Striatal and Tegmental Neurons Code Critical Signals for Temporal-Difference Learning of State Value in Domestic Chicks

摘要: To ensure survival, animals must update the internal representations of their environment in a trial-and-error fashion. Psychological studies associative learning and neurophysiological analyses dopaminergic neurons have suggested that this updating process involves temporal-difference (TD) method basal ganglia network. However, way which component variables TD are implemented at neuronal level is unclear. investigate underlying neural mechanisms, we trained domestic chicks to associate color cues with food rewards. We recorded activities from medial striatum or tegmentum freely behaving condition examined how reward omission changed firing. compare signals assumed method, simulated behavioral task form finite sequence composed discrete steps time. The three were prediction signal, target signal for updating, TD-error signal. In both tegmentum, majority categorized into types according fitness models, though these tended continuum spectrum without distinct differences firing rate. Specifically, two striatal successfully mimicked A linear summation was good fit activity one type tegmental mimicking present study thus demonstrates can convey critically required method. Based on theoretical studies, together tract-tracing data, propose novel model explain convergence represented could lead computation error neurons.