Abstract
During Pavlovian conditioning, if a cue (e.g., lever extension) predicts reward delivery in a different location (e.g., a food magazine), some individuals will come to approach and interact with the cue – a behavior known as sign tracking (ST) – and others will approach the site of reward, a behavior known as goal tracking (GT). In rats, the acquisition of ST vs. GT behavior is associated with distinct profiles of dopamine release in the nucleus accumbens (NAc), but it is unknown whether it is associated with different patterns of accumbens neural activity. Therefore, we recorded from individual neurons in the NAc core during the acquisition, maintenance, and extinction of ST and GT behavior. Even though NAc dopamine is specifically important for the acquisition and expression of ST, we found that cue-evoked excitatory responses encode the vigor of both ST and GT behavior. In contrast, among sign trackers only, there was a prominent decrease in reward-related activity over the course of training, which may reflect the decreasing reward prediction error encoded by phasic dopamine. Finally, both behavior and cue-evoked activity were relatively resistant to extinction in sign trackers, as compared with goal trackers, although a subset of neurons in both groups retained their cue-evoked responses. Overall, the results point to the convergence of multiple forms of reward learning in the NAc.
Significance Statement An individual’s tendency to interact with a cue that predicts reward – known as sign tracking – has been linked with impulsivity and addiction-related behaviors. Here, we show that, during learning, sign tracker rats – as compared with goal tracker rats, who preferentially interact with the site of reward – display different profiles of neuronal activity in the nucleus accumbens (NAc). The evolution of NAc activity is uniquely linked to the acquisition of sign tracking, but not goal tracking; however, after learning, NAc activity reflects the vigor of both behaviors. These findings imply that sign tracking and goal tracking result from different learning processes and engage distinct neural circuits that partially overlap in the NAc.
Footnotes
The authors declare no competing financial interests.
1. Dietrich School of Arts & Sciences, University of Pittsburgh (S.E.M.)
2. UPMC Competitive Medical Research Fund award (S.E.M.)
3. NARSAD Young Investigator Award 25185 from the Brain & Behavior Research Foundation (S.E.M.)
This is an open-access article distributed under the terms of the Creative Commons Attribution 4.0 International license, which permits unrestricted use, distribution and reproduction in any medium provided that the original work is properly attributed.
Jump to comment: