Citation

Galley N, Andres G, Reitter E. Vis. Veh. 1999; 7: 49-59.

Copyright

(Copyright © 1999, Elsevier Publishing)

DOI

unavailable

PMID

unavailable

Abstract

All saccade and blinks were measured for 20 subjects during 40 trips on the Autobahn lasting six hours. They were identified on-line out of the electrooculogram and stored on a hard disk together with the self-reported alertness/fatigue scores. The relationships of fixation duration, saccadic velocity, blink rate, blink amplitude, lid closure velocity and so on (in total 19 parameters) to subjectively scored fatigue and time on tasks as well as their interrelationships were analyzed. The main result is that subjectively scored alertness/fatigue can be predicted by oculomotor parameters - by blink parameters better than by saccade parameters - but the best predictor was time of task. The sign of the correlations between alertness/fatigue and the oculomotor parameters is different in persons; this necessitates an interpretation that in 'fatigue' at least two processes are at work; a deactivation one, for which we adopt the name 'sleep propensity function ; (Lavie, 1991) and an activation process, the struggle to keep awake. A factor analysis results in four dimensions, the first associated with blink intervals and amplitudes, the second with blink duration, the third with fixation duration and velocity of saccades and blinks and the last with saccade durations and amplitudes.