Walter W. Wierwille

Behavioral Measures of Aircrew Mental Workload

Behavioral research literature pertaining to the measurement of aircrew workload was classified into general categories of subjective opinion, spare mental capacity, and primary task metrics. Fourteen specific classes of workload measures related to these general categories were reviewed specifically in regard to aircrew workload assessment in the flight test and evaluation. Each class of measures was summarized in terms of background, applications, and implications for research and implementation. It was concluded that no one, single measure can be recommended as the definitive behavioral measure of mental workload. Due to the multidimensionality of workload, it appears that the most promising assessment procedure should include multiple measures of subjective opinions, spare mental capacity, and primary task measures as well as physiological correlates.

Evaluation of 16 measures of mental workload using a simulated flight task emphasizing mediational activity

As aircraft and other systems become more automated, a shift is occurring in human operator participation in these systems. This shift is away from manual control and toward activities that tap the higher mental functioning of human operators. Therefore, an experiment was performed in a moving-base flight simulator to assess mediational (cognitive) workload measurement. Specifically, 16 workload estimation techniques were evaluated as to their sensitivity and intrusion in a flight task emphasizing mediational behavior. Task loading, using navigation problems presented on a display, was treated as an independent variable, and workload-measure values were treated as dependent variables. Results indicate that two mediational task measures, two rating scale measures, time estimation, and two eye behavior measures were reliably sensitive to mediational loading. The time estimation measure did, however, intrude on mediational task performance. Several of the remaining measures were completely insensitive to mediational load.

A comparison of rating scale, secondary-task, physiological, and primary-task workload estimation techniques in a simulated flight task emphasizing communications load

Sixteen potential metrics of pilot mental workload were investigated regarding their sensitivity to communication load and their intrusion on primary-task performance. A moving-base flight simulator was used to present three cross-country flights. The flights varied only in the difficulty of the communications requirements. Rating scale measures were obtained immediately postflight; all others were taken over a 7-min segment of the flight task. The results indicated that both the Modified Cooper-Harper Scale and the workload Multi-descriptor Scale were sensitive to changes in communications load. The secondary-task measure of time estimation and the physiological measure of pupil diameter were also sensitive. As expected, those primary-task measures that were direct measures of communicative performance were also sensitive to load, whereas aircraft control primary-task measures were not, attesting to the task specificity of such measures. Finally, the intrusion analysis revealed no differential interference between workload measures.

On Solving Multifacility Location Problems using a Hyperboloid Approximation Procedure

Abstract An iterative solution method is presented for solving multifacility location problems involving rectilinear and/or Euclidean distances. The iterative procedure is based on the use of an approximating function involving hyperboloids, which in the limit approach the cones in the original objective function. Given that the hyperboloid approximation procedure converges, it is shown to converge to the optimum solution. Computational experience with the procedure is described.

The Influence of Motion and Audio Cues on Driver Performance in an Automobile Simulator

A highway driving simulator with a computer-generated visual display, physical motion cues of roll, yaw, and lateral translation, and velocity-dependent sound/vibration cues was used to investigate the influence of these cues on driver performance. Forty-eight student subjects were randomly allocated to six experimental groups. Each group of eight subjects experienced a unique combination of the motion and audio cues. The control group received a full simulation condition while each of the remaining five groups performed with certain combinations of motion and sound deleted. Each driver generated nine minutes of continuous data from which five performance measures were derived. Results indicate that the performance measures of yaw, lateral, and velocity deviation are significantly affected by the deletion of cues. In support of the hypothesis that driver performance is augmented by the addition of motion cues, statistically significant negative correlations were obtained between the number of motion cues present and the measures of yaw and lateral deviation. With respect to motion and audio cues, recommendations are made regarding simulator design criteria.

Evaluation of 20 Workload Measures Using a Psychomotor Task in a Moving-Base Aircraft Simulator

The sensitivity and intrusion of 20 pilot workload assessment techniques were compared using a psychomotor loading task in a three degree-of-freedom moving-base aircraft simulator. The primary task was an instrument landing system approach and landing, with measures taken between the outer and middle markers. Three levels of psychomotor load were obtained by combined manipulation of random wind-gust disturbance level and pitch stability. Two rating scale measures and one control movement measure demonstrated sensitivity to all levels of load. Additionally, one time estimation measure and one pulse rate measure demonstrated sensitivity to some levels of load. No intrusion was found. The results of this experiment indicate that the sensitivities of workload estimation techniques vary widely, and that only a few techniques appear sensitive to psychomotor load.

Physiological Measures of Aircrew Mental Workload

Physiological measures of aircrew mental workload were divided into fourteen specific classes. Each class was then summarized in terms of background, applications, and implications for research and implementation. It is concluded that several physiological measures appear promising, but that more research is needed to provide convincing evidence of viability. Physiological techniques can, however, be combined with other workload assessment techniques to provide a more complete understanding of the workload associated with given aircrew tasks.

Comparison of Five Mental Workload Assessment Procedures in a Moving-Base Driving Simulator

Five methods of measuring mental workload (secondary task performance, visual occlusion, cardiac arrhythmia, subjective opinion rating scales, and primary task performance) were compared for sensitivity to changes in operator loading. Each was used to differentiate among low, medium, and high levels of workload defined in terms of the application point of crosswind gusts in a driving task. The driving task was produced using an automobile driving simulator with a six-degree of freedom computer generated display, a four-degree of freedom physical motion system, and a four-channel sound system. Techniques of mental workload measurement that have shown promise in previous studies were used as a between-subjects factor, and subjects were presented with a within-subject factor of wind gust placement. Gusts at the front of the vehicle represented high workload levels, and gusts toward the center of the vehicle represented progressively lower levels of workload. The results showed significant differences among workload levels for subjective opinion scales and primary performance measures of lateral deviation, yaw deviation, and steering reversals. A relative sensitivity estimate of these would be, from highest to lowest sensitivity, steering reversals and yaw deviation, rating scales, and lateral deviation. The techniques of occlusion, cardiac arrhythmia, and secondary task performance yielded no significant workload effect.

On the measurement of pilot perceptual workload: a comparison of assessment techniques addressing sensitivity and intrusion issues

A flight simulator-based study was conducted to examine fourteen distinct mental workload estimation measures, including opinion, secondary task, physiological, and primary task measures. Both the relative sensitivity of the measures to changes in mental workload and the differential intrusion of the changes on primary task performance were assessed. The flight task was varied in difficulty by manipulation of the presentation rate and complexity of a hazard-perception task that required each of 48 licensed pilots to rely heavily on their perceptual abilities. Three rating scales (Modified Cooper-Harper, Multi-descriptor, and Workload-Compensation-Interference/Technical Effectiveness), two secondary task measures (time estimation and tapping regularity), one physiological measure (respiration frequency), and one primary task measure (danger-condition response time) were reliable indicants of workload changes. Recommendations for applying the workload measures are presented.

Driver steering reaction time to abrupt-onset crosswinds, as measured in a moving-base driving simulator.

A moving-base driving simulator was used in three experiments involving driver reaction time (RT) to simulated crosswind disturbances. Analyses were conducted on driver steering reaction time (RT) to the disturbances. Experiment 1 revealed that RT was significantly shorter when physical-motion cues were present. A second variable, vehicle yaw rate rise time, showed no effect. In Experiment 2, design parameters influencing aerodynamic behavior of a vehicle were adjusted. RT increased as the vehicle center of pressure (point of crosslvind application) moved rearward (rom the (ront axle. However, rearward movement of the center of pressure also produced less disturbance of the vehicle itself Changes in understeer and steering sensitivity yielded no significant effect. In Experiment 3, both uninitiated drivers and drivers with time on task were examined. Neither the first exposure to a step gust nor driving time up to 150 mirl caused significant changes in RT when perfurmance was compared with that of practiced, fresh drivers. Interexperiment comparisons using crosswind amplitude and shape as independent variables demonstrated that the amplitude and rise time of the crosswinds were critical determinants of steering RT.