Fig. 1.

The Response-outcome contingency conditions consisted of four blocks of trials, each of which selectively rewarded correct responses or punished incorrect responses. Reward conditions emphasized correct go responses (reward go – RG) or correct response inhibition (reward no-go – RN). Punishment conditions emphasized consequences of incorrect go responses (punish go – PG) or incorrect response inhibitions (punish no-go – PN). Each block was associated with specific gains or deductions for correct or incorrect responding. For trials of the stimulus–response condition RG, correct Go responses (Hit) were associated with high gains (+10 points) and correct inhibitions (CI) were rewarded with moderate gains (+1 point), whereas all incorrect responding, that is, commission errors (i.e. pressing the button when not required, CE) and omission errors (i.e. not pressing the button when required, OE) had no consequences (±0). The same was true for the RN condition, except that the magnitude of gains for Hit and CI were swapped. During trials of the punishment conditions (PG and PN), incorrect responses (i.e. CE and OE) were accordingly associated with high (−10 points) and moderate (−1 point) deductions. Correct responses had no consequences (±0). Neutral blocks without gains or deductions were presented before and between the four reinforcement blocks. Subjects received feedback about their response outcomes immediately after each trial.