To make decisions in a social context, humans have to predict the behavior of others, an ability that is thought to rely on having a model of other minds known as "theory of mind." Such a model becomes especially complex when the number of people one simultaneously interacts with is large and actions are anonymous. Here, we present results from a group decision-making task known as the volunteer's dilemma and demonstrate that a Bayesian model based on partially observable Markov decision processes outperforms existing models in quantitatively predicting human behavior and outcomes of group interactions. Our results suggest that in decision-making tasks involving large groups with anonymous members, humans use Bayesian inference to model the "mind of the group," making predictions of others' decisions while also simulating the effects of their own actions on the group's dynamics in the future.

To make decisions in a social context, humans have to predict the behavior of others, an ability that is thought to rely on having a model of other minds known as theory of mind. Such a model becomes especially complex when the number of people one simultaneously interacts is large and the actions are anonymous. Here, we show that in order to make decisions within a large group, humans employ Bayesian inference to model the "mind of the group," making predictions of others' decisions while also considering the effects of their own actions on the group as a whole. We present results from a group decision making task known as the Volunteers Dilemma and demonstrate that a Bayesian model based on partially observable Markov decision processes outperforms existing models in quantitatively explaining human behavior. Our results suggest that in group decision making, rather than acting based solely on the rewards received thus far, humans maintain a model of the group and simulate the group's dynamics into the future in order to choose an action as a member of the group.

Bribery is a common form of corruption that takes place when a briber suborns a power holder to achieve an advantageous outcome at the cost of moral transgression. Although bribery has been extensively investigated in the behavioral sciences, its underlying neurobiological basis remains poorly understood. Here, we employed transcranial direct-current stimulation (tDCS) in combination with a novel paradigm ( N = 119 adults) to investigate whether disruption of right dorsolateral prefrontal cortex (rDLPFC) causally changed bribe-taking decisions of power holders. Perturbing rDLPFC via tDCS specifically made participants more willing to take bribes as the relative value of the offer increased. This tDCS-induced effect could not be explained by changes in other measures. Model-based analyses further revealed that such neural modulation alters the concern for generating profits for oneself via taking bribes and reshapes the concern for the distribution inequity between oneself and the briber, thereby influencing the subsequent decisions. These findings reveal a causal role of rDLPFC in modulating corrupt behavior.

Humans frequently interact with other agents whose intentions can fluctuate over time between competitive and cooperative strategies. How does the brain decide whether the others’ intentions are to cooperate or compete when the nature of the interactions is not explicitly signaled? We used model-based fMRI and a task in which participants thought they were playing with another player. In fact, this agent was an algorithm alternating without signaling between cooperative and competitive strategies. A neurocomputational mechanism underlying arbitration between competitive and cooperative experts outperforms other learning models in predicting choice behavior. The ventral striatum and ventromedial prefrontal cortex tracked the reliability of this arbitration process. When attributing competitive intentions, these regions increased their coupling with a network that distinguish prediction error related to competition versus cooperation. These findings provide a neurocomputational account of how the brain dynamically arbitrates between cooperative and competitive intentions when making adaptive social decisions.

Learning one’s status in a group is a fundamental process in building social hierarchies. Although animal studies suggest that serotonin (5-HT) signaling modulates learning social hierarchies, direct evidence in humans is lacking. Here we determined the relationship between serotonin transporter (SERT) availability and brain systems engaged in learning social ranks combining computational approaches with simultaneous PET-fMRI acquisition in healthy males. We also investigated the link between SERT availability and brain activity in a non-social control condition involving learning the payoffs of slot machines. Learning social ranks was modulated by the dorsal raphe nucleus (DRN) 5-HT function. BOLD ventral striatal response, tracking the rank of opponents, decreased with DRN SERT levels. Moreover, this link was specific to the social learning task. These findings demonstrate that 5-HT plays an influence on the computations required to learn social ranks.

Background To show the equivalence between the specific binding ratios (SBR) of visually normal 123I-FP-CIT SPECT scans from patients to those from healthy volunteers (Hv) or patients without dopaminergic degeneration to allow their use as a reference database. Methods The SBR values of visually normal SPECT scans from 3 groups were studied: (1) suspected Parkinsonism and no diagnostic follow-up (ScanOnlyDB: n = 764, NM/CT 670 CZT, GE Healthcare), (2) no degenerative dopaminergic pathology after a 5-year follow-up (NoDG5YearsDB: n = 237, Symbia T2, Siemens Medical Solutions), and 3) Hv (HvDB: n = 118, commercial GE database). A general linear model (GLM) was constructed with caudate, putamen, and striatum SBR as the dependent variables, and age and gender as the independent variables. Following post-reconstruction harmonization of the data, DB were combined in pairs, ScanOnlyDB&NoDG5yearsDG and ScanOnlyDB&HvDB before performing GLM analysis. Additionally, ScanOnlyDB GLM estimates were compared to those published from Siemens commercial DB (SiemensDB) and ENC-DAT. Results The dispersion parameters, R2 and the SBR coefficients of variation, did not differ between databases. For all volumes of interest and all databases, SBR decreased significantly with age (e.g., decrease per decade for the striatum: − 4.94% for ScanOnlyDB, − 4.65% for NoDG5YearsDB, − 5.69% for HvDB). There was a significant covariance between SBR and gender for ScanOnlyDB (P < 10–5) and NoDG5YearsDB (P < 10–2). The age-gender interaction was significant only for ScanOnlyDB (P < 10–2), and the p-value decreased to 10–6 after combining ScanOnlyDB with NoDG5YearsDB. ScanOnlyDB GLM estimates were not significantly different from those from SiemensDB or ENC-DAT except for age-gender interaction. Conclusion SBR values distribution from visually normal scans were not different from the existing reference database, enabling this method to create a reference database by expert nuclear physicians. In addition, it showed a rarely described age-gender interaction related to its size. The proposed post-reconstruction harmonization method can also facilitate the use of semi-quantitative analysis.