Figure 1.

Predictions, task design, and key measures. A, Three possible patterns of anticipatory neural responses to effort. Left, Signals coding for effort per se would scale monotonically with effort regardless of choice. Middle, Signals coding for the decision cost associated with effort should be steeper across effort levels when individuals reject the effort. Right, Signals coding the anticipatory energization needed to accept the challenge should be steeper across effort levels when individuals accept the effort. B, Experimental paradigm. Prescan: participants received visually-guided effort training on a hand-held dynamometer. Levels 1–9 correspond to 10–90% MVC. In the fMRI scanner, participants chose between an effortful option associated with variable amounts of reward and effort and a non-effortful option with smaller reward. Postscan: outside the scanner, eight trials were randomly selected and participants executed the effort they chose in those trials to obtain the reward. C, Behavioral effort sensitivity. This individual measure was derived by calculating for each participant the slope of the probability to choose the effortful option across effort levels. D, Phasic pupil measure. Grand-mean of pupil width during decision-making showed a stereotypical dilation shortly following stimulus onset, peaking right after averaged response onset (purple line), and constricting down to baseline level around stimulus offset. Pupil rate (z/s) was calculated by subtracting pupil width at response from pupil width at stimulus onset, divided by response time (RT). stim=stimulus; resp=response.