When asked about mental workload most people think about difficult tasks or conditions in which many tasks have to be performed simultaneously, or have to be performed in limited time. There is nothing wrong with this assertion, in particular the word ‘difficulty’ reflects mental workload very well. However, I will argue (as others have) that not only increased task demands, but also reduced operator capacity affect mental workload. The interaction between the task and the operator (driver) determine mental workload. So, not only increased task demands (e.g., the addition of a task, such as using a route navigation device), but also decreased driver capacity (e.g., as a result of experience or more temporary factors such as fatigue or illness) increase mental workload.

Techniques to measure mental workload are generally classified into three categories: performance (behavior), self reports and physiology. Measures from all three categories have been used in traffic research and an overview of use and usefulness in different experiments will be given. The measurement of task performance, self reports and measuring physiology will be illustrated by examples from experimental studies performed in an instrumented vehicle and studies performed in an advanced driving simulator. The primary task of car driving is to maintain longitudinal and lateral vehicle control. Primary task performance is reflected in the amount of ‘weaving’ and in headway control. Secondary tasks are sometimes added to assess mental workload, while self reports reflect the subjective experience of mental workload. In addition to this, measurement of the drivers’ physiology, in particular heart rate, can indicate when mental workload is relatively low or high. The use of the so-called heart rate profiles and special ‘applied’ measures, such as mirror looking behavior, will receive specific attention.