Fig. 2: A hypothesised framework of how distinct dopaminergic circuits could be altered in schizophrenia.

People with schizophrenia show a learning paradox, characterised by increases in learning about neutral or irrelevant information, and a decrease in learning about reward-predictive information. This is associated with an increase in activity in ventral striatum to neutral or irrelevant information, and a decrease in activity to reward-predictive information [27, 29, 30, 81, 113]. One way to account for this deficit is to hypothesise that the inputs to the ventral striatum that regulate the balance between learning about reward-predictive, neural, and irrelevant information, are changed. Specifically, we would argue that the paradox seen in schizophrenia is consistent with a strengthening of orbitofrontal inputs to the ventral striatum, and a decrease in the inputs from prelimbic and lateral hypothalamic circuits. The feasibility of such a model is supported by the separable nature of afferents coming from these regions to ventral striatum [156, 159, 175, 197, 198]. Further research is necessary to test the validity of this model using preclinical studies in the context of reinforcement learning.