Figure 2. Task performance. A, top, Pedal trajectories of right and left forelimbs during task performance in hemiparkinsonian rat. Middle and bottom, Averaged population EMG activity (±SEM) of both forelimbs aligned to pedal-release onset, during right or left forelimb movements (left and right columns, respectively). B, EMG activity of left forelimb in healthy rat showing left muscles activity change associated with movement execution of the left forelimb. C, Rate of correct responses for each forelimb in hemiparkinsonian rat. In the first 4 d of training, the number of correct trials was higher for the right forelimb than the left forelimb; **p < 0.01, two-way ANOVA with Tukey–Kramer post hoc test. D, Correct rate after block change in hemiparkinsonian rat. Solid lines represent the number of trials needed to correctly change the response after a block change in training day 14. At the end of the training period, this number was similar for both forelimbs. The faint lines represent training day 6. E, The rat chooses the correct pedal based on the reward. Relative to healthy rats, lesioned rats exhibited a higher bias toward the right in the initial 4 d of training; **p < 0.01 compared to the corresponding day of healthy group, two-way ANOVA, Tukey–Kramer post hoc test. F, Rate of correct responses. The healthy rats showed a higher rate of correct responses during days 7–9, choosing the rewarded pedal >75% of the time; **p < 0.01 compared to the corresponding day of healthy group, two-way ANOVA with Tukey–Kramer post hoc test. G, Number of total trials per session.