Neural correlates of social influence on individual decision making and learning

This thesis describes investigations on how the brain processes social influences during decision making (herd behaviour) and learning (observational learning and vicarious rewards/punishments) using functional magnetic resonance imaging (fMRI). To probe the neural correlates of social influence on economic decision making a stock-picking task was designed where participants were able to observe the decisions of others. Analysis of the imaging data revealed that activity in the ventral striatum, an area of the brain heavily implicated in reward processing, tracks the subjective degree of social influence on participant’s decisions arising from the decisions of others. Activity in the amygdala was associated with alignment decisions, whereas anterior cingulated cortex was active when participants’ decisions went against the crowd. A second experiment used a social version of the 2-armed bandit task to investigate the neural correlates of social learning. Using a model-based fMRI analysis, it was found that the brain uses computationally similar processes during social learning and non-social learning. Specifically, it is possible to learn from observing another person’s actions and outcomes by processing social prediction errors (in the ventromedial prefrontal cortex) and action prediction errors (in the dorsolateral prefrontal cortex). Individual reward prediction errors were found to correlate with activity in the ventral striatum. A more traditional analysis focused on the similarities and differences in the processing of egocentric (one’s own) and allocentric (others’) outcomes. Ventromedial prefrontal cortex and insula processed both individual and vicarious outcomes, with the ventral striatum only responsive to individually-received rewards. Punishment to others elicited activity in areas of the mirror system such as the inferior frontal gyrus and superior temporal sulcus.EThOS - Electronic Theses Online ServiceGBUnited Kingdo