Physiological Measures of Mental Workload: Evidence from Empirical Studies

Abstract

Mental workload (MWL) is widely used in the design and evaluation of complex human–machine systems and can be measured by a number of physiological measures. However, the effectiveness of these measures seems unknown. This study was conducted to provide a comprehensive understanding of the effectiveness of physiological measures of MWL. Four electronic databases were systematically searched for empirical studies measuring MWL with physiological measures. Ninety-four studies were included for analysis. We identified 36 physiological measures and grouped them into electrocardiogram, eye movement, electroencephalogram, respiration, electromyogram, and skin measures. Thirty-three measures were reported to have significant associations with MWL, but their effectiveness varied. We also identified 11 physiological measures that were widely used and demonstrated high effectiveness in assessing MWL. However, their effectiveness did not remain consistent across different application domains. Our study offers insights into the understanding and selection of appropriate physiological measures to evaluate MWL in varied human–machine systems.