Proceedings of the Human Factors and Ergonomics Society Annual Meeting

Recent research suggests that errors made early in a software development project and carried on into testing and integration are the most costly type of errors to find and correct. Yet, there is almost a total absence of research examining the impact of tools and methodologies early in the process, such as in program design.

One approach to improving the design process has been the use of program design methodologies, which provide strategies to programmers for structuring solutions to computer problems. The basic difference among methodologies is the criterion used to decompose the problem into smaller units. The approaches basically vary along one dimension: the extent to which the decomposition relies upon data structures as an organizing principle for modularization. On one end of the dimension are data structure techniques that rely primarily on the data structures present in the specifications as the basis for modularization. On the other end of the dimension are techniques that rely primarily on operations as the basis for structuring the problem. In the former case, modules are organized around data structures, while in the latter, modules are organized around operations. Falling between the two extremes are techniques which rely partially on data structures and partially on operations as the basis for structuring the programs.

Using this dimension to classify methodologies, it was possible to generate programs decomposed in each of these ways, and to evaluate the effects of these decompositions in terms of the initial coding process, the quality of the resulting code, and the subsequent maintainability of the program. The focus of the research was on a comprehensive evaluation of programs produced by the different classes of methodologies.

Professional programmers were provided with the specifications for each of three problems and asked to produce pseudo-code for each specification. Each time the programmers worked on the program, they were asked to complete a summary sheet for the session. The intermediate versions of the programs and these summary sheets were collected for analysis. In addition, the participants were asked to complete a final questionnaire at the end of the project which provided us with information about each programmer's programming background, familiarity with the methodology, and reactions to the problems used in this research.

The measures collected were the time to design, number and types of design errors, time to code, number and types of coding errors, number and types of commands used to create the programs, amount of computer time needed to execute the program, complexity (as measured by several metrics including, for example, the McCabe metric), and consistency between different programmer's solutions.

The preliminary results suggest that there were differences in time to code, number of design errors detected, and consistency and complexity ratings among methodologies. These differences will be discussed in light of their impact on the comprehensibility, reliability, and maintainability of the programs produced.

This is, in part, a preview of the program for the International Conference on Building Use and Safety Technology to be held November 7–9, 1984 in Washington, D.C. Background information is also provided on current developments in the major U.S. model codes and safety standards organizations where there are discussions on applications of research, with a human factors orientation, especially in relation to circulation safety and egress concepts. Relative to what was the case only a few years ago such discussions are now fairly intensive. They are leading to revisions which can be regarded as revolutionary. New research agendas, opportunities and sponsoring arrangements are being considered in the same contexts.

The human response to the thermal environment depends in part on the temporal nature of the exposure. One aspect of the temporal dimension includes ramps or drifts that are brought about when the heating or cooling system is turned off. When this occurs, the shape of the drift depends upon the outside temperature and wind velocity, the solar load; the building shell, the location of the occupant and the indoor thermal load including the occupant density and machinery. The Honeywell Corporation in Arlington, Illinois has developed a model that will predict the temperature drifts at various locations within the occupied space when the HVAC system is shutdown. This regulation of the office temperature before workers leave for the day represents an energy saving strategy. The purpose of this study was to examine the comfort response of humans when exposed to drifts that began 0.5 h and 1.0 h before the normal departure of the employees. Drifts were examined at 4 locations: interior, west side and north side; a control “no-drift” condition was also used. The results showed for the 1.0 h drift that no source of variance was significant for the Thermal Comfort votes. Location, however was significant for the Thermal Sensation votes. For the 0.5 h drift the only significant source of variance was the location x sex interaction for the Thermal Comfort votes. A discussion of the implication of the results as an energy conservation strategy is provided.

Folkard, Monk and Lobban (1979) proposed a questionnaire which offered promise to help distinguish between personnel who differ in the degree to which their circadian rhythms adjust to shift work schedules. Fourteen male and ten female subjects completed two versions of the questionnaire. A period of irregular shift work was interposed between the two administrations. Significant differences were found in the way the questions were answered on the two administrations. Factor analysis products were not consistent for the two administrations nor with the original scale.

During a 72 h field test, two groups of elite, “ranger type” soldiers were evaluted five times (during the first 8 h, after 30 h of continuous field duty, after a rest period, after 30 more h of continuous field duty, and after a second rest). One group (7 soldiers) received two rests of 3 h each. The second group (6 soldiers) received two rests of 6 h each. Five military tasks (map reading, decoding messages, vehicle/aircraft recognition, assembly and disassembly of the M-16 rifle, and preparation of an AN/PRC radio for operation) were used. These tasks were considered by the commander to be representative of duties that the soldiers would do and were considered to be completely familiar to the soldiers. In addition, a peg into pegboard task and a subjective evaluation of fatigue questionnaire were given.

There was no significant effect of time or difference between the two groups. The soldiers completed these tasks of short duration (less than 3 min) without decreasing performance—even after extended periods of sleep loss. These performance results occurred even though subjectively the soldiers reported they were tired.

Typically, human performance is projected to fluctuate in a circadian manner, usually in association with body temperature rhythm. Such sinusoidal relationships are isomorphic in that they are, under most normal circumstances, phase-locked in environmental time. As body temperature rises so performance is improved. However, increase in temperature above the zenith does not result in increased capability. Rather, such elevation has the dual and opposing action of speeding performance while elevating error rate to produce an overall diminution of efficiency (e.g., Allan and Gibson, 1979; Nunneley, Reader, and Maldonado, 1982). It has been suggested that deep body temperature controls the speed of performance, while skin temperature mediates error through the control of comfort associated with peripheral variation. To overcome distress connected with peripheral increase, yet to utilize the superior speed elicited by alteration of central temperature change, the use of selective head heating has been explored. The present paper synthesizes data from selective heating studies and indicates the potential gains which may be realized in complex machine operation. Results from this paradigm have suggested facilitation in stimulus assimilation, decision making capability and response selection and execution (e.g., Hancock, 1983; Hancock and Dirkin, 1982; Holt and Brainard, 1976). This facilitation is not linear in temperature change. To a 1°C increase there appears a change in decision criterion such that individuals become more conservative in response. Above this threshold value, veridical gains in performance are manifest with a reduction in both task latency and error rate. Simple physiological accounts of such behavior are largely inadequate as the precise physiological effect has yet to be thoroughly documented. An initial outgrowth of this work is the potential usage of fluctuation in head temperature as a dynamic physiological reflection of imposed mental workload (e.g., Hancock, Meshkati, and Robertson, 1984). These performance facilitations under relatively slow temperature changes are contrasted with decrement observed under rapid temperature fluctuation. A potential relationship with the notion of attentional resource capacity is explored.

The field study reported here investigated the various ways in which women are handicapped and constrained in their behavior, or socially discomforted by the design of common objects and places. The survey showed that most “discrimination by design” was due to failures of ergonomics, environmental interaction effects with feminine dress and biology, or being placed on unwanted display. Women evidenced a remarkable variety of ingenious and casual adaptations to these circumstances — most of which are avoidable through sensitive human factors design. The ubiquity of objects and settings that place women at a disadvantage and the relative lack of awareness of these effects suggest that such conditions will persist for a long time.

Several studies have utilized the psychophysical approach to generate lifting capability data for industrial workers. In all these studies, certain assumptions were made. These assumptions, as reported recently, are not valid and, therefore, lifting capability data based on the psychophysical approach must be subjected to certain adjustments. This paper reports new data on lifting capability of the industrial workforce and compares it with two previous studies. Data from all the three studies were adjusted for time duration and, since the three studies compared favorably, integrated to develop comprehensive lifting capability data bases. The effects of lifting frequency, box size, height of lift, and age of workers are discussed.

The determination of the maximum acceptable weight of lifting using the psychophysical method was examined over a four hour period. Six females lifted a standard milk carton case from the floor to a height of 0.79 m. The subjects chose the amounts of liquid acceptable to them by instructing a helper to increase or decrease the weight of the case. The amount of liquid acceptable to subjects at 4 lifts/min after four hours of lifting was not different from the weight estimated by them after 20 minutes. However, at 8 lifts/min the acceptable amount of liquid decreased with time. These results suggest that a 20 minute period may be sufficient to estimate the maximum acceptable weight for the frequency of 4 lifts/min, but not for the frequency of 8 lifts/min.

The primary objective of the present study was to explore whether the human machine has an optimal value or a range where it performs at maximum physiological efficiency during repetitive lifting tasks. Ten trained male subjects participated in this study. The frequency of lift and height of lift were the independent variables. The psychophysical methodology was employed in this investigation. Maximum physiological efficiency is defined as the power output per unit oxygen consumption. The results indicated that for maximum physiological efficiency in repetitive lifting tasks of a duration of 1/2 hour the frequency of lift should be between 5 and 11 lifts/min with 9 lifts/min being optimal.

A study was conducted to compare dynamic lifting task with a simulated static lifting task. 17 subjects 12 males and 5 females volunteered to participate in this study. The independent variables were the type of task (static and dynamic), frequency of lift (2, 4, and 6 lifts/min), and muscle group (back lift, leg lift, and arm lift). The psychophysical methodology was used in this study. The results showed highly significant differences between maximum acceptable static effort (MASE) and maximum acceptable dynamic effort (MADE) for both males and females. The frequency was also statistically significant. The muscle groups were significant for females and not significant for males. On the average MADE values were 45% of those of MASE values for females and 38% for males. The ratio of female to male data, on the average, were 39% for MASE values and 44% for MADE values. Regression models for the prediction of maximum of MADE from MASE and other task and worker variables were developed.

Thirty industrial subjects took part in a manual lifting task, using different handle positions on a container and different angles between handle and container. Lifts were from floor to waist, waist to shoulder and floor to shoulder. Upper extremity body angles were measured, with heart rate and rated perceived exertion. As in previous static holding experiments, it was found that handle positions with both horizontal and vertical stability gave good results. As a result of this work, handle positions are recommended in the middle of the front edge of a box (at 60°) and in the middle of the lower edge (at 50°). Such an arrangement will minimize wrist deviation and slippage angle between handle and hand.

The historical development of back compression and muscle strength modeling is reviewed. A microcomputer version of these models was developed. Examples are given to demonstrate the applicability of these models in the analysis and design of railroad jobs.

Memorization has been an often-used technique for testing cognitive construct organization by programmers. This study looks at three different memorization techniques and introduces a new method for examination of syntax and logical constructs.

Subjects had to discover the structure of a logical network whose links were invisible. The behaviour of the network components after a failure gave clues to network structure. Since similar failure diagnosis tasks often draw on spatial processes, a good deal of spatial complexity in the network should affect ease of discovery. Results supported this conjecture and showed that increasing the amount of evidence available to subjects about the existence of complex links was of little benefit. These results raise the question of whether inferences about spatially complex pathways were simply not made, or whether they were made but not retained because of a high load on memory resources.

Four groups of subjects (N = 8 in each group) were presented training, based on the cognitive aspects of a basic flight maneuver, at different times during the acquisition of the motor skills needed to perform that maneuver. Results indicate that the performance of the group that received extensive cognitive training prior to attempting the associated motor skills was significantly better (p<.05) than the groups that received the same cognitive training interspersed during the initial learning of the motor skills. This was true even when the interspersed training was preceded by extensive cognitive pretraining. Results also show that cognitive training interspersed with motor skill learning produced a deterioration in performance as the performance of the two groups receiving the training was below that of the control group which received no cognitive training whatsoever. Transfer of training to a similar maneuver indicated that the groups that were initially trained using extensive cognitive pretraining performed significantly better than the groups that received only interspersed cognitive training and the control group.

Fitts' law has been universally cited as an index of difficulty or predictor of movement time (MT) for rapid aiming tasks since it was first published in 1954 (Fitts 1954). Many researchers report a remarkable correlation of Fitts' law and the observed movement times in aiming tasks. Other researchers report discrepancies, however, between observed movement time and the law, especially at low and high movement times, which correspond, respectively, to short movements to a large target, and long movements to a small target. These discrepancies suggest that while the law predicts MT well for some human motions, the true basis for the law may not be known, and, as a consequence, that there may exist conditions where its application is appropriate and yet others where different laws should be used.

Fitts suggested the law as a model of the rate-limit of human information processing and movement. But this rate-limit model is not the only interpretation possible. Rapid movement of the hand to a target can be modeled from a different view-point: namely, as a control system. This paper gives the analytical results for several models: a first order model where it is assumed that the hand velocity can be directly controlled, and a second order model where it is assumed that the hand acceleration can be directly controlled. The results show that the first and second order models with the linear control-law produce a MT function with the exact form of the Fitts' law, and yet it has nothing to do with information theory.

Training technology has traditionally addressed issues that have been limited to the acquisition of knowledge at skill levels far below those of highly skilled performers. Knowledge about the changes that occur as a result of extensive experience are potentially useful in understanding highly proficient performance. Empirical evidence suggests that experts differ from novices in the cues used to achieve a target performance. It appears that recognitional capacities of experts differ from those of novices. A Reperatory Grid Technique was the means used to investigate differences in recognitional capacities of expert and novice software specialists. Task dimensions of a variety of algorithms for the problem of critical path analysis were elicited from subject matter experts by means of a matching technique. Expert and novice computer programmers then used the empirically determined dimensions to rank nine algorithms.

Discriminant analyses performed on the data indicated successful separation of the novice and expert groups on the majority of dimensions used. Furthermore, the analyses indicated which individuals were correctly classified as experts or novices solely on the basis of scores on the dimensions. The results of this study provide support for the hypothesis that as expertise is achieved, the perceptual cues used by the proficient performer differ from those of novices.

Assistance systems are beginning to become necessary components in human-computer systems. The present research describes an initial attempt at providing a performance-based assistant for a complicated information retrieval task. The assistant was implemented based upon an expert profile methodology. The resultant expert performance model was used to diagnose inappropriate novice search strategies. A suggestive and forceful assistant were compared against an unassisted novice subject group. The suggestive assistant highlighted appropriate search procedures for novice subjects on a keypad with a touch entry display. The forceful assistant also highlighted appropriate search procedures but eliminated inappropriate search procedures from the touch keypad. The experiment demonstrated a marked improvement in the search strategies and performance of assisted novices. These enhanced search styles were not without a cost. In this implementation, both the suggestive and forceful assistants were very intrusive on trials with advice. Consequently, further research should address the development of assistance systems that naturally transfer expertise to inexperienced users.

The performances of thirty first-time users of an online library catalog system were studied. The subjects were asked to conduct five online searches for documents available in the library and to think out loud as they did so. An analysis of the errors made and the associated verbal protocols provided a number of insights into the underlying mental processes.

A critical component of computer interface design is the specification of a file management system. Although much is known about how humans organize and classify information and objects, there have been few investigations of the performance consequences of alternative computer filing systems. This paper presents the results of an experimental study of how the organizational capabilities of interactive computer filing systems affect the ability of users to manage sets of information. This study compared the use of flat versus hierarchical, and keyword versus non-keyword filing systems. Each of 16 subjects performed a two-part filing and retrieval task using one of four filing system configurations. Subjects first organized and arranged a set of 50 computer files according to the constraints and capabilities of one of the four filing systems. During retrieval the subjects were required to locate certain files and perform common computer operations on them. Task completion time is used as a measure of ease of filing and retrieval for each system. Strategies of file retrieval as a function of the filing system are also discussed.

Medical rehabilitation is a rapidly growing field in which the discipline of human factors can make substantial contributions. Some of the major problem areas include the development of a taxonomy of activities of daily living, the design and evaluation of orthotic and prosthetic devices including environmental control systems, and the structuring of effective rehabilitation training programs. Specialists in human factors should recognize the impact of their potential assistance in medical rehabilitation and should be encouraged to meet this challenge.