Embodying addiction: A predictive processing account

Worth a full read, but here’s a quick NotebookLM summary:

This article explores addiction from a predictive processing (PP) perspective, arguing that addiction is not solely a brain disease but rather a breakdown in the dynamic relationship between an individual and their environment. It challenges the traditional reward learning model of addiction, which posits that substance use hijacks the brain's reward system. Instead, PP proposes that addiction arises from aberrant precision estimation, where an individual overestimates the reliability of drug-seeking and using behaviors, leading to a distorted perception of the environment and a neglect of other life opportunities. The authors further suggest that bodily feelings play a crucial role in this process, providing feedback on the rate of error reduction and influencing precision weighting. By viewing addiction as a disruption of the agent-environment system, the article encourages a broader understanding of addiction's societal and environmental causes, promoting a more holistic approach to treatment and prevention.