Adaptation development and after-effects in the three experiments.
Gain changes computed relative to pre-exposure are represented as a function of the phases (pre-, four blocks of exposure, post-). For the backward (A) and forward (B) experiments the intra-saccadic (blue), post-saccadic (purple) and no-shift (black) sessions are shown separately. For the acceleration/deceleration (C) experiment, the acceleration (red), deceleration (green) and no-shift (black) sessions are represented separately. The symbols represent mean gain changes for pre- and post-exposure phases as well as for each block of the exposure phase. The shaded areas in the exposure phase represent one standard error of the mean (SEM). The error bars in post-exposure phase show SEMs. The asterisks indicate significant differences in after-effect (computed as post- versus pre- difference) between both the intra-saccadic and post-saccadic sessions with the no-shift session: *** P<0.001 (post-hoc Bonferroni tests). Note that the baseline gain of saccades measured during the pre-exposure phase was submitted for each experiment to a two-way ANOVA with session and saccade direction (rightward vs leftward) as factors. No main effect and no interaction was detected for any of the three experiments (backward: F2,18<0.71, P>0.5; forward: F2,18<1.28, P>0.3; acceleration/deceleration: F2,8<3.76, P>0.07). This indicates that baseline gain was similar for the three sessions of each experiment. Moreover, because of the lack of effect of the saccade direction factor, data of the rightward and leftward saccades were pooled for this figure and for subsequent analyses.