Behavior Economic Measures of Social Reward in Children with Autism

Background:  

There is an emerging supposition that the social impairments that are part of the autism phenotype originate from a failure to experience social interactions as rewarding (Dawson et al., 1998; Dawson, Carver, Meltzoff, Panagiotides, & MacPartland, 2002; Dawson, Webb, & McPartland, 2005; Mundy, 1995; Rochat & Striano; Schultz, 2005). As such, there is a need for objective measures of social reward. Such a task is ideally suited to the subfield of psychology known as behavioral economics, the methods of which emphasize precise and direct quantification of reward value. Within this discipline, reward value is commonly measured based upon how hard an individual will work for a particular outcome. In the case of social reward, children with ASD would be expected to work less at a task for the opportunity to engage in social interactions than would typically developing (TD) children. Progressive ratio (PR) methodologies examine an individual’s consumption of a reward at various “prices”, which are represented by the number of times they must perform a task for the reward. Data from PR procedures are examined in terms of consumption at several key points along the price continuum. For example, the break point is the most an individual will work for a reward. O_max is the point along the price/responding function at which working begins to decelerate as a result of increases in price. 

Objectives:  

The present study employed a PR arrangement to compare the degree to which children with autism and TD peers valued social attention. 

Methods:  

Fourteen children, ages 3 - 8 participated, 5 with an ASD diagnosis and 9 TD individuals. Each participant completed a PR procedure that involved working at a task (i.e., pressing a microswitch). The procedures compared each participant's performance on a PR arrangement for two different types of rewards: 30 s of attention from a novel adult, or access to a highly preferred item. The number of tasks that had to be completed increased within each series until one minute elapsed without any task completion.

Results:  

Four out of five participants with autism showed higher breakpoints and Omax values for preferred items than for the social attention. On average, participants in the ASD group emitted 732% more responding towards preferred item than attention. This pattern was reversed in the TD group, with seven of nine showing higher breakpoints and Omax values for attention than preferred items.  On average, participants in the TD group emitted 460% more responding for attention than for preferred items.

Conclusions:  

Behavioral economic measures of social reward effectively quantified the degree to which individuals with autism value social reward. This method effectively differentiated between individuals from ASD and TD groups, and offers a level of precision in quantification of social reward that is much greater than in existing measures.