Obsessive–compulsive disorder as a disorder of attention

Abstract

An influential model holds that obsessive–compulsive disorder (OCD) is caused (in part) by distinctive personality traits and belief biases. But a substantial number of sufferers do not manifest these traits. I propose a predictive coding account of the disorder, which explains both the symptoms and the cognitive traits. On this account, OCD centrally involves heightened and dysfunctionally focused attention to normally unattended sensory and motor representations. As these representations have contents that predict catastrophic outcomes, patients are disposed to engage in behaviors and mental rituals designed to forestall these outcomes or to produce more precise information. The same representations also cause the cognitive traits characteristic of sufferers.

Searching for an anchor in an unpredictable world: A computational model of obsessive compulsive disorder.

“…Fourth, recently, Levy (2018) has suggested that OCD is related to increased precision for sensory PEs and actions, leading to increased attention to normally automatic processes. Whereas we concur with the general idea, we argue that this is the specific result of excessive uncertainty regarding state transitions.…”

Obsessive Compulsive Disorder: A Pathology of Self-Confidence?

“…Neil Levy has recently suggested OCD may be due to an imbalance in the relative influence of top-down predictions and incoming sensory information [7]. In OCD, the agent anticipates threats such as contamination.…”

“…Based on extant dual-process theories, such findings have been interpreted as evidence for distinct learning mechanisms underlying implicit and explicit evaluations: automatic formation of associative links between co-occurring events (e.g., associative link between A and B) and controlled generation and truth assessment of mental propositions about the relation between co-occurring events (e.g., A prevents B). However, the observed dissociation may also reflect differences in the retrieval of stored propositional information, given that (i) implicit and explicit evaluations differ in terms of their relative speed and (ii) fast evaluations are more likely affected by incomplete retrieval of stored information (e.g., retrieval of A is related to B rather than A prevents B) [7]. Thus, different from the argument that the observed dissociation provides evidence for functionally distinct learning mechanisms, it can be explained by retrieval-related processes without any assumptions about distinct learning mechanisms or distinct memory systems.…”

A Psychophysical Window onto the Subjective Experience of Compulsion

“…Moreover, we show that intentional binding can pit two ideas that are fundamental to current sensorimotor theories of compulsion against each other. These are the idea of excessive conscious monitoring of action in OCD [ 8 , 9 , 10 , 11 ] and the idea that patients with OCD compensate for diminished conscious access to “internal states”, including states of the body, by relying on more readily observable proxies [ 8 , 12 ]. These ideas make distinct predictions regarding the cause of an altered experience of action and agency in OCD, and ultimately differ in their implications for therapeutic approaches.…”