Mental Health and Bayesian Reasoning

Title: "Bayesian reasoning implicated in some mental disorders"

Link: https://www.sciencenews.org/article/bayesian-reasoning-implicated-some-m...

Reputability Statement: This article appeared in ScienceNews, which is the magazine published by the Society for Science and the Public (SSP). SSP was founded almost a century ago and runs some of the largest national science competitions for high school students.

Comments: Some modern scientists argue that the human mind effectively uses Bayes’ theorem to make decisions and rationalize. This article discusses recent evidence that some mental disorders might actually be rooted in a flawed use of Bayesian reasoning by the human mind. The human mind is constantly receiving inputs from sensory neurons and is responsible for integrating those inputs to produce an output. Since information is often imperfect or incomplete, the human mind must frequently make inferences about (for example) what another person’s body language suggests.

 

The article gives the example of interpreting a person’s smile. Typically, smiles are associated with happiness or friendliness; so, if you see someone smile, you’d be inclined to think that he/she is being friendly with you, based on your past experiences with people smiling at you. However, in the case of a person with a mental disorder, his/her mind might come to a different conclusion about what a smile means, if it inputs different probabilities into Bayes’ equation. Therefore, this person might interpret the smile to be antagonistic and react in a different manner as most other people would.

 

I thought this was an interesting example of how Bayes’ theorem applies in real life. This finding suggests that psychologists and scientists should look more closely at what might be causing this difference in Bayesian reasoning between people with mental disorders and people without those disorders. If the root causes of these differences can be identified, then targeted treatments could potentially be developed to recalibrate patients’ Bayesian reasoning functions.