Habit Learning in OCD: Preliminary Data from a Spanish Sample

Abstract

Introduction Instrumental learning involves goal-directed and habitual systems. The Slips-of-Action Task (SOAT) is extensively used to measure habit tendencies and the likelihood of making erroneous responses for devalued outcomes. The SOAT provides a Devaluation Sensitivity Index (DSI), a measure of the balance between relative goal-directed and habitual learning. Individuals with Obsessive-Compulsive Disorder (OCD) often engage in repetitive actions, suggesting a potential deficit in goal-directed control and an increased reliance on habitual learning. Previous literature has shown that medicated OCD adults performed worse on the SOAT task than healthy controls. Objectives To compare habit learning performance in an unmedicated sample: - Goal 1: Between OCD and Healthy Controls (HC) - Goal 2: Across four groups: adult OCD, adult HC, children OCD, and children HC Methods Participants: Eighty-three participants (44 OCD patients and 38 healthy controls) completed the study with usable task data. The 44 OCD patients comprised 17 adults (mean age: 26.76 years, SD: 8.61 years) and 27 children/adolescents (mean age: 12.84 years, SD: 2.59 years). The 38 healthy controls included 17 adults (mean age: 30 years, SD: 7.49 years) and 21 children/adolescents (mean age: 14.1 years, SD: 2.19 years). All participants were unmedicated. Measures: Participants completed an adapted version of the “Fabulous Fruit Game”, which included an instrumental training phase to learn Stimulus-Response-Outcomes (S-R-O) associations and a SOAT to assess the strength of learned S-R-O associations. DSI was calculated by subtracting the percentage of responses made toward devalued outcomes from the percentage of responses made toward still valuable outcomes. Behavioral Analyses: Student’s t-test comparing individuals with OCD to HC and a ONEWAY ANOVA to examine group differences across multiple categories. Results Goal 1: DSI comparison between individuals with OCD and HC revealed a significant difference, with HC demonstrating superior performance (t (60.9) = 2.60, p = .012, Cohen’s d = .546). Goal 2: The overall DSI comparison across adult OCD, adult HC, children OCD, and children HC showed a non-significant difference (F(3) = 3.407, p = 0.22). However, post hoc analysis revealed significant differences between Adult HC and Youth OCD (I-J Scheffe = 28.82, p = .033), indicating superior performance in adult HC. Conclusions This study highlights altered Habit Learning in unmedicated OCD individuals, supported by significant DSI differences compared to HC. Age-related distinctions were observed, emphasizing the need for age-sensitive interventions in understanding and addressing habit-related challenges in OCD. Disclosure of Interest None Declared