Fig. 1.

Schematic of the tokens task. In each trial, 15 tokens jumped 1 by 1 every 200 ms from the central circle to 1 of the 2 lateral circles (i.e., Jump₁ to Jump₁₅). Subjects had to indicate by a right index or right middle finger key press (i.e., F9 and F8 keys, respectively) which lateral circle they thought would receive more tokens (i.e., left or right circle, respectively) at the end of the trial. They could provide their response whenever they wanted between Jump₁ and Jump₁₅. For a correct response, the subjects earned, in euro cents, the number of tokens remaining in the central circle at the time of the response. Hence, the reward received for a correct response decreased over time, as depicted on the upper right side of the figure (green trace). Example presented on the lower left side of the figure represents a correct response provided between Jump₉ and Jump₁₀ (i.e., the score indicates that 6 tokens remained in the central circle at the moment the right circle was chosen). In contrast, if subjects chose the incorrect lateral circle, they lost €0.07, regardless of their reaction time. As such, the penalty score was fixed, as shown in red on the upper right side of the figure: the lower middle example represents an incorrect choice of the left circle. Thus the reward-to-penalty ratio decreased over time, producing an increasing sense of urgency over the course of a trial. In the absence of response (“Time Out” trial, lower right side example), subjects were neither rewarded nor penalized (score = 0). For representative purposes, the Time Out message is depicted below the circles in this example, although it was presented above them in the experiment.