Pre-saccadic perceptual facilitation: top-down covert shift of attention or automatic enhancement of visual processing?

Abstract

Immediately before making an eye movement discrimination performances at the saccade goal is improved. Is this facilitation top-down covert attention or an automatic enhancement of visual processing? We compared the performance of ten subjects in identifying a target letter (b,d,p,q, 4AFC) in their upper or lower visual field (2 possible target locations at 7° eccentricity on the vertical meridian) both in a covert attention task and while planning a saccade, in both cases guided by a central arrow cue. Because we suspect that better performance in the lower visual field is a marker for covert attention, we ask whether the same bias is present for the pre-saccadic facilitation. To investigate the top-down influences, we manipulated the cue validity probability. When the cue validity was 90%, there was a clear lower field preference in the pre-saccadic letter identification performance that correlated strongly with the bias observed in the covert attention condition, suggesting that pre-saccadic facilitation could be mainly due to covert attention. However, with 50% and 10% cue validity, performance in the covert and pre-saccadic conditions differed. Covert performance followed the cue validity (i.e. attention was allocated according to the probability of target appearance, not simply the cue direction). Pre-saccadic discrimination remained better at the cued location (saccade target), despite showing improvement at the non-cued location as cue validity decreased. Importantly, the lower field preference was not present for discriminations away from the saccade target (invalid trials) and therefore the bias was de-correlated from the covert performance. In summary, reducing cue validity induces a dissociation between covert attention and pre-saccadic facilitation, breaking the lower field bias. We conclude that pre-saccadic facilitation is an automatic processing enhancement that can be combined with additional covert attention. Meeting abstract presented at VSS 2012