S.Parkes, E. Coutureau et al. inCerebral Cortex

June 29th, 2017

Insular and Ventrolateral Orbitofrontal Cortices Differentially Contribute to Goal-Directed Behavior in Rodents. Parkes SL, Ravassard PM, Cerpa JC, Wolff M, Ferreira G, Coutureau E. Cereb Cortex. 2017 May 25:1-13. doi: 10.1093/cercor/bhx132. [Epub ahead of print] PMID: 28541407

CNRS, Institut de Neurosciences Cognitives et Intégratives d’Aquitaine, UMR 5287, Université de Bordeaux. / INRA, Nutrition et Neurobiologie Intégrée, UMR 1286, Université de Bordeaux

Appropriate decision-making is critical for adapting to a changing environment and relies on an ability to predict and control events. Goal-directed behaviour facilitates decision-making by allowing organisms to flexibly adapt their actions according to changes in the consequences of those actions. Here, we uncover how distinct regions of the cerebral cortex modulate this complex behaviour in rats.

We adopted a chemogenetic approach to study the involvement of insular cortex (IC) and ventrolateral orbitofrontal cortex (vlOFC) in goal-directed behavior. Specifically, we used DREADDs (Designer Receptor Exclusively Activated by Designer Drugs) to express an inhibitory receptor in either IC or vlOFC. We confirmed using in vitro electrophysiology that activation of the inhibitory DREADD, via application of its agonist CNO, led to a substantial decrease in the firing rate of neurons expressing the receptor. Rats were then trained on the instrumental outcome devaluation task, which is commonly used to study goal-directed behaviour in both rodents and humans. This task involves three distinct phases: action acquisition, devaluation and a choice test. In the first phase, rats learned to associate actions with desirable outcomes. Then, one of these outcomes was devalued and, finally, we assessed the rats’ choice between the two actions. If the rat is behaving in a goal-directed manner it should bias its choice away from the action that delivers the now devalued outcome and towards the action that delivers the still valued outcome.

Shauna Parkes the first author: ” We show that inhibition of IC during the choice phase abolished goal-directed control but inhibition during the initial acquisition phase was without effect. IC is therefore necessary for action selection based on the current value of the available outcomes but not for learning action-outcome relationships. By contrast, we observed a deficit in goal-directed behaviour after vlOFC inhibition but only following a reversal of the instrumental relationships. That is, rats with vlOFC inhibition were unable to update and learn new action-outcome relationships. Our results reveal distinct yet complementary roles for insular and orbitofrontal cortices in adaptive behaviour. We show that IC is required for the retrieval of outcome value to guide choice whereas vlOFC is used to detect changes in the causal relationship between actions and their specific consequences. These findings not only increase our understanding of the cortical bases of decision-making but also have implications for psychiatric disorders characterised by impairments in goal-directed control, including schizophrenia, depression and obsessive compulsive disorder”

Top row (from left to right): The image on the left shows the expression of the DREADD inhibitory receptor (in red) in the insular cortex. The middle image shows the behaviour of the rats: Inhibition of IC during the acquisition phase caused no impairment at the choice test. Indeed, rats were able to correctly bias their choice towards the action associated with the still valued outcome (white bar). By contrast, inhibition of IC during the choice test caused impaired goal-directed behaviour such that rats selected the devalued (black bar) and valued (white bar) actions equally. The boxed image on the right illustrates the marked decrease in firing rate of neurons expressing the inhibitory receptor when CNO is applied in vitro.

Bottom row (from left to right): The image on the left shows the expression of the DREADD inhibitory receptor in the ventrolateral orbitofrontal cortex. The middle image shows the behaviour of the rats: Inhibition of vlOFC was without effect and rats in both groups biased their choice towards the action associated with the still valued outcome (white bar). However, following a reversal of the action-outcome relationships (image in blue), rats were unable to perform in a goal-directed manner and selected both actions equally.

Shauna Parkes / Post-doc / Team DECAD (Etienne COUTUREAU) INCIA