Proceedings of the Human Factors and Ergonomics Society Annual Meeting

Many vigilance tasks require discrimination of infrequent signal events from frequent non-signals. During performance on these tasks, signal detections often decline. But this does not generally signify a loss of vigilance - if one can rely on signal detection analyses showing that signal discriminability remains constant during a vigil, and hence that neither attention nor signal processing have waned. The paper confirms that signal detection theory does provide a good fit to vigilance data, and that the analyses can therefore be relied on. The paper also shows that probability matching (of signal reports to signal occurrence) occurs. In the main, it is the adjustment of report rate towards matching by an alert observer, in control of his performance, that produces the vigilance decrement in detections.

The present paper discusses vigilance phenomena in terms of the automatic/controlled processing theory(Schneider & Shiffrin, 1977) of attention and human information processing. Fisk & Schneider (1981) interpreted vigilance decrements within the automatic/controlled process framework and discussed system design implications as well as training methods for vigilance environments. New results, concerning the effects of various stressors on automatic and controlled processes, are presented to highlight why, when theorizing about vigilance, it is more important to think about mode of processing than task complexity, stage of processing, etc.

Several experiments are described showing that the vigilance decrement can be reversed or retarded by a suitable level of complexity in symbolic functions needed for critical signal detection. This effect is most likely based upon motivational rather than learning factors and is dependent upon the selection of a complex task that is no more capacity demanding than a simple one.

The objective of this panel session is to discuss the roles of human factors engineering with respect to the area of gerontology. The focus will be on two elderly subpopulations, the well and active elderly and the debilitated, sick elderly.

With respect to the well elderly, the discussion will center around the increasing numbers of elderly people in the labor force and how human factors engineers need to provide for this population in the design of jobs, equipment and work environments. General trends in employment patterns for the elderly population will be discussed and statistics will be given on the number of elderly in the population, the types of roles the elderly are assuming and where the elderly will be employed. Since the use of computer technologies is pervasive across occupational settings, there will be an emphasis on aging and the use of computers. This part of the discussion will focus on software design, equipment design and training.

This symposium brings together several researchers studying different aspects of human-computer interaction. All of the researchers share the assumption that it is important to understand the cognitive processes involved in human-computer situations. Each presentation focuses on a unique aspect of the cognitive component of human-computer situations, however.

Human-computer interaction is a growing area that includes many human factors issues. Physical factors in human-computer situations are common across the range tasks for which computers are used and are similar to traditional ergonomic considerations in other contexts. When the focus is on human performance in a particular task, however, the important issues become the design of the software interface and the cognitive components involved in manipulating the interface and understanding the system behind it.

We hope that this set of talks will focus interest on “cognitive human factors” and provide insight into which components of the cognitive system are relevant to particular kinds of tasks and which methodologies might be appropriate in different research contexts. Thus, the tasks studied and methodologies discussed are diverse. Two studies focus on procedural skills (text editing and image processing). The researchers in these cases describe plans and strategies and rely on naturalistic observation and formal simulation to characterize the observed behavior. Another study focuses on database manipulation and utilizes a more traditional experimental methodology. The researcher describes the characteristics of different memory representations resulting from different database interfaces and shows the importance of “interface-representation compatability.” Yet another study concerns the task of software development. The researcher examines problem solving strategies and the interaction between development methodology and the programming problem to be solved. Finally, one presentation explicitly explores the use of a particular theoretical orientation and research tool, signal detection theory, in the general context of human-computer interaction.

The foregoing papers are joined, yet divided, by their approaches to the concept of interface, and express those differences in the choice of technique of measurement and the emphasis on distinct cognitive processes. The interface definitions include: (a) the characteristics at the machine surface and this is demonstrated in the Bennett study on accessing data bases and Boebm-Davis' work with programming methodologies; (b) the medium or task environment between the operator and the machine such as Parasuraman's focus on signal detection models; (c) a software module that is adaptable or responsive to the operator that is introduced by Howard et al. in the image processing study; and (d) the processing elements embedded within the operator as is characterized in Robertson's research on text-editing. It is evident that each of these approaches is a facet of human-computer interaction and integrative models are required before adequate understanding of human performance on this task can be realized. Computer architecture has been highly influential in suggesting structural elements that underlie the information-processing models which are often invoked to explain the results of human-computer interaction studies. It would appear that those models have run their course. Another perspective, derived from software programming concepts, has had a fresh impact on the interpretation of human-computer performance; in the main, those concepts that have been adopted are related to applications software and these notions are present in some of the foregoing reports. The next step in programming concepts is the emergence of modules of user-interface systems that will link hardware and application-software systems. At least one theorist (J.A. Fodor) has developed an argument that that suggests the characteristics of the human-processing elements that would comprise such a module; however, this development is not clearly anticipated in these papers.

Computer graphics are becoming an increasingly important component of the human-computer interface. Computer graphics can be effectively used to present many types of information. However, the psychological processes underlying visual imaging and reasoning are not well understood, especially in a systems context. This paper is concerned with how individuals mentally represent and reason about complex systems. Relevant basic and applied research is briefly reviewed and implications for human-computer interaction are discussed.

This presentation explores how people learn from mistakes that they make when learning a computer text editor. The development of complex procedural skills is characterized by the frequent appearance of errors. Learners gain information about the characteristics of a system as they make errors and plan how to recover from error states. This project examines the processes of error recognition, explanation, and recovery planning–and how these processes influence the development of plans for using an interactive system.

This position paper is based on research concerned with the usefulness of various program design methodologies in producing maintainable software. The position is taken that these methodologies will produce maintainable code only when the strategies they provide to programmers for dividing the problem up into more manageable parts are consistent with their internal representation of the problem.

Several related experiments are being conducted to investigate human performance in an image processing and interpretation task. In this task, users select operators from a set of image enhancement transformations in order to modify an image and facilitate interpretation. In several trials, subjects were required to identify images of ship hulls as belonging to one of four categories. Operator sequences, selection times, and categorization responses were recorded. Subjects adopted strategies that lead to the enhancement of spatial frequencies important for the categorization task. The strategies were easily identified, emerged in an orderly manner, and changed in predictable ways in response to changes in the frequency content of the task images. We are currently developing a production-based model to account for the operator sequences and selection times.

The use of signal detection theory (SDT) in evaluating attentional and decision-making performance in the context of human-computer interaction is reviewed. SDT provides a means for distinguishing between accuracy and criterion setting in decision-making environments. This is useful in evaluating the effectiveness of the decision-making performance of an intelligent machine, a human user, or a human-machine system. SDT is also useful in deciding how best to allocate subtasks and functions in human-computer monitoring systems. Applications of SDT to the evaluation of rule bases in expert systems and the design of computer assistance in human-computer monitoring are discussed.

A twenty week program of one hour a week training sessions for in-house personnel is nearing completion. The ourpose is to inculcate recent graduates with programmatic skills and to upgrade the skills of older human factors personnel. The course contains materials in three general areas, namely: 1) An overview of human factors functions at Lockheed, 2) the specific process and some examples, and 3) a description of other technologies and their relation to human factors.

The discussion focuses on the development of one area of human factors/ergonomics education in the International Business Machines Corporation. This chronology, detailing the inception, evolution and current status of an IBM ergonomics education program, should provide valuable insights into the rationale behind the growing success of occupational ergonomics in IBM.

As a result of a rapid increase in fire fatalities among residents in board and care homes, the National Bureau of Standards developed an Evacuation Difficulty Index for use in determining how much building fire protection should be required in individual board and care homes. The Index is a ratio, where the numerator is a sum of the assistance needs of residents, and the denominator is the availability of staff. The assistance needs of residents are rated on behaviorally-anchored scales that represent “risk factors.” The risk factors were identified by an expert panel's informal analysis of everyday assistance needs that must also be provided during fire emergencies, an approach adapted from research on synthetic job validation. The panel also helped determine scale values for risk factors. Three studies of inter-rater reliability led to further revisions, and the system was field tested in 151 facilities.

What is legal and not in violation of any building code or ordinance is not sufficient to guarantee safety of the humans involved in many cases of public use. The objects and places that people encounter and use in everyday life (e.g., stairways, elevators, and shopping carts) need to be consistently designed and manufactured. Consistency in this vein will allow for a positive transfer of training and thereby allow for the human filtering function to process the stimuli received from the objects and places. An annotated discussion of several examples is provided.

The FAA is engaged in a major effort to safely reduce the vertical separation standard to 1,000 feet between aircraft flying above 29,000 feet and to develop system performance requirements which, if met, will ensure safe operation of the National Airspace System. To accomplish this we are proposing an overall program structure of defining risk/safety criteria, a risk assessment program, and a risk management program. Included in this effort is an extensive in situ data collection effort. Hopefully, the conclusion of this effort will result in providing to management the quantitative decision tools and information necessary to (1) make the decision for the safe reduction of the vertical separation standard and (2) to implement changes to the National Airspace System to support new separation procedures.

The U.S. Nuclear Regulatory Commission and Sandia National Laboratories sponsored a project to evaluate psychological scaling techniques for use in qenerating estimates of human error probabilities. The project evaluated two techniques: direct numerical estimation and paired comparisons. Expert estimates were found to be consistent across and within judaes. Converaent validity was good, in comparison to estimates in a handbook of human reliability. Predictive validity could not be established because of the lack of actual relative frequencies of error (which will be a difficulty inherent in validation of any procedure used to estimate HEPs). Application of expert estimates in probabilistic risk assessment and in human factors is discussed.

A current trend in cockpit design is to incorporate synthesized speech to present secondary information. Multiple resource theories of information processing support this, but theories of stimulus-central processing-response compatibility suggest that spatial information presented visually may have some advantages over speech. An experiment was run comparing tracking and response performance when information was presented pictorially and by speech. Pictorial subjects responded quicker and improved more with learning than did speech subjects. More research on the spatial advantages of pictorial displays is needed before too many speech displays are incorporated into the cockpit.

The possibility of reducing task interference in complex aviation environments by taking advantage of the verbal-spatial short term memory dichotomy is explored in a dual-task paradigm. Eighteen subjects performed verbal and spatial retention memory tasks concurrently with intervening verbal and spatial cognitive tasks. Both number and processing code of the intervening tasks were manipulated. Support was found for code-specific interference such that concurrently performed tasks of the same code disrupted performance more than concurrent tasks of different codes. In addition spatial memory was found to be more fragile than verbal memory. Implications of the findings to mental workload reduction include task scheduling, presentation format, and assignment.

Subjective assessments of workload are becoming increasingly important in the evaluation of new systems. Two popular methods were compared in the present investigation: (1) the Subjective Workload Assessment Technique (SWAT) which was developed around the use of conjoint analysis to create interval scales, and (2) a technique under development at NASA that utilizes an individually weighted workload score from a set of nine bipolar ratings. Both methods were applied in a laboratory experiment that required rating a number of single- and dual-task trials of tracking and/or a spatial transformation task. The dual transformation-tracking task results were reviewed. The results for the two assessment techniques were remarkably similar, indicating that the subjective experience of workload is sufficiently robust to be resistant to variations in the measuring technique. Also, both subjective assessment techniques were successful in measuring the differences in task difficulty as indicated by a multivariate analysis of performance. Finally, the specific strengths and weaknesses of each assessment technique were reviewed.

The desirability of employing speech response in a dynamic dual task situation was discussed from a multiple resource perspective. A secondary task technique was employed to examine the time-sharing performance of five dual tasks with various degrees of resource overlap according to the structure-specific resource model (Wickens, 1980). The primary task was a visual/manual tracking task which required spatial processing. The secondary task was either another tracking task or a spatial transformation task with one of four input (visual or auditory) and output (manual or speech) configurations. The results show that the dual task performance was best when the primary tracking task was paired with the visual/speech transformation task. This finding was explained by an interaction of the S-C-R compatibility of the transformation task and the degree of resource competition between the time-shared tasks. Implications on the utility of speech response were discussed.

The effects of the complexity of stimuli on reaction times in double stimulation tasks were investigated within the concept of cognitive capacity allocation. The results showed that the complexity of the first stimulus (S1), and the complexity of the second stimulus (S2), and their interaction terms had significant effects on reaction time to S1 (RT1), and reaction time to S2 (RT2). An increase in RT2 (and RT1) was observed when the complexity of S1 and/or the complexity of S2 increased. However, the increase in RT2 due to the increase in the complexity of S1 was greater than the increase in RT1 due to the increase in the complexity of S2.

Voice radio traffic was collected at National Training Center exercises in Ft. Irwin, CA. The traffic consisted of five one-hour samples covering peak-period traffic on several command nets. This traffic underwent a detailed exploratory analysis and identified several human performance issues which will be subjected to further evaluation in a follow-on effort. Sample issues covered in this effort include expected wait time, transmission rates, station utilization factor, message categorization, interference, and dialogue patterns.

In this paper I argue that a human factors specialist should supervise the translation of operating instructions. Pitfalls to avoid in translating text and a procedure (which I call the “Advocacy Method”) for generating reliable translations are presented.

An experiment utilizing 28 participants was conducted to evaluate two different user interfaces for creating and building lists of phone numbers using only a Touch Tone phone. Two different methods of use were compared and evaluated. Predefined tasks were given to the participants, and individual keystrokes, incorrect responses, and “inefficient” responses were recorded. Questionnaires were administered to obtain user opinion for assisting in making marketing decisions. The analysis of the performance data identified a number of problem areas. System-initiated time outs and a lack of user control resulted in frequent errors and confusion. The method of accessing the system and the terminology employed throughout were found to be obstacles to ease of use. To remedy these problems a third method of use was developed and its benefits are presented in this paper.