Distinct Neural Correlates of Expected Value and Bayesian Posterior Estimates in Active Avoidance
Discuss this preprint
Start a discussion What are Sciety discussions?Listed in
This article is not in any list yet, why not save it to one of your lists.Abstract
Choosing whether or not to act is dictated by whether you believe you can affect the outcomes you experience and the cost of acting. If you believe bad things will happen regardless of what you do, there is no incentive for you to perform the often costly actions needed to avoid negative events. This results in helplessness. Here, we used two behavioural tasks, computational modelling, and fMRI in 40 healthy participants to investigate the neural representations of expected value and a posterior avoidability estimate derived from a generalised prior belief concerning helplessness. Based on previous work, we expected the ventromedial prefrontal cortex (vmPFC) to represent the expected value and areas involved in episodic memory, such as the hippocampus and default mode network, to represent trial-by-trial posterior avoidability estimates. Contrary to this, we found that BOLD signals in the lateral orbitofrontal cortex (OFC) were correlated with both estimates. A combined analysis showed that these neural representations are independent and spatially distinct in the OFC. We discuss how these results indicate that the lateral OFC may be an area where information about prior information and utility is integrated to form a final decision.