This is a review of an interesting study from Abhishek Bannerjee & Burkhard Pleger’s lab. Summarising what I learnt from this study.

Summary

Flexible decisions have always intrigued and caught the attention of neuroscientists. This is an interesting paper which talks about the role of OFC & S1 in carrying out a Go/No-Go tactile reversal learning task in humans. The study prior to this was done in mice as a proof of concept.

Main Findings:

Engagement of OFC and primary somatosensory cortex (S1) is explored in this paper. Flexibility in human behavior is linked to activity in OFC which sends top-down signals to the S1. Using fMRI and RSA, here they showed that there is distinct but short-lived activity in OFC during rule learning and rule switch. While in S1, the activity is more sustained and is observed both during naïve learning and after rule reversal. Further more, they show that activity in contralateral S1 reflects sensory input while ipsilateral S1 mirrors the value outcome during re-learning. And this activity in ipsilateral S1 is dependent on the OFC activity.

Experiment Task:

Tactile Go/No-Go reversal learning— In this task, participants have to learn the reward associations with the tactile stimuli. One of the stimuli has a reward probability of 0.7 while the other has 0.3, subsequently this reward association is reversed after some trials and the participant has to re-learn the rule of the game.

The whole trial is divided into 4 subparts: learning naïve, learning expert, reversal naïve, reversal expert. Brain activity corresponding to these specific periods is thus used in the observation and analysis of the experiment.

Conclusion & Inferences:

Human lOFC (lateral orbitofrontal cortex) encodes value outcome during the learning. Signals from lOFC influence the ipsilateral activity of S1. The fact that activity in lOFC is transient and short termed—present only during rule switch, suggests that the activity in neurons of lOFC must be closely linked with attentional mechanisms via some fast scale neuromodulation. Moreover, it must also be receiving inputs from higher order neurons of sensory cortices of the brain, as it has to encode or adjust to the new rule and this information is only available through sensory circuits.

Future influences of this article could be:

• To investigate at circuit level how lOFC encodes outcome value.
• To understand at circuit level, feature difference in new rule learning and its reversal in the OFC and S1.
• To develop a computational understanding of fast scale neuromodulation in OFC.