Differentiating Active and Passive Fatigue States with the Use of Electroencephalograph

With advances in automation technology, it is becoming essential to understand how automation affects human operators. A concern for the implementation of automation technology is the interactive effects it has with operator cognitive fatigue. Desmond and Hancock (2001) proposed that two types of fatigue can arise depending on the nature of the task: active and passive. Active fatigue results when operators must make constant perceptual-motor adjustments during high task demands, while passive fatigue results from operators executing little or no perceptual-motor adjustments during low task demands, similar to when automation is employed. The purpose of this study was to use electroencephalographic (EEG) indices of workload, engagement, and a candidate marker of strain under fatigue in conjunction with performance and subjective measures to differentiate active and passive fatigue states. Participants (N = 84) performed a generalized flight simulator for 62 min either under active, passive, or control conditions. Passive fatigue was characterized by reduced EEG engagement and initially elevated and stable ratios of Fz theta to POz alpha power compared to active fatigue. Subjective measure results indicated that passive fatigue was characterized by reduced ratings of alertness and workload compared to active fatigue. No performance differences were observed between fatigue conditions; however, an overall speed-accuracy trade-off was observed from pre to post fatigue induction. This study demonstrated that different fatigue states produce different effects on EEG indices. These results have potential applications for developing augmented cognition technologies that deliver appropriate fatigue countermeasures in automated operational environments.