Abstract
Background: We investigated the mechanisms underlying both sustained and transient covert attention. Two models have been proposed to explain how attention affects perception: contrast gain and response gain. Whereas contrast gain predicts a multiplicative increase in sensitivity (characterized by a threshold shift in the contrast response function), response gain predicts an additive increase in firing rate (characterized by a change in slope and asymptote).
Methods: Observers performed a 2AFC orientation discrimination task on a ±4° tilted Gabor (4cpd), presented at one of eight iso-eccentric (4°) locations. Psychometric functions were measured by presenting the target Gabor across a range of contrasts. In each block, either a neutral (baseline), transient (deploying attention exogenously), or sustained (deploying attention endogenously) cue preceded the target Gabor. Demands of the task were adjusted such that observers' psychometric functions did not asymptote at 100% accuracy, ensuring enough room at the upper bounds for response gain to manifest itself.
Results: For all observers, sustained attention caused a shift in threshold across the psychometric function, supporting a contrast gain model. However, with transient attention both a threshold shift and a change in asymptote were observed. These findings suggest that whereas sustained attention operates strictly via contrast gain, transient attention may be better described by a mixture of both contrast and response gain mechanisms.
Supported by the National Science Foundation under grant # BCS-9910734, and by the National Institute of Health under grant # 1 F31 NS051111-01.