Reduced Preference for Social Rewards in Toddlers with Autism: Relations with Symptoms and Treatment Response

Background: Research indicates that Autism Spectrum Disorder (ASD) is associated with reduced sensitivity to social rewards. Children with ASD vary in terms of their degree of preference for social versus nonsocial rewards, and it is expected that variations in such preferences might correlate with symptom severity, especially social impairments. Furthermore, given that early intervention utilizes social rewards to motivate learning and skill acquisition, variation in social reward preference might predict rate of skill acquisition and outcome in early intervention research.

Objectives: The purpose of the current study is to assess, via parent questionnaire, preferences for social versus nonsocial rewards in toddlers with ASD, and assess the relations between variation in such preferences and symptom severity and response to early intervention.

Methods: The current sample consists of 48 children diagnosed with ASD (M CA = 23.5 months; SD = 4.9, range 18 to 30 months; 37 M, 11F) who participated in a randomized clinical trial of early intensive behavioral intervention, based on the Early Start Denver Model (ESDM). Participants met diagnostic criteria for Autistic Disorder or Pervasive Developmental Disorder-Not Otherwise Specified (PDD-NOS) as measured by the ADI-R and ADOS-WPS as well as DSM-IV criteria by an experienced clinician. At baseline, parents completed a 201-item reinforcer questionnaire (RQ) to assess their children’s preference for a wide range of social and nonsocial reward stimuli. The questionnaire used a Likert scale, with scale values ranging from 0 (not rewarding) to 4 (extremely rewarding). Additionally, parent reports of adaptive behavior and observational measures of joint attention, social orienting, and symptom severity were collected. To examine the relationship between items endorsed on the RQ and participant characteristics, the items from the RQ were sorted into four categories (social, nonsocial activities, tangibles, and edibles) and correlation analyses were conducted.

Results: According to parent report, young children with ASD are more likely to prefer nonsocial, as compared to social, reward stimuli. Nonsocial activities, such as riding in the car, playing with glue, listening music, or lining up objects, were most frequently endorsed, followed by edible stimuli. Children who showed the greatest preference for nonsocial stimuli, as compared to social stimuli, showed more significant impairments in social orienting skills, whereas those children with greater preferences for social stimuli had significantly better social orienting skills. Further analysis will examine the relationship between RQ categories and additional child characteristics, including response to early behavioral intervention after 2 years.

Conclusions: This is the first report, to our knowledge, to assess the relation between variations in social reward preference and degree of autism symptom severity. Results show that young children with ASD tend to prefer nonsocial as compared to social stimuli, and that variation in their degree of preference for social stimuli is correlated with symptom severity, as reflected in social orienting, joint attention, and other domains. Future analyses will explore whether variation in preference for social rewards moderates response to early behavioral intervention. Knowledge of children’s preference for social versus nonsocial stimuli can be useful in designing individualized strategies for intervention.