Fig. 1

Multi-category valuation paradigm and behavioral pleasantness ratings. a Four outcome categories were included in the paradigm. These categories, resulting from the crossing between outcome valence (reward/punishment) and outcome type (primary/secondary), were: viewing pleasant faces, receiving electric shocks on one’s ankle, monetary gains, and monetary losses. Each of these outcome categories was represented by a distinct shape, which the participants learned beforehand. These shapes served as cues that predicted the potential upcoming outcome in the current trial. b Four levels/intensities were included within each outcome category. Participants only received the predicted outcome in 1/3 of the trials; in the remaining trials nothing was delivered. In those actualized trials, participants could receive one of four different levels/intensities of the predicted outcome, which was indicated by the levels to which the corresponding shapes were filled. For instance, in monetary gains, the amount of money delivered could be $1, $5, $20, or $100. For other outcome categories see the Methods section. c Trial timeline. Each trial began with a cue presentation. Participants had to provide their pleasantness rating for the cue within 5.5 s. Two buttons on a botton box moved a cursor to the left and to the right along the 1–9 scale, and a third button was used to register the selection. A brief delay period of 0.5 s and a presentation of the outcome for 2 s then followed. For actualized shock trials, a 2 ms electric shock of the specified level was delivered during this time. For non-actualized trials of all categories, the message “No Outcome” appeared on the screen. Participants then had another 4 s to rate the outcome pleasantness. The trial length was held constant at 12 s, and the inter-trial interval was jittered between 10 s and 12 s. d Pleasantness ratings for cues predicting outcomes of different levels and of different categories (average across 18 participants, error bars denote SEM). A statistically significant two-way interaction between valence and magnitude indicated opposing trends of pleasantness ratings as a function of magnitude in reward and punishment categories. See Supplementary Fig. 2 for the pleasantness ratings for delivered and non-delivered outcomes