Abstract
Background: Studies that mix trials with prosaccades and antisaccades have shown that there are carry-over effects between one trial and the next. In particular, a preceding antisaccade leads to increased latencies of the following response, particularly if it is a prosaccade. Whether this antisaccade effect results from effects generated in the execution of the antisaccade or possibly in the planning and preparation process is unknown. Objective: We studied whether prolongation of the latencies of subsequent prosaccades after antisaccade trials depended upon execution of the antisaccade or occurred regardless of whether the antisaccade was actually performed. Methods: 9 subjects were tested using blocks of randomly ordered prosaccades and antisaccades. An instructional cue at fixation indicated whether a prosaccade or antisaccade was required when the target appeared 2 seconds later. On 20% of the antisaccade trials, the target did not actually appear (antisaccade cue/no-target trials). We analysed the latencies of all correct prosaccades or antisaccades that were preceded by correctly executed trials. Results: As expected the latencies of prosaccade trials preceded by prosaccades were 20ms shorter (mean 189ms) than the latencies of prosaccade trials preceded by antisaccades (mean 209ms). Prosaccades preceded by trials where antisaccades were cued but not executed because the target did not appear also showed prolonged latencies (mean 214ms). These differed from the latencies of trials with preceding prosaccades but not from those of trials with preceding antisaccades. Conclusion: The effects of a prior antisaccade in prolonging the latencies of subsequent saccades is generated by the planning of the antisaccade. This may reflect persistence of the known pre-target preparatory activity seen in neural recordings of the superior colliculus and frontal eye field.
Meeting abstract presented at VSS 2014