ABSTRACT
Saccades (voluntary high velocity rotations of the eye) are often used as overt indicators of shifts of visual attention even though attention shifts and saccades are not always coincident. Smooth eye movements may provide more accurate indicators because they are involuntary, but are, nevertheless, affected by cognitive factors. We first review the evidence that selective attention affects smooth eye movements. A recent experiment by Kowler, van der Steen, Tamminga, and Collewijn (1984) showed that subjects could select one of two full-field, identical, superimposed patterns of randomly positioned dots, one stationary and the other moving, as the target for smooth eye movements. There was virtually no influence of the background on smooth eye movement velocity. Effective selection of the target was due to attention and not stimulus variables because target and background were physically identical. We now report that performance of a concurrent visual task (detection of the disappearce of dots from either the stationary or the moving field) was faster and more accurate when dots disappeared from the target than from the background. This result would not be predicted from the differences in the retinal velocities of target and background because the retinal velocity of the background was low enough (about 1 deg/sec) so that visual acuity and contrast sensitivity were not impaired. We conclude that the smooth oculomotor subsystem and the visual system share the same attentional mechanism. This implies that smooth eye movements provide accurate indicators of selective attention during performance of a visual task with stationary and moving stimuli.
