TOO MUCH DELIBERATION? CAUTIOUS DECISION-MAKING IN OCD

Abstract

abstract Introduction Compulsive behaviours are typical symptoms of Obsessive-Compulsive Disorder (OCD) which may reflect difficulties to commit to ultimate decisions and may be conceptualized as a means to accumulate sufficient evidence prior to a decision. Here we investigate the process of evidence accumulation in OCD in perceptual discrimination, hypothesizing greater evidence accumulation in OCD patients. Methods Twenty-eight OCD patients (16 F/12 M) and 35 healthy control (20 F/15 M) subjects were tested with a low-level visual perceptual task (random dot motion task), whereby different coherent levels for motion were defined to measure high and low uncertainty along with two response conflict tasks as control tasks (Flanker task and reinforcement learning conflict task). Regression analysis across all coherence levels (which accounted for visual detection threshold) and hierarchical drift diffusion modeling (HDDM) [1] were used to characterize response strategies between patients with OCD and healthy controls in the random dot motion task. Results Behavioural and computational modelling approaches show that patients required more evidence under high uncertainty perceptual contexts, as indexed by longer response time intercepts and higher decision boundaries. HDDM, which defines a decision when accumulated noisy evidence reaches a criterion level (a decision boundary) [2] , further showed slower drift rate towards the decision boundary reflecting poorer quality of evidence entering the decision process in patients under low uncertainty. With incentives to emphasize speed, OCD patients lost the previous effect of uncertainty on reaction time and reversed the difference in decision boundaries, accumulating less evidence in low uncertainty trials compared to healthy volunteers, without compromising accuracy. There were no significant differences between groups in the confidence ratings and in reaction time after receiving positive or negative feedback in the random dots task. There were also no significant differences between groups on accuracy and reaction time measured by the control tasks. Our findings in the perceptual task were unrelated to high-level visual perceptual deficits and response conflict. Conclusions This study extends the literature of evidence gathering in OCD from probabilistic to perceptual decisions. We provide evidence for impaired decision-formation processes in OCD, with a differential influence of high and low uncertainty contexts on evidence accumulation (decision threshold) and on the quality of evidence gathered (drift rates). We further emphasize that OCD patients are sensitive to monetary incentives in heightening speed in the speed-accuracy tradeoff, improving evidence accumulation and shifting away from pathological internal monitoring. Finally, our lack of sensitivity to external feedback in OCD subjects highlights the differential role of implicit cost and external feedback in decision formation in OCD. These findings may have relevance for therapeutic approaches.