The brain estimates and updates risk using Bayes‑like priors; exposure therapy works because repeated safe experience corrects overly pessimistic priors. Framing compulsions, panic and ritualistic avoidance as failures of prior calibration explains why confronting feared states produces durable change.
— This framing reframes clinical treatment, public health messaging, and cultural debates about avoidance (e.g., trigger warnings, safetyism) in terms of probabilistic learning and evolutionary adaptation.
Kristen French
2026.09.15
80% relevant
The article supplies new neural-level evidence about how the brain computes risk versus reward: rather than simple evidence accumulation, two localized OFC patches push opposite signals like a see‑saw. This refines and challenges the existing 'brain as Bayesian/risk estimator' idea by adding a distinct mechanistic motif (local antagonistic patches in orbitofrontal cortex) and by reporting predictive power (choice predicted ~0.5 s before action) from UCSF/UC Berkeley intracranial recordings published in Nature Neuroscience.
Steven Gussman
2026.03.03
100% relevant
Steven Gussman’s article uses his OCD experience and the success of exposure therapy (e.g., resisting checking rituals until the feared outcome fails to occur) to argue the brain updates probabilities in a Bayesian manner.
← Back to all ideas