Abstract
Attention-Deficit/Hyperactivity Disorder (ADHD) is a disorder with core symptoms of overactivity, inattention, impulsivity, or a combination of the above. Many researchers have proposed models of ADHD, some of which attempt to explain atypical reward sensitivity, but fewer have addressed punishment sensitivity. The studies in this thesis aimed to examine behavioural sensitivity to reward and punishment in children with ADHD.
Control participants were recruited through Dunedin area schools, and clinical participants through referral from the Otago District Health Board. Both studies used the same participant pool. The first study used a signal detection task. Each trial presented an array of red and blue alien faces on a computer screen. Participants judged whether there were ‘more red’ or ‘more blue’ characters. Correct responses were occasionally followed by a multi-component reward. Initially, correct ‘more red’ responses were rewarded four times more often than correct ‘more blue’ responses. After 20 rewards, this distribution reversed, with correct ‘more blue’ responses rewarded four times more often than correct ‘more red’ responses. After another 20 rewards, the distribution reverted to the original arrangement, until the participant received 20 more rewards. Participants then experienced a period with no reward available. Response bias was measured during both phases. During the reinforced phase, both groups were sensitive to reinforcement, but children in the ADHD group displayed lower bias levels. Initially, both groups displayed bias toward the more frequently reinforced alternative. This reversed when the reinforcement contingencies reversed, and then switched again when the original arrangement was reinstated. This effect was larger for controls; children with ADHD seemed less able to track changes in the contingencies. During extinction, both groups continued responding, but control participants showed changes in bias similar to those displayed during the reinforced phase of the task, whereas children in the ADHD group showed consistent bias towards the alternative most recently associated with more frequent reinforcement, suggesting greater control exerted by the most recently received reward or specific reward instances. This is consistent with previous reward sensitivity research.
In the second study, the computer screen displayed two concurrent games. On each trial, participants clicked on one of the games, and experienced either a win, worth 10 points, a loss, which deducted 5 points, or no outcome. Wins were accompanied by a celebratory sound and short animation, and losses by “evil laughter”. The reinforcement schedule was equal across games, but Game 1 responses had a 16% chance of being punished, and Game 2 responses a 4% chance of being punished. Punishment appeared to exert greater control over the responding of the participants with ADHD than controls; they experienced fewer punishment outcomes and avoided the game associated with more frequent punishment more than the control group. This is inconsistent with some previous studies, which have found reduced punishment sensitivity in children with ADHD.
The findings from these studies suggest that children with ADHD may be less capable of tracking unsignalled changes to reinforcement contingencies, and they may also have increased punishment sensitivity.