. A fatigue index must also allow individuals to make self-assessments to contribute to health management. To accomplish this objective, a system has been designed and manufactured to calculate average measured values and standard deviation and to plot X-s or X-R control charts using data storage for self-assessments using the fatigue measurement index as a CFF and by achieving simple and accurate measurements by the adjusting method.
A Code Key-In Time Estimation Method for College Students BIBA 624-629
  S. Kishino; Y. Hayashi
An estimation method of standard time required for almost every student to key-in one's code written in Pascal programming language is proposed. Actual time students need to key-in varies depending on their knowledge. This method does not estimate an individual key-in time. So it is not our interest to evaluate of each student's ability.
   The features of our estimation process are as follows. First, we defined short parts of the Pascal code named reading units and assumed that all students read a code and type keys with every reading unit which is fixed for all students. Secondly, we adopted a method Method Time Measurement (MTM) to determine the required time. MTM is one of the Predetermined Time Standard (PTS) methods and is usually applied to the field of industrial work study. In the third place, we introduced only six parameters that varies Key-in time. They were average moving distance of fingers (MD(cm)), average time to read a reading unit (RT(second)), occurrence ratio of input errors (IER), occurrence ratio of modification errors (MER), and two real numbers (R1 and R2: 0<=R1<=R2<=1) which determine the difficulty in moving fingers and searching characters.
Direct Control of the Computer Through Electrodes Placed Around the Eyes BIBA 630-635
  James Gips; Peter Olivieri; Joseph Tecce
A system has been developed that allows an individual to communicate with a Macintosh computer solely through electrodes placed around his eyes. The user controls the cursor on the screen simply by moving his eyes and head. The user can spell out messages on the Macintosh, play tic-tac-toe, and even play commercial video games by controlling the computer through the electrodes.
Qualitative and Quantitative Analysis of Human Decision-Making Complexity BIBA 636-641
  Val F. Venda; Hal W. Hendrick
The principles of macroergonomics, coupled with the theories of mutual multi-level adaptation and transformation dynamics are proposed as a basis for analyzing human decision-making complexity and optimizing system design. The concept of macroergonomics is described. Studies relating cognitive, decision-making, and organizational complexity to system design are noted. The application of hybrid intelligence theory is proposed as an effective means of managing organizational and decision-making complexity. The process of mutual adaption being applied at Northern Telecom Canada, Ltd., including a concrete methodology, is described.
Quantitative Evaluation and Performance Prediction BIBA 642-647
  Yan M. Yufik; Thomas B. Sheridan; Valery F. Venda
The growing complexity and less than hundred percent reliability of modern man/machine systems, combined with the high cost of operator mistakes, makes it exceedingly important to obtain practical quantitative methods that would allow designers to mitigate complexity and predict operator performance at the early stages of design. Although the need for such methods has been long recognized, their development and acceptance proved to be slow. This paper tries to underscore some of the bottlenecks, and invite a discussion on how they can be overcome. We start by an example substantiating the value of quantitative predictive methods. Here, we refer to the Army/NASA Aircrew-Aircraft
   Integration Program (A{cubed}I) at NASA Ames Research Center. This example is followed by a general definition of the levels of man-machine adaptation. Current quantitative methods for performance prediction and operational complexity assessment are discussed next. The paper concludes by summarizing what we believe are the key points in making quantitative methods more useful for practical applications.
Data Chunking and Interface Display Design BIBA 648-653
  V. Venda; L. Beltracchi
Information chucking within an interface is an integration of related data, items, and facts to adapt information to human cognitive strategies. Studies (NUREG/CR-5977) found that the chucking of data into an information model did improve the efficiency of the decision making process. For power plant process data, a computer-driven Rankine Cycle display automates the integration and mapping of data into process functions and the heat engine cycle. The opinions and viewpoints herein are those of the authors and they do not necessarily reflect the criteria, requirements, and guidelines of the U.S. Nuclear Regulatory Commission.
The Quadrigram as a Model for Dynamic HCI Studies and Design BIBA 654-659
  Val Venda; Doug Strong; Ian Dromey
The Quadrigram is presented as a model for use by HCI specialists for studying dynamic interaction in human-machine-environment systems. The Quadrigram describes the most important parameters and parameter deviations in a system including machine-environment factors, human perception, real performance strategies, human performance dynamics and environmental influence. The Quadrigram introduces a new perceptual framework for problems involving human performance in complex environments and has applications in the modelling of dynamic situations in ergonomics, human-computer interaction, psychology and automatic control systems.
Performance of a Usability Assessment Model: A New Zealand Case Study BIBA 660-665
  Sarah V. Burger
An attempt to address how to assess system usability resulted in a methodology based on an integrated approach to usability assessment which combines critical, user, and quantitative evaluation methods. Though this model incorporates crucial cognitive, ergonomic, and interface design principles, its effectiveness in industry based environments needed to be tested. This paper describes the performance of this model compared to an industry based practice in New Zealand, and discusses the implications of these results.
Problem Discovery in Usability Studies: A Model Based on the Binomial Probability Formula BIBA 666-671
  James R. Lewis
Product developers want their products to be as easy to use as possible, but must consider constraints such as cost and schedule. The primary goal of many usability studies is to discover design problems. After discovery, designers can take steps to eliminate or minimize problem impact. This paper shows that problem discovery in usability studies is consistent with the binomial probability formula. The problem discovery curves from two recent studies lend empirical support to this problem discovery model. One practical application of the model is to help estimate appropriate sample sizes for problem discovery usability studies. This model can help usability researchers simultaneously consider cost (minimized by running as small a sample as possible) and risk (minimized by running as large a sample as possible) to maximize the efficiency of a study.
Modelling Collective User Satisfaction BIBA 672-677
  F. M. T. Brazier; Zs. Ruttkay
In many situations in human-human interaction people strive for collective satisfaction of all partners. In this paper we claim that in certain types of information retrieval systems it is more appropriate to aim at collective user satisfaction than at individual user satisfaction. We discuss two techniques to model and maintain collective user satisfaction.
Explanation Evaluation Based on Referential Models BIBA 678-683
  Alain Giboin
As opposed to the <> of Explanatory Knowledge-Based Systems (EKBS) evaluation, we present the <>. From this alternative standpoint, the evaluation referent is not only defined in terms of criteria, but also and specially in terms of the referential models which underlie them. Referential models are physical or mental models to which the evaluator refers during her activity, and which allow her to generate criteria. As a result, referential models motivate the criteria. To show the relevance of the view, we report the results of an empirical study dealing with the evaluation of explanations about <>. In conclusion we outline some implications of the results for an EKBS evaluation methodology.
From the User Interface to the Database Management System: Application to a Geographical Information System BIBA 684-689
  M. Mainguenaud
This paper presents the links between a visual programming language, Cigales, and a database management system in the context of a Geographical Information System. We study the way of filling the gap between the semantic level of a visual query language and a data base query language.