Fig. 1 | Nature Communications

Fig. 1

From: The cost of obtaining rewards enhances the reward prediction error signal of midbrain dopamine neurons

Fig. 1

HLC saccade task. a The HLC saccade task. Cost cues (C cue) signaled the amount of effort that was required to achieve a potential reward. A long fixation was required during the delay period in high-cost trials. The reward cue (R cue) indicates whether the monkeys could obtain a reward or not. b The time course of the gaze angle during the delay period. The upper and lower panels show the horizontal and vertical gaze angles, respectively. Green and purple lines indicate the gaze angle in high-cost trials (50 trials in each panel) and in low-cost trials (50 trials in each panel), respectively. c The number of errors during the delay period in the high-cost and low-cost trials (**P< 0.01; two-tailed paired t test; t67 = 8.8, P = 4.8 × 10−15, n = 68 for Monkey P; t83 = 26.6, P ≈ 0, n = 84 for Monkey S). Black circles and error bars indicate mean and SEM. d Success rates in the high-cost and the low-cost trials (two-tailed paired t test; t67 = 0.51, P = 0.61, n = 68 for Monkey P; t83 = 0.79, P = 0.43, n = 84 for Monkey S). e RTs to the cost cues (**P < 0.01; two-tailed paired t test; t67 = 20.4, P ≈ 0, n = 68 for Monkey P; t69 = 2.0, P = 1.2 × 10−3, n = 70 for Monkey S). f RTs to the reward cues (**P < 0.01; two-tailed paired t test; Monkey P (n = 68): HC+ vs. LC+, t67 = 3.5, P = 9.2 × 10−4; HC− vs. LC−, t67 = 24.5, P ≈ 0; HC+ vs. HC−, t67 = 21.6, P ≈ 0; LC+ vs. LC−, t67 = 28.5, P ≈ 0; Monkey S (n = 70): HC+ vs. LC+, t69 = 5.6, P = 4.4 × 10−7; HC− vs. LC−, t69 = 4.8, P = 8.4 × 10−5; HC+ vs. HC−, t69 = 18.0, P ≈ 0; LC+ vs. LC−, t69 = 5.9, P = 1.1 × 10−7)

Back to article page