Mental Workload Assessment

We all feel stressed out and strained when we have work to do. Not only that, we experience situations like this even if we are just studying. More often, we feel pressured just by thinking the amount of exams to be prepared for, or for that next project that is necessary for a good promotion in the company. Mental workload is the right term for the stress and strain we experience, especially with regards to studying and working.   The Hanover College defines mental workload as â€Å"the feeling of mental effort or the level of use of the human operators limited resources† (n.d.).   In short, mental workload is a demand placed upon humans (Xiaoli, n.d.). When there is too much mental workload, it might lead to errors. Preventing this makes mental workload important to be understood. However, due to the many factors that must be considered in discussing mental workload, defining it becomes difficult. Mental workload is important in driving and aviation and design. In fact, most of the studies conducted about mental workload were about driving and aviation and task demands. This is perhaps due to the fact that a driver is required to do not just one but many tasks. Moreover, even though a driver is experienced, accidents still occur. De Waard (1996) conducted a study on mental workload among drivers. He said that driving a car looks like a pretty simple task for everyone. Driving schools provide comprehensive lessons and manuals on how to drive safely. But no matter how good a driver can be, accidents cannot be avoided. Moreover, these accidents are attributed to human failure. Human failure is still increased due to several factors. First is the increasing number of vehicles on the road. There is a demand on the human information processing system, and also increase in the likelihood of vehicles colliding. Second, people drive well into old age. However, older people tend to suffer from problems in terms of divided attention performance. It all started with the car radio, and then car phones and other technological devices. The driver must divide his attention to all these systems besides controlling the vehicle. Lastly, those drivers in a diminished state may endanger him. Most of the time, drivers set out at night for the longer journeys to avoid traffic. Driving at night can cause him sleepiness and fatigue. Aside from this, the driver can also be intoxicated (de Waard, 1996). Xiaoli (n.d.) presented the factors which affect driver workload, including the following: fatigue, monotony, sedative drugs and alcohol. Environmental factors also affect drivers, such as traffic demands, automation and road environment demands. There are different techniques in assessing mental workload, including the following: performance measures, physiological measures, and subjective task measures (or self-report measures) (Luximon & Goonetilleke, 2001). Primary and secondary task measures comprise the performance, or system output measures. An overview of each assessment technique will be discussed in the context of traffic research (driving or aviation). Performance Measures In Xiaoli’s (n.d.) slide presentation, he said that the measures usually belonging to this category are speed of performance, number of errors made and reaction time measures. Outside the laboratory, these become task-specific. De Waard (1996) said that most of primary-task measures include speed or accuracy measures. Aside from this, De Waard (1996) explained that primary-task performance establishes the efficiency of man-machine interaction. Not just the primary-task performance but also other workload measures must work together so that valid conclusions can be drawn about man-machine interaction. There are several approaches in the measurement of performance measures. First is the analytical approach (Meshkati, Hancock, Rahimi and Dawes, 1995). According to Welford (1978, cited in Meshkati, Hancock, Rahimi and Dawes, 1995), the analytical approach considers the detail at the actual performance of the task that will be assessed. Not only the overall achievement is examined but also the manner in which it is achieved. Another assessment technique is the synthetic methods. These start with a task analysis of the system. Task analytic procedures are then used to identify the specific performance demands placed on the operator. The third approach is the multiple measurement of primary task performance. This approach is very useful when individual measures of primary task performance do not show enough sensitivity to operator workload. On the other hand, Xiaoli (n.d.) indicated that secondary-task performance are about factors such as time estimation or time-interval production and memory-search tasks. The assumption associated with secondary task measure says that an upper limit exists on the ability of a human operator to gather and process information (Meshkati, Hancock, Rahimi & Dawes, 1995). The way to measure secondary-task performance is through another task included to the primary one. De Waard (1996) mentioned about the multiple-resource theory. The theory says that â€Å"the largest sensitivity in secondary-task measures can be achieved if the overlap in resources is high† (De Waard, 1996). According to Hancock, Vercruyssen and Rodenburg (1992), a person must have the ability to synchronize their actions with the dynamics of differing environmental demands so that he can survive and prosper in uncertain conditions. This means that the person must have some degree of autonomy with respect to space and time. However, secondary-task measures have disadvantages to consider. According to De Waard (1996), time sharing is not very efficient if the same resources are utilized. Moreover, additional instrumentation is required in secondary-task measures. Not only that, but there is lack of operator acceptance. There are also possible compromises to system safety. Subjective Task Measures There is much talk about the self-report measures, which is also called subjective measures. In fact, for De Waard (1996), self-report measures are advantageous because they can better show the real meaning of mental workload. These measures’ subjectivity is what makes self-report measures strong. Muckler and Seven (1992, as cited in De Waard, 1996) explained that self-report measures are strong because the awareness of the operator about the increasing effort used must give subjective measures an important role to play. Moreover, performance and effort are incorporated in self-report measures. Additionally, individual differences, operator state and attitude are also considered. Xiaoli (n.d.) said that the primary advantages of subjective task measures are high face validity, ease of application and low costs. However, there are also limitations in these measures. First is that there might be confusion of mental and physical load in rating. There might also be an exhibition of the operator’s inability to differentiate between external demands and actual effort or workload experienced. Second, limitations can be seen in the operator’s ability to introspect and rate expenditure correctly. Hancock, Brill, Mouloua and Gilson (2002) added that another disadvantage of self-report measures is that they cannot be used for online workload assessment. Physiological Measures According to De Waard (1996), physiological measures showed sensitivity to global arousal or activation level and in some stages in information processing. One advantage of this is that physiological responses do not need an obvious response by the operator. Additionally, most cognitive tasks do not need overt behavior. Moreover, some of the measures can be collected continuously. Kramer (1991, cited in De Waard, 1996) showed some of the disadvantages of these measures. First is that there must be specialized equipment and technical expertise to be able to utilize these measures. Second is the presence of signal-to-noise ratios. Kramer furthered that in operator-system performance, the operator’s physiology is not directly involved, unlike in primary-task performance. Other physiological measures involved in driving are pupil diameter, endogenous eye blinks, blood pressure, respiration, electrodermal activity, hormone levels, event related potentials, and electromyogram. De Waard (1996) furthered that not all measures are sensitive to workload when it comes to performance. There are instances when dissociation between these measures of different categories was reported. He said that dissociation occurs between measures when they do not correspond to changes in the workload, or if there is an increase in one measure and a decrease in another. Performance is thus affected by the amount of resources invested and the demands on working memory. Hancock, Brill, Mouloua and Gilson (2002) said that although physiological measures present global assessments of workload, they do little to balance the demands of tasks on sensory systems. In addition, physiological measures provide little or no information about what sensory systems are most taxed. To measure mental workload, two groups must be considered (Gopher & Donchin, 1986, cited in De Waard, 1996). Self-report measures, physiological measures and performance measures are included in the first group. This group supposes that it is probable to achieve a global measure of mental workload. The second group includes secondary-task measures and some of the physiological measures. This group is concerned about those diagnostic procedures and has something to do with the theories of multiple resources. References De Waard, Dick. (1996). The measurement of drivers’ mental workload. The Netherlands: The Traffic Research Center VSC. Hancock, P.A., Brill, J.C., Mouloua, M., & Gilson, R.D. (2002). M-SWAP: On-line workload assessment in aviation. Paper presented at the 12th International Symposium on Aviation Psychology. Dayton, OH. Hancock, P.A., Vercruyssen, M., & Rodenburg, G.J. (1992). The effect of gender and time-of-day on time perception and mental workload. Current Psychology: Research and Review,. 11, 203-225. Hanover College. (n.d.). Mental Workload. Retrieved October 27, 2007 from http://psych.hanover.edu/classes/hfnotes3/tsld022.html Luximon, A. & Goonetilleke, R. (2001). Simplified subjective workload assessment technique. Ergonometrics, 44, 229-243. Meshkati, N., Hancock, P.A., Rahimi, M., & Dawes, S.M. (1995). Techniques of mental workload assessment. In J. Wilson and E.N. Corlett, (Eds.). Evaluation of human work: A practical ergonomics methodology. (Second Edition), London: Taylor and Francis. Xiaoli, Yi. (n.d.). Measurements of mental workload. [Slide presentation]. Available on http://www.slideshare.net/ESS/measurement-of-mental-workload/