42 Behavioral Correlates of Action-Control in Children with Attention-Deficit/Hyperactivity Disorder

Abstract

Objective: Children with attention/deficit-hyperactivity disorder (ADHD) exhibit motivational and cognitive impairments that affect daily life functioning. These impairments may reflect a deficit in action-control; the process by which voluntary actions are selected and executed based on prior reinforcement learning. It consists of parallel opposing processes; goal-direction and habit formation. Using the outcome-devaluation paradigm, we previously showed that children with ADHD rely on reflexive habitual, at the expense of goal-directed, behavior to deploy their actions. The current study investigates action-control using a contingency degradation paradigm, which involves outcome overvaluation as opposed to outcome devaluation. We hypothesize that children with ADHD will display a habitual behavior, while healthy controls (HC) will use goal-directed behavior to control their actions. Participants and Methods: We tested 19 ADHD and 14 HC participants (age 6-10 years) for this study. Children with ADHD were recruited from Children’s Specialized Hospital and underwent a structured clinical diagnosis. All participants were screened for ADHD and other neurologic or psychiatric conditions that could contribute to attention impairment using the SNAP-IV rating scale. Participants completed a set of the Woodcock-Johnson® IV assessments. They were tested using an outcome-overvaluation computer-based task. During learning, participants acquired stimulus-reward associations in the acquisition phase, as well as the overvaluation phase. In the latter, one of the rewards was delivered in a similar contingency to the acquisition phase (valued), while the other reward was randomly accompanied by an extra reward in 10% of the trials (overvalued). After the overvaluation phase, participants were presented with two stimuli (associated with a valued, and an overvalued outcome) and were asked to choose one stimulus in extinction. Choosing the overvalued at a higher rate was assigned as goal-directed behavior, while choosing both stimuli at the same rate was assigned as habitual behavior. Results: Independent-samples t-test showed that children with ADHD scored significantly higher than HC in the following measures: ADHD_inattention, ADHD_hyperactive/Impulsive, ADHD_combined, inattention/overactivity, Conner’s index, inattention domain, hyperactive/impulsive domain, and general anxiety disorder screening (P-value for all is <0.001). Results from the computer-based task showed that both groups acquired action-outcome associations during the first two phases of the task. During the extinction phase, HC, as compared to ADHD, responded at a higher rate on the stimuli that were associated with the overvalued outcome (t(31)=2.1, p=0.043); indicating higher tendency to show goal-directed behavior. Further, paired-samples t-test showed that there was no significant difference between response rate on the valued vs. overvalued stimuli in the ADHD group (t(18)=1.027, p=0.318), while there was a difference trending towards significance in response rate in the HC group (t(13)=-2.00, p=0.067). These results show that ADHD responded habitually, while HC responses were goal-directed. Conclusions: Our results indicate that children with ADHD are less likely than HC to engage in goal-directed behavior as opposed to habitual responding. This is consistent with our previous research highlighting a deficit in action-control in ADHD.