Bayesian predictive coding hypothesis: Brain as observer’s key role in insight

Abstract

Hypnosis, defined by focused attention and reduced peripheral awareness, integrates psychological processes such as attention, expectancy, and imagery. Self-hypnosis is a process where an individual induces a hypnotic state in themselves to achieve specific goals, such as stress reduction, behaviour modification, or overcoming phobias. This practice involves deep relaxation and heightened focus, allowing suggestions to bypass the conscious mind and influence the subconscious. In contrast, Insight meditation, also known as Vipassana, is a form of mindfulness meditation rooted in Buddhist traditions. Practitioners observe their thoughts, emotions, and bodily sensations as they arise and pass away, gaining deep insights into the nature of reality, impermanence, and the workings of the mind. In this paper, I present a Bayesian Predictive Coding Hypothesis as a theoretical framework to compare Self-hypnosis with Insight meditation that proposes brain processes information by generating and updating predictions about sensory inputs using Bayesian inference where perception is not a passive reception of stimuli, but an active construction based on prior knowledge and expectations. Assuming linear system dynamics and Gaussian noise, I propose that a Kalman filter model functions as an optimal observer subserving Insight meditation, priming the measurement model necessary for effective cognitive control in Self-hypnosis. This Bayesian measurement model is crucial for planning therapeutic cognitive control interventions in functional neurological disorders, which are hypothesized in this paper to be forms of maladaptive learning in adults.