Mesolimbic mechanisms of reinforcement

Albert Einstein College Of Medicine, Inc, Bronx NY

Investigators

Abstract

Project Summary When a stimulus predicts reward at a specific location, animals learn to approach that location in response to the stimulus. When the stimulus no longer predicts reward, the approach behavior extinguishes. Extinction of cued reward approach is thought to be the consequence of ?negative reward prediction error? signals ? neural signals that a predicted reward was not obtained. These signals may be carried by a brief pause in the firing of dopamine neurons when reward is not obtained at the predicted time. Consistent with this hypothesis, recent optogenetics evidence indicates that extinction can be reduced or eliminated by stimulating dopamine neurons in the ventral tegmental area (VTA) at the predicted time of reward. However, the neural mechanisms downstream of the VTA that contribute to this form of extinction, and to its rescue by VTA dopamine neuron stimulation, remain unknown. Our lab has demonstrated that excitations of neurons in the nucleus accumbens (NAc) in response to reward- predictive cues are causal to the approach behavior elicited by such cues. Therefore, we hypothesize that the negative reward prediction error signal carried by a pause in VTA dopamine neuronal firing causes extinction via a mechanism that results in reduction in the magnitude of the cue-evoked excitation of NAc neurons on trials subsequent to the reward omission. We will test this hypothesis by recording the firing of NAc neurons in TH-Cre rats expressing channelrhodopsin in VTA dopamine neurons. We expect that cue-evoked excitations will be gradually reduced when the reward is omitted, and that this reduction will be prevented by optical stimulation of VTA dopamine neurons at the predicted time of reward delivery. In addition, we will determine whether stimulation of dopamine release within the NAc at this time point is sufficient to prevent extinction caused by inactivation of dopamine neurons in the VTA. Together, these experiments will constrain the possible neural mechanisms by which negative reward-prediction error signals cause behavioral extinction to those involving a reduction in the cue-evoked excitation of NAc neurons.