Talk by Dr. Paul Chorley
Department of Informatics, University of Sussex, United Kingdom
- 31 Jul 2013 14:00
- 31 Jul 2013 15:30
- INM-6, Bldg. 15.22, Seminar Room 3009, 1. OG
Spiking Model of Phasic Dopamine Signals
Dopaminergic neurons in the mammalian substantia nigra display characteristic phasic responses to stimuli which reliably predict the receipt of primary rewards. These responses have been suggested to encode reward prediction-errors similar to those used in reinforcement learning theory. In this talk I describe a spiking neural network model of dopaminergic activity in which such precisely-timed phasic signals are generated by the joint action of short-latency excitation and long-latency inhibition, in a network undergoing feedback neuromodulation of both spike-timing dependent synaptic plasticity and neuronal excitability. Significantly, sensitivity to recent events is maintained in the model by selective modification of specific striatal synapses efferent to cortical neurons, which exhibit stimulus-specific and temporally extended patterns of activity. The model shows, in the presence of significant background activity; (i) a shift in dopaminergic responses from reward to reward-predicting stimuli, (ii) preservation of a response to unexpected rewards and (iii) a precisely-timed below-baseline dip in activity observed when expected rewards are omitted, therefore providing an explicit mechanistic account of the major features of the observed phenomenology. A brief discussion of the model assumptions in light of recent advances in our understanding of the neuromodulatory action of dopamine will also be presented.