This is a plot of the number of citations turned up by a simple “Web of Knowledge” search for papers containing the words “dopamine” and “reinforcement learning”, against year of publication. The rise, dating from approximately the time of publication of the first computational theory of phasic dopamine function, is rapid. There are, as far as I know, two computational theories of phasic dopamine function. One from Schultz, Dayan and Montague (1997) and one from our team here in Sheffield (Redgrave and Gurney, 2006)
Refs
Schultz, W., Dayan, P., & Montague, P. R. (1997). A neural substrate of prediction and reward. Science, 275, 1593-1599.
Redgrave, P., & Gurney, K. (2006). The short-latency dopamine signal: a role in discovering novel actions? Nature Reviews Neuroscience, 7(12), 967-75.