Dopamine role still not settled

vta pfc
No idea why the prefrontal cortex neuron is Australian, but here you go. Cartoon made by me with free (to the best of my knowledge) clipart elements. Feel free to use to your heart’s content.

There have been literally thousands of pages published about the dopamine function(s). Dopamine, which made its stage debut as the “pleasure molecule”, is a chemical produced by some neurons in your brain that is vital to its functioning. It has been involved in virtually all types of behavior and most diseases, from pain to pleasure, from mating to addiction, from working-memory to decision-making, from autism to Parkinson’s, from depression to schizophrenia.

Here is another account about what dopamine really does in the brain. Schwartenbeck et al. (2015) trained 26 young adults to play a game in which they had to decide whether to accept an initial offer of small change or to wait for a more substantial offer. If they waited too long, they would lose everything. After that, the subjects played the game in the fMRI. The authors argue that their clever game allows segregation between previously known roles of dopamine, like salience or reward prediction.

As expected with most fMRI studies, a brain salad lit up (that is, your task activated many other structures in addition to your region of interest), which the authors address only very briefly. Instead, they focus on the timing of activation of their near and dear midbrain dopamine neurons, which they cannot detect directly in the scanner because their cluster is too small, so they infer their location by proxy. Anyway, after some glorious mental (and mathematical) gymnastics Schwartenbeck et al. (2015) conclude that

1) “humans perform hierarchical probabilistic Bayesian inference” (p. 3434) (i.e. “I don’t have a clue what’s going on here, so I’ll go with my gut instinct on this one”) and

2) dopamine discharges reflect the confidence in those inferences (i.e. “how sure am I that doing this is going to bring me goodies?”)

With the obvious caveat that the MRI doesn’t have the resolution to isolate the midbrain dopamine clusters and that these clusters refer to two very distinct population of dopamine neurons (ventral tegmental area and substantia nigra) with different physiological, topographical, and anatomical properties, and distinct connections, the study adds to the body of knowledge of “for the love of Berridge and Schultz, what the hell are you DOIN’, dopamine neuron?”.

Reference: Schwartenbeck, P., FitzGerald, T. H., Mathys, C., Dolan, R., & Friston K. (October 2015, Epub 23 July 2014). The Dopaminergic Midbrain Encodes the Expected Certainty about Desired Outcomes. Cerebral Cortex, 25:3434–3445, doi:10.1093/cercor/bhu159. Article + FREE PDF

By Neuronicus, 8 October 2015