Task-switching with antisaccades versus no-go trials: a comparison of inter-trial effects

Antisaccades involve the suppression of a pre-potent prosaccade and a vector inversion to generate the novel ocular motor response of looking away from the target. Antisaccades have also been found to prolong the latencies of saccades in upcoming trials, an effect that we attribute to a form of immediate plasticity in the ocular motor system. Our goal was to determine whether the inter-trial effects of antisaccades were similar to that of no-go trials, where subjects must suppress making a saccade when the target appears without substituting a novel ocular motor response. We tested 12 subjects with two different blocks of saccadic trials. In one, prosaccades randomly alternated with antisaccades. In the other, prosaccades alternated with no-go trials. We analyzed the error rates and latencies of prosaccades that followed antisaccades versus no-go trials, compared to repeated prosaccades, to determine if inter-trial effects were present for both types of responses that required prosaccade suppression. No-go responses increased the error rates of prosaccades in the following trial less than antisaccades did. However, no-go trials had the same effect on the latencies of upcoming prosaccades as antisaccades. The inhibitory effect that prolongs the latencies of prosaccades after antisaccades likely stems from the need to inhibit a prosaccade, a function that is also required in no-go trials. The greater impairment of prosaccade accuracy after an antisaccade may reflect either additional control mechanisms involved in vector inversion or a different form of inhibitory control that operates during antisaccades and not during no-go responses.