Figure 3. Pupil dilation during VGT. A, Pupil modulation time courses for VGT with and without sensory cues over the different trial epochs. Modulation is computed as percentage change from baseline in pupil size (measured as area subtended by the pupil in angular units from the point of view of the camera) over the course of last 500 ms of the decision phase, the anticipation phase, and the feedback phase. B, AUC measure of pupil dilation from baseline over the decision and anticipation periods for VGT with the sensory cues. C, AUC pupil dilation over the decision and anticipation periods without sensory cues. D, Pupil dilation with respect to baseline following wins and 0 outcomes for VGT with and without sensory cues. Pupil dilation is plotted as log-transformed AUC measure depicted in A. E, Pupil dilation with respect to baseline for safe (higher probability prospect, solid lines) versus risky (lower probability prospect, dotted lines) choices. F, Pupil dilation with respect to baseline as a function of the chosen prospect's probability. G, Pupil dilation with respect to baseline as a function of the chosen prospect's magnitude. H, Heat maps of fixation durations during baseline (first 200 ms of the decision phase) and the analyzed pupil response period (final 500 ms of decision plus the anticipation phase). *p < 0.05. ***p < 0.0005.