BACKGROUND: Fatigue's negative impact on safety represents one of the top threats to military transportation. Biomathematical models have been developed to predict the response to fatigue; however, current models do not take into account stable individual differences in fatigue susceptibility. Readiness Screening Tools (RSTs) can capture individual differences in fatigue response, but cannot predict performance long-term. The objective of this study was to combine an existing biomathematical model of fatigue with existing RST-derived measures to determine current ability to predict individual differences in fatigue response. We hypothesized that the predictive ability of the biomathematical model could be significantly improved by incorporating cognitive and oculometric measures shown to be sensitive to individual differences in fatigue response. METHODS: Data on multiple cognitive and oculometric measures were collected at rested baseline and then every 3 h across 25 h of continual wakefulness. Results characterized actual fatigued performance at the group and individual levels. Actual performance was compared to predicted performance decrements over the same time period. The unique variance explained by each approach was then combined to determine if RST-derived individual difference measures added significant predictive power to the model. RESULTS: Addition of individual-difference sensitive RST measures to an existing fatigue model significantly increased the amount of variance in performance explained by the model from 13.8 to 35.7%. DISCUSSION: Simply leveraging RSTs' ability to capture individual differences in fatigue susceptibility can substantially improve biomathematical prediction of fatigued performance.
BACKGROUND: Fatigue's negative impact on safety represents one of the top threats to military transportation. Biomathematical models have been developed to predict the response to fatigue; however, current models do not take into account stable individual differences in fatigue susceptibility. Readiness Screening Tools (RSTs) can capture individual differences in fatigue response, but cannot predict performance long-term. The objective of this study was to combine an existing biomathematical model of fatigue with existing RST-derived measures to determine current ability to predict individual differences in fatigue response. We hypothesized that the predictive ability of the biomathematical model could be significantly improved by incorporating cognitive and oculometric measures shown to be sensitive to individual differences in fatigue response. METHODS: Data on multiple cognitive and oculometric measures were collected at rested baseline and then every 3 h across 25 h of continual wakefulness. Results characterized actual fatigued performance at the group and individual levels. Actual performance was compared to predicted performance decrements over the same time period. The unique variance explained by each approach was then combined to determine if RST-derived individual difference measures added significant predictive power to the model. RESULTS: Addition of individual-difference sensitive RST measures to an existing fatigue model significantly increased the amount of variance in performance explained by the model from 13.8 to 35.7%. DISCUSSION: Simply leveraging RSTs' ability to capture individual differences in fatigue susceptibility can substantially improve biomathematical prediction of fatigued performance.
Authors: Janet M Mullington; Sabra M Abbott; Judith E Carroll; Christopher J Davis; Derk-Jan Dijk; David F Dinges; Philip R Gehrman; Geoffrey S Ginsburg; David Gozal; Monika Haack; Diane C Lim; Madalina Macrea; Allan I Pack; David T Plante; Jennifer A Teske; Phyllis C Zee Journal: Sleep Date: 2016-04-01 Impact factor: 5.849