The Neural Correlates of Probabilistic Classification Learning in Obsessive-Compulsive Disorder: A Pilot Study

Abstract

Individuals suffering from obsessive-compulsive disorder (OCD) have been found to show deficits in implicitly learning probabilistic associations between events. Neuroimaging studies have associated these implicit learning deficits in OCD individuals with aberrant activation of the striatal system. Recent behavioral studies have highlighted that probabilistic classification learning (PCL) deficits in OCD individuals only occur in a disorder-specific context, while PCL remains intact in a neutral context. The neural correlates of implicit learning in an OCD-specific context, however, have not yet been investigated. Using functional magnetic resonance imaging during a neutral (prediction of weather) and an OCD-specific variant (prediction of a virus epidemic) of a PCL paradigm, we assessed brain activity associated with implicit learning processes in 10 participants with OCD and 10 matched healthy controls. Regions of interest (ROIs) were the striatum and the medial temporal lobe. ROI analyses revealed a significantly higher activity in the bilateral putamen and the left hippocampus of OCD participants as compared to healthy controls during both PCL tasks. The group differences could partly be subsumed under a group × task interaction effect with OCD participants showing a significantly higher activity than healthy controls in the left putamen and the left hippocampus in the OCD-specific task variant only. These results suggest a compensation of aberrant striatal activity by an augmented engagement of the explicit memory system particularly in a disorder-relevant context in OCD participants.

What Is the Weather Prediction Task Good for? A New Analysis of Learning Strategies Reveals How Young Adults Solve the Task

“…Indeed, the WPT or the Ice Cream Task, a modified version of the WPT in which participants must predict the flavor of ice cream preferred by a cartoon figure based on the different accessories worn by that figure ( Shohamy et al, 2004a ), have been extensively used in clinical populations to evaluate the behavioral consequences of known or presumed structural or functional basal ganglia deficiencies: for example in schizophrenia ( Keri et al, 2000 ; Weickert et al, 2002 ; Foerde et al, 2008 ; Horan et al, 2008 ; Gomar et al, 2011 ; Karcher et al, 2019 ; Fernandez et al, 2021 ), Tourette syndrome ( Keri et al, 2002 ; Marsh et al, 2004 ), bulimia nervosa ( Labouliere et al, 2016 ), attention-deficit hyperactivity disorder ( Gabay and Goldfarb, 2017 ), Parkinson and Huntington diseases ( Knowlton et al, 1996a , b ; Shohamy et al, 2004a , b ), and children with acquired or developmental basal ganglia pathology ( Mayor-Dubois et al, 2010 ). Researchers have also used these tasks to assess memory capacities in individuals with neurodevelopmental disorders thought to be characterized by some procedural memory deficits, such as autism ( Brown et al, 2010 ; Obeid et al, 2016 ), obsessive–compulsive disorder ( Exner et al, 2014 ; Kelmendi et al, 2016 ; Hansmeier et al, 2018 ), specific language impairment ( Kemeny and Lukacs, 2010 ; Mayor-Dubois et al, 2014 ; Obeid et al, 2016 ), or developmental dyslexia ( Gabay et al, 2015 ). However, to properly interpret the results of investigations employing these tasks, it is critical to precisely define the cognitive processes actually engaged during their resolution, and not rely simply on the fact that overall task performance may be more or less impaired in individuals with certain pathologies.…”

“…In sum, despite the fact that several studies have shown that the WPT does not necessarily assess striatal-dependent implicit memory in healthy young adults ( Knowlton et al, 1996a ; Poldrack et al, 2001 ; Gluck et al, 2002 ; Hopkins et al, 2004 ; Foerde et al, 2006 , 2007 ; Lagnado et al, 2006 ; Meeter et al, 2006 ; Newell et al, 2007 ; Price, 2009 ; Li et al, 2016 ), this task is still used to determine whether specific populations exhibit impairments of this memory system ( Knowlton et al, 1996b ; Keri et al, 2000 , 2002 ; Weickert et al, 2002 ; Marsh et al, 2004 ; Shohamy et al, 2004a , b ; Foerde et al, 2008 ; Horan et al, 2008 ; Brown et al, 2010 ; Kemeny and Lukacs, 2010 ; Mayor-Dubois et al, 2010 , 2014 ; Gomar et al, 2011 ; Exner et al, 2014 ; Gabay et al, 2015 ; Kelmendi et al, 2016 ; Labouliere et al, 2016 ; Obeid et al, 2016 ; Hansmeier et al, 2018 ; Karcher et al, 2019 ; Fernandez et al, 2021 ). Moreover, knowing that different learning strategies can lead to above chance performance in the WPT, it is fundamental to determine how participants solve the task ( Ashby and Maddox, 2005 ).…”

Distinct Response Inhibition Patterns in Obsessive Compulsive Disorder Patients and Pathological Gamblers

“…This phenomenon was reported also in unaffected relatives of OCD subjects (71). In addition, OCD patients demonstrated impairment of implicitly learning probabilistic associations between events (72). Furthermore, flexibility is required when responses that previously associated with positive or negative outcomes should be switched to new opposite consequences.…”