The Computational Anatomy of Psychosis

Abstract

This paper considers psychotic symptoms in terms of false inferences or beliefs. It is based on the notion that the brain is an inference machine that actively constructs hypotheses to explain or predict its sensations. This perspective provides a normative (Bayes-optimal) account of action and perception that emphasizes probabilistic representations; in particular, the confidence or precision of beliefs about the world. We will consider hallucinosis, abnormal eye movements, sensory attenuation deficits, catatonia, and delusions as various expressions of the same core pathology: namely, an aberrant encoding of precision. From a cognitive perspective, this represents a pernicious failure of metacognition (beliefs about beliefs) that can confound perceptual inference. In the embodied setting of active (Bayesian) inference, it can lead to behaviors that are paradoxically more accurate than Bayes-optimal behavior. Crucially, this normative account is accompanied by a neuronally plausible process theory based upon hierarchical predictive coding. In predictive coding, precision is thought to be encoded by the post-synaptic gain of neurons reporting prediction error. This suggests that both pervasive trait abnormalities and florid failures of inference in the psychotic state can be linked to factors controlling post-synaptic gain – such as NMDA receptor function and (dopaminergic) neuromodulation. We illustrate these points using biologically plausible simulations of perceptual synthesis, smooth pursuit eye movements and attribution of agency – that all use the same predictive coding scheme and pathology: namely, a reduction in the precision of prior beliefs, relative to sensory evidence.

Losing Control Under Ketamine: Suppressed Cortico-Hippocampal Drive Following Acute Ketamine in Rats

“…Extrinsically, our connectivity results were predicted by theoretical considerations (Adams et al, 2013) and corroborate a recent fMRI-based analysis showing increased cortico-cortical and CA1-PFC interactions following ketamine in rat (Gass et al, 2014). Corlett et al (2011) place the multi-faceted effects of ketamine on network interactions within a predictive coding framework, prescribing a Bayesian model by which synaptic dysfunction can be linked to symptoms of psychotic illness.…”

“…These synaptic effects may underlie a functional cortico-hippocampal dysconnection that is dominated by overly precise bottom-up prediction error signals. This fits comfortably with a failure to attenuate sensory precision and consequent false inference (and resistance to illusions) that attends the trait abnormalities of schizophrenia (Adams et al, 2013).…”

“…Individual differences in the expression of this pathophysiology may explain the spectrum of symptoms, clinical outcome, and treatment response across schizophrenic patients . From a predictive coding perspective (Dolan et al, 1999;Fletcher and Frith, 2009;Adams et al, 2013), a malfunction of neuromodulation (eg, dopaminergic transmission) is expected to cause a shift in the relative precision of regional inputs, resulting in a predominance of bottomup, sensory-driven signals, which report sensory prediction errors (Adams et al, 2013). This shift has been proposed as an early (prodromal) deficit that underlies abnormal learning over longer-time periods, in order to accommodate (explain away) unexpected bottom-up inputs (Corlett et al, 2011).…”

On the functions, mechanisms, and malfunctions of intracortical contextual modulation

“…The evidence reviewed here is in broad agreement with the computational model of psychosis presented by Adams, et al (2013). They present strong grounds for the view that many schizophrenic abnormalities, including delusional and hallucinatory inferences, arise from aberrant encoding of precision, i.e.…”

Visual mismatch negativity (vMMN): A review and meta-analysis of studies in psychiatric and neurological disorders

“…Predictive coding provides a framework for theoretical models of perceptual inference, cognition, learning and decision making (Friston, 2005(Friston, , 2010den Ouden et al, 2012), and it might provide visual MMN research with a principled probabilistic approach to test neurophysiologically grounded hypotheses in clinical conditions (Adams, Stephan, Brown, Frith, & Friston, 2013;Corlett, Honey, Krystal, & Fletcher, 2011;Corlett et al, 2007;Friston et al, 2014;Stephan, Bach et al, 2016;Stephan et al, 2006;Stephan, Binder, et al, 2016;Stephan et al, 2009). Thus, it might help to understand better disease mechanisms and dissect heterogeneous clinical groups into well-defined subgroups to guide diagnosis, predict response to treatments or conversion to psychosis, or track disease progression.…”