The Predictive Coding Account of Psychosis

Abstract

Fueled by developments in computational neuroscience, there has been increasing interest in the underlying neurocomputational mechanisms of psychosis. One successful approach involves predictive coding and Bayesian inference. Here, inferences regarding the current state of the world are made by combining prior beliefs with incoming sensory signals. Mismatches between prior beliefs and incoming signals constitute prediction errors that drive new learning. Psychosis has been suggested to result from a decreased precision in the encoding of prior beliefs relative to the sensory data, thereby garnering maladaptive inferences. Here, we review the current evidence for aberrant predictive coding and discuss challenges for this canonical predictive coding account of psychosis. For example, hallucinations and delusions may relate to distinct alterations in predictive coding, despite their common co-occurrence. More broadly, some studies implicate weakened prior beliefs in psychosis, and others find stronger priors. These challenges might be answered with a more nuanced view of predictive coding. Different priors may be specified for different sensory modalities and their integration, and deficits in each modality need not be uniform. Furthermore, hierarchical organization may be critical. Altered processes at lower levels of a hierarchy need not be linearly related to processes at higher levels (and vice versa). Finally, canonical theories do not highlight active inference-the process through which the effects of our actions on our sensations are anticipated and minimized. It is possible that conflicting findings might be reconciled by considering these complexities, portending a framework for psychosis more equipped to deal with its many manifestations.

Keywords: Bayesian brain; Cognition; Delusions; Hallucinations; Learning; Perception; Predictive coding; Schizophrenia.

Figure 1

Schematic illustration of Bayesian predictive coding as an explanatory framework for psychosis. (A) Predictions are encoded at higher levels of a hierarchical system and are sent as predictive signals to lower levels (downward arrows on the left). Whenever the incoming sensory data violate these predictions, a prediction error signal is sent to update the predictive model at higher levels (upward arrow on the right). Formalized as Bayesian inference, predictions (prior) and sensory data (likelihood) are represented in the form of probability distributions. The posterior results from the combination of prior and likelihood according to Bayes’ rule, weighted by their respective precisions π (which is the inverse of their variance σ; see first equation), and updates the predictive model (third equation). The fourth equation rearranges the third to show that the new posterior mean is simply the old prior mean added to a precision-weighted prediction error. (B) In psychosis, the balance between predictions and sensory data has been proposed to be disrupted, with a decreased precision in the representation of priors and increased precision of the likelihood . This imbalance biases Bayesian inference toward the likelihood and away from the prior, resulting in the abnormally strong weighting of prediction error. Candidate mechanisms for decreased prior and increased likelihood precisions are hypofunction of glutamatergic N-methyl-D-aspartate receptors (NMDA-Rs) and increased dopamine (DA) activity, respectively. Some psychotic phenomena may be explained by a compensatory increase in feedback signaling at higher levels of the hierarchy (bold arrow, upper left).