Dissociable and Paradoxical Roles of Rat Medial and Lateral Orbitofrontal Cortex in Visual Serial Reversal Learning

M E Hervig, L Fiddian, L Piilgaard, T Božič, M Blanco-Pozo, C Knudsen, S F Olesen, J Alsiö, T W Robbins

Cerebral Cortex (2020)

doi: 10.1093/cercor/bhz144

One sentence summary: by training rats to learn a deterministic touchscreen serial visual reversal learning (RL) task, the authors found that the lesions of lOFC impaired RL during perseverative phase, while the lesions of mOFC, or IL/PrL, facilitated RL during perseverative phase or the overall performance, respectively.

The improvement of RL after the lesions of mOFC was accompanied by a significant increase of lose-shift strategy in behavioral choice, indicating an enhanced sensitivity to negative feedback.

How to interpret the facilitation effects after IL/PrL lesions? The authors proposed an opposing framework as the one proposed in the study shared yesterday (2022-02-28vmPFC: a brake in the brain for reversal learning - 简书 (jianshu.com)): a shift from habitual toward goal-directed behavior. I think this is more reasonable. A habitual behavior means, after reversal, animals keep choosing the previously rewarded, but now unrewarded stimulus, regardless of the reversed stimulus-reward contingencies, which would lead to an impairment of RL. On the contrary, goal-directed behavior means animals adapt their choices based on the outcome valence, which would lead to a quick stop in choosing the previously rewarded, but now unrewarded stimulus. That is, a facilitation of RL. So this interpretation is much more acceptable. Be notice it's just a theoretical account.

Then how about the observed facilitation effects after BLA lesions? The authors failed to provide an even seemingly convincing explanation. I think this case perfectly exemplifies the difficulties in understanding the brain if each area was treated as an "isolated" working unit. It's a failure of the loss-of-function philosophy in dissecting the roles of one brain area!