Fig 1.
Experimental design and time course of one trial.
(A) We used a 2 × 2 within-subject design, in which individuals had to accept or reject a monetarily costly action benefiting a POS ORG or a profitable action entailing a moral cost (allowing the transfer of money to a NEG ORG), either in presence or absence of observers (“public” versus “private”). (B) The amounts of the potential transfers to the organizations and of the potential costs or benefits to the subjects were varied independently across trials. In each trial, the organization potential gains ranged from 4 to 32 Euros, by increments of 4 Euros. The subjects’ potential benefits (in the NEG ORG) or costs (in the POS ORG) varied from 1 to 8 Euros, by increments of 1 Euro. This manipulation resulted in 64 different dilemmas. (C) For both the private and public conditions, each trial began with the presentation of an offer that the participant could either accept or reject by pressing a left or right button response, respectively. Then, a “feedback” screen was shown, consisting of an unchanged screen in the private condition (lasting 500 ms after choice) to keep the chosen option private (no one in the scanner room could see the choice), or in highlighting the chosen option by increasing its font for 1.5 s (while the other option disappeared) in the public condition to further emphasize the presence of observers during this condition. ITI, inter-trial interval; NEG ORG, negatively evaluated organization; POS ORG, positively evaluated organization.
Fig 2.
(A) For the NEG ORG trials, an increase in moral costs (amount potentially transferred to NEG ORG) reduced the likelihood of a transfer being accepted, while potential monetary benefits increased this likelihood. For the POS ORG, an increase in the subject’s monetary cost decreased the likelihood of accepting the offer, while an increase in the moral benefits (POS ORG potential gain) significantly increased the acceptance rate. (B) Effect of audience on accepting offers for both organizations. For the NEG ORG, participants were more likely to accept the offers (i.e., earn money at a moral cost) in private than in public. In the charity condition, individuals were more likely to accept offers (i.e., make prosocial decisions) in public than in private. **p < 0.01. (C) RTs showing interactions between organization types and accept or reject decisions (***p < 0.001), regardless of audience/privacy effect. Participants were faster to accept transfers in the POS ORG condition and to reject them in the NEG ORG condition, suggesting opposite default strategies for the two organizations. See S1 Data. NEG ORG, negatively evaluated organization; POS ORG, positively evaluated organization; RT, response time.
Fig 3.
In the NEG ORG context, bilateral insula and lateral PFC activity correlated with DV at the time of decision-making.
A. BOLD response in the bilateral insula and lateral PFC was positively correlated with the DV in the NEG ORG. B. Heatmaps were created by averaging parameter estimates versus baseline within anterior insula and lateral PFC clusters for each cell of a 4 × 4 monetary benefit versus moral cost matrix (resulting from collapsing the original 8 × 8 matrix, see Methods). Color coding reflects the strength of the neural response, with dark red representing the strongest activation and dark blue representing the lowest activation. See S1 Data. fMRI data corresponding to this figure can be found at the following URL: https://neurovault.org/collections/5028/. BOLD, blood-oxygenation-level–dependent; DV, decision value; fMRI, functional MRI; lPFC, lateral PFC; NEG ORG, negatively valued organization; PFC, prefrontal cortex.
Fig 4.
In the positively valued organization condition, ventral putamen activity correlated with DV, and ventromedial PFC was engaged with prosocial choices.
(A) Activity in ventral putamen correlating with the DV in the POS ORG context. Heatmaps were created by averaging parameter estimates versus baseline within the ventral putamen clusters revealed by the DV correlation for the POS ORG for each cell of a 4 × 4 monetary cost versus moral benefit matrix (resulting from the collapsing of the original 8 × 8 matrix, see Methods). Color coding reflects strength of neural response for each condition, such that dark red represents the strongest activation and dark blue represents the lowest activation. (B) The vmPFC showed stronger activation when subjects chose the prosocial option (i.e., accept) than when they chose the selfish one (i.e., reject) in the POS ORG context. Plots of mean PSCs are shown for illustrative purpose. The color-coded heatmap was created by averaging parameter estimates versus baseline within the vmPFC cluster for each cell of the 4 × 4 monetary cost versus moral benefit matrix. See S1 Data. fMRI data corresponding to this figure can be found at the following URL: https://neurovault.org/collections/5028/. DV, decision value; fMRI, functional MRI; PCS, percent signal change; PFC, prefrontal cortex; POS ORG, positively valued organization; vmPFC, ventromedial prefrontal cortex.
Fig 5.
Brain responses in public relative to private.
Being observed increased activity in the bilateral INS, ACC, and right TPJ. Activations are overlaid on an average anatomical scan of all subjects (cluster-wise FWE-corrected p < 0.05 and voxel-wise uncorrected p < 0.001). Plots of mean PSCs are shown for illustrative purpose. For the POS ORG and the NEG ORG, respectively, light pink and light blue represent the private trials, while dark pink and dark blue are used for the public trials. Error bars indicate SEM. See S1 Data. fMRI data corresponding to this figure can be found at the following URL: https://neurovault.org/collections/5028/. ACC, anterior cingulate cortex; fMRI, functional MRI; FWE, family-wise error; INS, insula; NEG ORG, negatively valued organization; POS ORG, positively valued organization; PSC, percent signal change; TPJ, temporoparietal junction.