Pathway-specific control of reward learning and its flexibility via selective dopamine receptors in the nucleus accumbens

Satoshi Yawata, Takashi Yamaguchi, Teruko Danjo, Takatoshi Hikida, and Shigetada Nakanishi

PNAS (2012)

doi: 10.1073/pnas.1210797109

The conceptual gap or the scientific question:  how D1- and D2- pathways contribute to reward-based learning?

How the authors proposed the question: in fact, it’s an old question. It belongs to one style of scientific research: tackling the old question with novel techniques. 

Brief summary: by using reversible neurotransmission blocking(RNB) and dopamine (DA) receptor agonist or antagonist, the authors found that the disruption of DA D1-receptor expressing direct pathway impaired the initial learning, while the disruption of DA D2-receptor expressing indirect pathway impaired new learning or switch learning after the initial learning.

To dissect the functional roles of the D1- and D2- pathways in reward-based learning, the researchers trained mice to learn multiple tasks within a cross maze. In the visual cue test (VCT), animals learned to associate reward with specific visual cue located outside the maze. In the direction response task (DRT), animals learned to turn at a fixed direction to get reward. To test the behavioral flexibility, mice were asked to learn a new VCT with the reward baited at the opposite arm after the initial learning of the VCT, something similar to reversal learning. Switching from the VCT to the DRT was used to mimic the extra-dimensional shift task in monkey studies. The authors found that deficient D1-receptor mediated signaling impaired initial learning of both tests, while no significant increase in perseverative errors was observed during switch learning. On the contrary, disruptions of the D2-receptor signaling significantly increased the perseverative errors while the initial learning was unaffected. 

Inspired by the study, we can infer that: 1. dopamine suppressed the indirect pathway under physiological conditions; 2. omission of reward decreased dopamine concentration, which disinhibited indirect pathway; 3. disinhibited indirect pathway suppressed the previously learned behaviors; 4. D2 agonist enhanced the dopamine’s modulation on the indirect pathway, and then suppressed it; 5. suppressed indirect pathway led to a failure to inhibit previously learned behaviors after context switch.

The ongoing question(s) inspired by the current study: Under which conditions, the dopamine release will be suppressed to disinhibit the indirect pathway and then facilitated reward-based learning after context switch? One possibility is through the modulations of the cholinergic interneurons (shared in https://www.jianshu.com/p/f57b8cf9345f). And then, how the cholinergic interneurons are modulated?