Figure 2

Effects on pupil size, given by linear regression at each time point. For each participant, the pupil traces were correlated with the incentive on the current trial, the incentive on the previous trial, and time on task. The coefficient of each of these factors was plotted as a function of time. (a) The influence of reward on pupil diameter is shown. Positive values indicate that with higher incentives, the pupil was larger; conversely negative values indicate that reward made the pupil smaller. Comparisons of these coefficients with zero, and with each other, are shown. Reward increased pupil size in all three groups, but controls were significantly more reward sensitive than PD patients when OFF medication (unpaired comparison). Also, PD patients were more reward sensitive when ON compared with when OFF (paired comparison). All statistics are calculated for P<0.05 controlling for family-wise error using permutation. (b) The previous trial’s incentive influenced the pupillary response to the subsequent incentive cue. Negative values signify that larger previous incentives result in a smaller pupil diameter on the current trial. There was a significant effect in all groups, and no significant differences between groups. (c) As the task progressed, pupillary dilatation effects might vary. To account for this, time on task (i.e., trial number) was included as a regressor. Positive values signify the pupillary response to the cue is more dilated later in the session, compared with earlier in the session. The value was not significantly different from zero in any group, although there was a late-in-trial (1,372–1,683 ms after cue) effect of medication, resulting in smaller pupils later in the session when ON compared with OFF. This might be attributable to increased fatiguability associated with D2 agonists. PD, Parkinson’s disease.