Saccades vs. Novelty: the joint influence of saccades and repetition on perceived stimulus duration.

The duration of visual stimuli that occur just before and during saccades may be perceived as half of those that occur during fixation (Morrone et al., Nature neuroscience 2005). On the other hand, novel stimuli are perceived as lasting longer than repeated stimuli (Pariyadath & Eagleman, PLoS One 2012). Since these temporal distortions do not follow a scalar association with durations, the Weber’s law has failed to justify both effects. Here, we investigated how these time distortions interact. Seven participants were asked to judge the duration of a test stimulus (three parallel vertical lines) that appeared after a pre-test stimulus (three parallel horizontal lines), randomly presented between 1 and 3 times. Duration of the pre-test stimuli was fixed (200 ms), but duration of the test stimulus was varied (140, 170, 230 and 260 ms). The experiment was performed in three blocks (saccade with fixed spatial / different retinal pre/test retinal locations, fixation control with fixed stimulus locations, and fixation control with different saccade-matched retinal stimulus locations). In the saccade block, participants were cued to perform a saccade 100 ms before presentation of novel stimulus. Consistent with previous findings, we found a significant underestimation of test stimulus duration in saccade trials that followed a single pre-test stimulus (29.3% of longer trials perceived longer compared to 61.9% and 69.3% in the control fixation blocks). However, saccadic time compression was considerably reduced when repeated stimuli preceded a novel test stimulus (67.2% and 70.0% of longer trials perceived longer for 1 and 2 repetition of pre-test, respectively). These results indicate that saccadic time compression and the novelty / reptetion effect do not interact in a scalar accumulative manner. Instead, they can be explained by a nonlinear vector model based on the phase differences between time units (Ghaderi et al., PloS one 2018).