HCI Bibliography Home | HCI Journals | About IJHCI | Journal Info | IJHCI Journal Volumes | Detailed Records | RefWorks | EndNote | Hide Abstracts
IJHCI Tables of Contents: 01020304050607080910111213141516171819

International Journal of Human-Computer Interaction 9

Editors:Michael J. Smith; Gavriel Salvendy; Masamitsu Oshima
Publisher:Ablex Publishing Corporation
Standard No:ISSN 1044-7318
Links:Table of Contents
  1. IJHCI 1997 Volume 9 Issue 1
  2. IJHCI 1997 Volume 9 Issue 2
  3. IJHCI 1997 Volume 9 Issue 3
  4. IJHCI 1997 Volume 9 Issue 4

IJHCI 1997 Volume 9 Issue 1

Neural Networks in User Modeling and Intelligent Interface

Introduction BIB 1
  Nong Ye
Neural Networks Approach to User Modeling and Intelligent Interface: A Review and Reappraisal BIBA 3-23
  Nong Ye
This article provides background and an overview of the neural networks approach to user modeling and intelligent interface. Activities in user modeling and intelligent interface are discussed. Intelligent architectures of various neural network models are introduced to illustrate how neural networks solve problems and learn knowledge of problem solving. The introduction of neural networks helps in associating intelligent capabilities of neural networks with the activities in user modeling and intelligent interface. The practical utility of neural networks in supporting user modeling and intelligent interface is demonstrated by reviewing neural networks developed in this area. Issues relating to the design of neural networks in various contexts of user modeling and intelligent interface are also addressed.
Modeling a User's Domain Knowledge with Neural Networks BIBA 25-40
  Qiyang Chen; A. F. Norcio
This article presents a neural network approach for user modeling. A set of neural networks is utilized to represent and infer users' task-related characteristics. These networks function as associative memories that can capture the causal relations among users' characteristics for the system adaptation. It is suggested that this approach can be expected to overcome some inherent problems of the conventional stereotyping approaches in terms of pattern recognition and classification of user characteristics.
A Comparison of Three Methods of Task Analysis: Cognitive Analysis, Graph-Matrix Analysis, and Self-Organizing Networks BIBA 41-57
  John F. McGrew
Three methods of performing a task analysis are compared: cognitive analysis, graph-matrix analysis, and self-organizing networks. Cognitive analysis relies on the ability of an observer to abstract and generalize over situations. Graph-matrix analysis is valuable for its precision and inclusion of details. Neural networks have an ability to generalize uninfluenced by observer bias. Comparison demonstrates that each method misses some important but different aspect of human-computer interaction. The cognitive analysis included infrequently used information that was not captured by direct observation. The graph-matrix analysis included frequency of use information and details missed by the cognitive analysis. The self-organizing network generated an alternative view of the task structure that was not influenced by observer bias. It showed that the underlying structure for the user-computer interaction in this study was the structure of the computer system itself.
A Neural Network Approach to Tracking Eye Position BIBA 59-79
  Bryn Wolfe; David Eichmann
The design of a neural network-based eye tracker is presented. A series of experiments with counterpropagation neural networks convert synthetic video images into eye coordinates by an enhanced feed-forward neural network with multiple winning hidden layer nodes. Difficulties encountered during the design process are discussed. The results show that accurate, fine-grained tracking of a human's eye position is possible by processing the video image collected from a goggle-mounted miniature charge-coupled device (CCD) camera.
Automatic Setting of Article Format Through Neural Networks BIBA 81-100
  Nong Ye; Baijun Zhao
The automatic format setting of journal articles for reducing the workload of computer users involves two processes: automatic acquisition of article format and automatic recall of article format. Several neural networks have been explored to implement the two processes. The advantages and disadvantages of these neural networks are evaluated in comparison with capabilities of conventional computer programs. A heteroassociative back-propagation network has been developed for the automatic acquisition process. This network excels over computer programs because of its abilities in learning and generalizing implicit knowledge from examples. A bidirectional associative memory network, a Boltzman network, and an autoassociative back-propagation network have been investigated for the automatic recall process. None of them excel over computer programs in terms of recall accuracy.

IJHCI 1997 Volume 9 Issue 2

Workload when Using a Mouse as an Input Device BIBA 105-118
  Arne Aaras; Ola Ro
A newly developed mouse, which seemed to give the operator a more neutral forearm position, was compared with a traditional mouse. The muscle load was recorded by electromyography from the extensor digitorum communis, extensor carpi ulnaris, and trapezius muscles. The muscle load of the forearm was significantly less when using the new mouse compared with the traditional one. This was true for the extensor digitorum communis regarding the static (p = .0005) and the median (p = .001) values of the amplitude distribution function (ADF) and the number of periods per minute when the muscle load was below 1% of maximum voluntary contraction (p = .03). The same clear tendency was also found regarding the static muscle load from the extensor carpi ulnaris (p = .06). These results indicate the need for reducing the pronation of the forearm when working with a mouse.
The Effect of Pictogram-Based Interface Design on Human-Computer Performance BIBA 119-131
  Laura A. Miller; Kay M. Stanney
Metaphors are commonly used in human-computer interface design. These metaphoric designs, however, often do not extend beyond the surface-level interface. It stands to reason that a more valid test of the benefits of such context-based designs would be derived from an analysis of metaphors that more closely approximate human-environment interaction. In this study, the influence of pictogram-based versus symbolic interface designs on human-computer performance was empirically evaluated. First, a Goals, Operators, Methods, Selection rules analysis was performed to determine how closely each of the designs assimilated the actual procedures an individual would use in real life. As expected, the symbolic design was found to break from the analogy at an earlier stage than the pictogram-based design, thus providing impetus for empirical testing. Novice and expert participants were given a set of computer-based tasks to perform using both designs. Dependent variables were performance time, total number of steps to task completion, and subjective reports of workload. The performance results indicated that the novices significantly benefited from the use of the pictogram-based design in terms of performance time and efficiency. In addition, the performance of experts was significantly more efficient while using the pictogram-based design. These results provide empirical evidence to support the current trend in pictorial human-computer interface designs.
Desktop Agent Manager (DAM): Decision Mechanism BIBA 133-149
  Deborah Ann Guerrera Ercolini; Mieczyslaw M. Kokar
Desktop software agents are background processes that notify a computer user of certain predefined events. The complexity of desktop computing is increasing due to the proliferation of software agents. A desktop agent manager (DAM) may become an important component of desktop computing. This article focuses on the decision mechanism for the DAM to decide which agents should be allowed to access the user, which agents' results should be stored for future review by the user, and which of the agents should be filtered out. We prove the validity of the concept of a DAM by developing an architecture that includes both a prototype of the DAM and a simulator of various types of agents. In this article, we present the results of our simulations and analyses of the DAM decision mechanism.
Toward a Theory of the Effectiveness of Multimedia Systems BIBA 151-168
  Martijn Hoogeveen
This article presents steps toward a theory of the basic effects that multimedia systems have on people. These effects are grouped into knowledge transfer, entertainment, and data processing. Although based on empirical investigations, several components of the theory are controversial and in need of further critical empirical investigation. This article criticizes the optimistic multimedia paradigm, the dominant conviction among multimedia system developers and users that adding multimedia functionality to information systems (always) leads to improved information and knowledge transfer. Adding multimedia functionality is not sufficient for these and other learning effects. The role of independent variables that facilitate multimedia effects is reviewed and a more moderate, realistic multimedia paradigm is proposed.
Effects of Color Combination and Typography on Identification of Characters Briefly Presented on VDTs BIBA 169-181
  Kong-King Shieh; Ming-Te Chen; Jiunn-Huei Chuang
This study examines the effects of text and background color combination and typography of characters on visual performance and visual fatigue of individuals working on a visual display terminal (VDT). Correct percentage of performance, critical fusion frequency, near point of convergence, subjective visual fatigue, and preference ranking of color combinations of 40 participants were collected and analyzed. Color combination had no significant effects on visual performance. Typography significantly affected performance. Aesthetically pleasing but more cluttered characters were detrimental to visual performance. Characters of higher frequency and fewer strokes were identified more accurately. Participants also showed differential preference for color combinations. Red-on-green was ranked inferior to color combinations generally used in computer software. Participants showed no visual fatigue during the VDT work.
Individual Differences in the Use of Command Line and Menu Computer Interfaces BIBA 183-198
  S. J. Westerman
This article presents an experimental investigation of the process of computer-based command generation. The comparative cognitive demands imposed by menu and command line interfaces are examined in relation to individual differences in expertise and cognitive ability. Three-way interactions between associative memory, expertise, and command generation method indicated similarities in the performance of expert participants with low associative memory and that of novices. Spatial memory also interacted with expertise, with novices with low spatial memory performing more poorly than any other group. Implications for interface design are considered.

IJHCI 1997 Volume 9 Issue 3

Development and Evaluation of the Windows Computer Experience Questionnaire (WCEQ) BIBA 201-212
  Laura A. Miller; Kay M. Stanney; William Wooten
The software market has been inundated with Windows-based application programs claiming increased usability and convenience. Although this trend is indeed prolific, it has resulted in two important implications: (a) an increase in the need to select employees with high levels of Windows-based computer expertise or to identify current employees who require enhanced training, and (b) an increase in the need to measure user expertise to support human-computer interaction research. Despite these increasing demands, questionnaires used to determine general computer experience are scarce. Furthermore, questionnaires regarding computer experience in a Windows environment are seemingly nonexistent. A reliable means of measuring experience in a Windows environment could substantially facilitate both human-computer interaction research and training. This article describes the procedures used to develop and test the reliability of the Windows Computer Experience Questionnaire (WCEQ). A test-retest correlation revealed that the WCEQ is a reliable measure of computer experience. Furthermore, a subsequent factor analysis revealed that the WCEQ is composed of four main factors: general Windows experience, advanced Windows experience, formal instruction, and reliance on help functions.
Heuristic Walkthroughs: Finding the Problems Without the Noise BIBA 213-234
  Andrew Sears
Inspection-based evaluation techniques are popular because they require less formal training, are quick, can be used throughout the development process, do not require test users, and can result in finding numerous usability problems. A new technique is described that combines the benefits of heuristic evaluations, cognitive walkthroughs, and usability walkthroughs. This technique, a heuristic walkthrough, provides more structure than heuristic evaluations but less than cognitive walkthroughs. The result is an effective task-oriented evaluation technique that is easy to learn and apply. Methods are proposed for comparing the validity, thoroughness, and reliability of evaluation techniques. Finally, heuristic walkthroughs are compared to heuristic evaluations and cognitive walkthroughs in a controlled study. The results indicate that heuristic walkthroughs are more thorough than cognitive walkthroughs and more valid than heuristic evaluations. In other words, heuristic walkthroughs resulted in finding more problems than cognitive walkthroughs and fewer false positives than heuristic evaluations.
The Effect of the Interface on Learning Style in a Simulation-Based Learning Situation BIBA 235-253
  Sissel Guttormsen Schar; Christoph Schierz; Francois Stoll; Helmut Krueger
The applicability of a cognitive theory of implicit and explicit learning to computer-aided learning was tested with a two-factor experimental design. The interface factor tested whether a direct manipulation and a command-based interface can induce the users to implicit or explicit learning, respectively. The complexity factor tested whether the benefit of using one of the interfaces (i.e., learning modes) depends on the task complexity (high or low). According to the theory, the success of using one of the learning modes depends on characteristics of the tasks. The application of the theory to computer-aided learning would be to characterize the learning task and then choose the interface that best fits the task requirements. The ability of the user interfaces to induce the learning modes was supported by the results. The explicit learning mode observed in the users of the command-based interface resulted in better performance for both complexity conditions. Although the users of the direct manipulation interface did not fail to learn, their performance was inferior to those of the other group. Based on the results, task complexity is not suggested as a criterion for selecting interface.
Analysis of Information Utilization (AIU) BIBA 255-282
  Jan Gulliksen; Magnus Lif; Mats Lind; Else Nygren; Bengt Sandblad
Analysis of information utilization (AIU) is a method for describing and analyzing how information entities identified in information analysis are being used in the work situation. AIU aims at complementing existing methods for user interface design by identifying additional requirements for human-computer interaction. The method focuses on aspects of computer-supported work, related to cognitive load, aspects of which end users often not are explicitly aware.
   For skilled workers, in a professional work environment, the efficiency of the user interface is extremely important. We earlier stressed that important criteria for design of user interfaces deals with making the interface "obvious" to the users, by minimizing the cognitive load associated with the handling of it.
   AIU is performed through observation interviews in which human-computer experts interview representative users about their work situation and observe physical information-handling routines. The analysis identifies work tasks in terms of judgments and decision-making situations, requirements concerning the tasks that may have to be performed simultaneously, important features and priorities of the information, actions the user can initiate, and so on. The method supports the interface designer with human-computer interaction requirements structured for a workspace-oriented design. It is integrated in a user-centered development model and supports the simultaneous development of competence, organization, work activity, and information technology.
   This article describes the background and contents of the method and how the results of the analysis are documented and used in the design process. A number of application projects have shown that AIU makes it possible to capture aspects of human work and information processing that are important to the design of better interfaces. AIU is not a new method for system development but complements today's methods for task and information analysis with more design-relevant information. For illustration, an example is presented, describing interface design based on AIU in a system development project in the Swedish National Tax Board.
Preference-Based Human Factors Knowledge Repository for Designing User Interfaces BIBA 283-318
  Constantine Stephanidis; Demosthenes Akoumianakis
This article proposes and describes a representational framework and a supporting tool environment for embedding and propagating human factors expertise into high-level user interface design and development platforms. The proposed framework allows user interface designers to elicit, accommodate, and articulate user interface guidelines and results of experimental studies into reusable, evolutionary, and "living" design cases. The building blocks of the representational framework are a set of primitive constructs for consolidating the semantics of human factors knowledge into a design representation that characterizes the physical level of interaction. This is achieved through the development of a logical framework based on preference constraints and an initial set of preference and indifference expressions. The preference constraints provide a reasoning engine and a proof strategy for compiling a preference ordering of competing design alternatives and subsequently aggregating them into indifferent classes of design options per interaction element. The article also reports the implications of the proposed technique for user interface designers and the underlying requirements of user interface development platforms. Finally, the representational sufficiency of the proposed approach is discussed in the context of recent case studies aiming to consolidate human factors knowledge into a reusable repository supporting the ergonomic design of user interfaces in two example application domains, namely nonvisual hypermedia accessible to blind people and interpersonal communication aids for speech-motor-impaired and language-cognitive-impaired users.

IJHCI 1997 Volume 9 Issue 4

Longitudinal Studies of Human-Computer Interaction

Introduction BIB 323
  Pascale Carayon
Temporal Issues of Quality of Working Life and Stress in Human-Computer Interaction BIBA 325-342
  Pascale Carayon
This special issue of the International Journal of Human-Computer Interaction presents longitudinal research on quality of working life and stress in office computer work. These recent longitudinal studies examine the relation between people and their work environments in a dynamic manner. Conceptual and methodological issues related to the dynamic study of quality of working life and stress in human-computer interaction are discussed in this article.
A Longitudinal Study of Work Characteristics and Health Complaints among Insurance Employees in VDT Work BIBA 343-368
  Kari Lindstrom; Tuula Leino; Jorma Seitsamo; Ilkka Torstila
Employees using visual display terminals (VDT) in their work in the head office and branch offices of a Finnish insurance company were followed for 8 years. Their job satisfaction and health complaints were studied in relation to VDT-use characteristics and job demands and stressors. A total of 144 employees participated 3 times in a questionnaire survey: in 1985 before a major change was implemented in the data system used, just after its implementation in 1987, and later after the gradual increase of VDT application and enlargement of the data system in 1993. Most of the employees were women (65%) in customer service work. The original group of 477 decreased to these 144 employees during the follow-up due to employees leaving the company, moving to other departments, retiring, and nonresponding. The dropouts did not differ considerably from the retainees in the central variables of the study. The follow-up results showed an increase in the daily amount of VDT work and a better mastering of VDT applications. Only minor changes in job demands were found. A slight increase occurred between 1985 and 1987 in attention and psychomotor demands. The greatest increase was found in job insecurity between 1987 and 1993. The level of job dissatisfaction, psychological symptoms, and eye discomfort decreased somewhat, and neck and upper limb complaints increased between the two later study phases. The factors explaining job dissatisfaction and subjective symptoms were partly different in the 3 study phases. In the last study phase, job dissatisfaction was dependent on job insecurity, being earlier explained mainly by psychological and social job demands. Neck and upper limb complaints were in the beginning clearly related to high physical workload and later more directly dependent on VDT-use characteristics. Eye discomfort was explained inconsistently by various factors. In conclusion, the results of the subjective symptoms became more dependent directly on VDT characteristics after the major implementation of new VDT applications and as the follow-up proceeded. The results also indicated that in changing work life no constant explanatory model for the subjective symptoms could be found.
Longitudinal Study of Work Stress among Information System Professionals BIBA 369-381
  Yuko Fujigaki; Kazuko Mori
In order to examine the effect of a stressor changing over time on information system professionals, a longitudinal study on work stress was conducted. Data on job events, urinary catecholamine, salivary cortisol, and subjective symptoms were collected for 10 male engineers who were observed every 2 weeks for 5 months and every week for the following 2 months. Results show that adrenaline reflects reactions to the acute job events whereas cortisol seems to capture the chronic state of work-stress reaction. This study allows us to specify the job events that affect stress, which can be useful for intervention.
Toward a Multilevel Model in Longitudinal Studies on Computerization in Offices BIBA 383-405
  Pekka Huuhtanen
The theoretical framework of longitudinal field studies on office automation has expanded from the stress paradigm to the implementation and planning of change. Research has started to focus on the design processes of both the computer applications and work organizations simultaneously as opposed to the earlier descriptions of the effects of automation on stress and well-being. New technology should have a variety-increasing rather than a variety-decreasing effect on operations, by supporting flexible and adaptive work organizations instead of formalized hierarchical control and tight standardization of work routines. The goals of planning, the ways of organizing it, and leadership styles are decisive, as regards the benefits and quality of changes. In addition, assessment of possible new health and safety risks at the worksite level should be organized. One future risk factor is information overflow in a multimedia environment with increasing productivity demands.
   In this article, key psychological questions concerning the studies of new technology implementations in office settings from the health and safety perspective are discussed. A conceptual review of some previous studies is presented, with special emphasis on longitudinal research. The focus is more on research questions and methodology than on the presentation of empirical results as such. A multilevel process model for the study of change is presented that combines technological, organizational, and psychological change processes on both organizational and individual levels. Concerning the continuous change of information technology, work organization, and the age structure of labor force, future research needs are discussed.
New Information Technologies, Job Profiles, and External Workload as Predictors of Subjectively Experienced Stress and Dissatisfaction at Work BIBA 407-424
  Cristian Korunka; Sabine Zauchner; Andreas Weiss
Implementation cycles of information technologies in modern workplaces have become shorter, and employees have to constantly adapt to changing work situations. Presented here are preliminary results from a comprehensive research project, the 2nd Vienna Implementation Study, which investigated the effects on workers of continuous implementations, that is, the implementations of new information technologies on workplaces already equipped with computers. In a longitudinal research design, strain and dissatisfaction of 466 employees in 10 different companies are investigated. In each company, an implementation of a new information technology occurs during a specified time period. Seven measurements are taken over a 22-month period. Contextual factors postulated to moderate employee reactions to the new information technology include the following: type of implementation, implementation style (e.g., implementation management, participation), job profiles, external workload of employees, and personality factors.
   Preliminary analyses of the first two measurements of subjectively experienced stress and dissatisfaction (2 subscales representing strain and job satisfaction) in 9 companies are presented here. Compared to a control sample (the first 2 measurements in companies in which the implementation has not occurred), employees in the implementation sample had an increase in subjectively experienced stress and no changes in dissatisfaction shortly after the implementation. Both preimplementation values and changes in subjectively experienced stress and dissatisfaction seemed to be highly influenced by contextual factors at an organizational level. Regarding job characteristics (decision latitude) and external workload (e.g., family, children, and household responsibilities), employees with low decision latitude at their workplaces and a high external workload showed the strongest increases in subjectively experienced stress after the implementation. Preimplementation levels of subjectively experienced stress and dissatisfaction also seemed to be affected by decision latitude and external workload.
Implementation of Office Automation and its Effects on Job Characteristics and Strain in a District Court BIBA 425-442
  Eila Jarvenpaa
Using a longitudinal case study design, the implementation of office automation was examined at a district court office during a 4-year period. The aims of the study were to examine the effects of the implementation of office automation on perceived job characteristics and on the personnel's short-term mental strain, long-term mental strain, and job satisfaction. All occupational groups at the district court, that is, the office workers, the judges, and the court clerks, participated in the study. After the implementation, the office workers perceived their job as slightly more interesting than before the implementation. After the implementation, their short-term mental strain was slightly lower than before it. However, this effect seemed to diminish in the long run. On the other hand, a slight increase in the office workers' long-term strain was found. Moreover, the judges' job characteristics changed slightly.
   Using a longitudinal research design, the development of job stress among the personnel at the district court could be followed up, and the changes in job stress could be compared according to the phases of the implementation of office automation.
Use of Design Methods, Team Leaders' Goal Orientation, and Team Effectiveness: A Follow-Up Study in Software Development Projects BIBA 443-454
  Sabine Sonnentag; Michael Frese; Felix C. Brodbeck; Torsten Heinbokel
This article reports an empirical study on effectiveness of software development teams. It was predicted that both the extent to which design methods are used and team leaders' goal orientation is related to team effectiveness. Follow-up data (6-12 months after first assessment) were available from 25 German and Swiss software development projects. Hierarchical regression analysis showed that use of design methods and team leaders' goal orientation predicted team efficiency and changeability of the software product. In addition, goal orientation predicted maintaining schedule and budget. Practical implications of this study refer to the necessity to consider both the use of design methods and leadership issues when aiming at effective software development projects.