Matthew Smith¹, Chris Scholes¹, Neil W Roach¹; ¹University of Nottingham

Our ability to detect visual stimuli is suppressed immediately before and during saccadic eye movements, but the mechanisms underlying this change in sensitivity are not fully understood. One proposal is that saccadic suppression is driven by extra-retinal signals generated during motor preparation which act to cancel the perceptual consequences of rapid eye movements. In primates, neuronal activity in superior colliculus prior to a saccade has been identified as a candidate extra-retinal signal. Pre-saccadic activity in superior colliculus build-up activity is stronger before pro-saccades (where the eyes move towards a target) than before anti-saccades (where the eyes are instead diverted away from a target). Here, we asked whether behavioural suppression, measured in humans, also differs between the two saccade types. To test this, we employed a pro/anti-saccade task while measuring participants’ contrast sensitivity for detecting a transient horizontal grating presented in the upper or lower half of the screen. The time course of saccadic suppression was characterized by flashing the grating at three different time points relative to an individual's median saccadic reaction time and exploiting natural variations in saccadic reaction time to sample the entire perisaccadic time window. We found that suppression differed between the saccade types in the pre-saccade window (from 50ms before the eye starts to move up to saccade onset), with suppression beginning significantly later for anti-saccades than for pro-saccades. This effect is not explained by differences in saccade amplitude or peak velocity. Given that our experimental paradigm was designed to minimize passive suppression mechanisms (e.g., blur, masking), our results suggest that active saccadic suppression is altered by behavioural context. Changes in behavioural suppression may be linked to differences in superior colliculus build-up activity for pro- and anti-saccades.