Autism Spectrum Disorder Is Characterized By Structural Under-Connectivity in Reward Circuitry

Background: The idea that the reward system dysfunctions play a central role in etiology of autism spectrum disorder (ASD) has gained increased interest over the last decade due to the potential of unifying currently disjoint characteristics of ASD such as social deficits and repetitive behaviors under a single theory. The brain regions associated with the reward circuitry have been repeatedly reported as being affected in ASD. However, the literature still lacks a systematic connectivity analysis of the reward circuitry in ASD, as well as changes in its integration with other circuitries corresponding to distinct neuropsychological mechanisms.  

Objectives: The aim of this work is to provide a systematic connectivity analysis of the reward system in ASD, by studying diffusion imaging based structural connectivity changes within the reward circuitry, as well as between the reward circuitry and several other circuitries.  

Methods: A dataset of 97 male participants with ADOS/expert clinician confirmed ASD (age: 12.6  2.9 years) and age-matched 75 male TDCs with diffusion tensor imaging (DTI) data were studied.  Data processing entailed generating connectivity maps (connectomes) of the entire brain, using a parcellation scheme of 82 cortical and subcortical gray matter regions (i.e. nodes of connectome). Network measures encoding connectivity changes at multiple levels including individual regions, clusters of regions, and specific functional subnetworks were calculated and compared between the groups using univariate parametric statistical tests.  

Results: Our results suggested significant structural connectivity differences related to subcortical and several cortical regions traditionally associated with the reward system. Almost all regions of the reward circuitry including caudate, putamen, pallidum, hippocampus, accumbens area, amygdala, and orbitofrontal cortex showed local connectivity differences. Connectivity was significantly weaker in the ASD group in the reward circuitry as a whole, and in its integration with circuitries associated with other neuropsychological systems, including those composing the social brain.  

Conclusions: Regions of reward circuitry were, both individually and as a whole, affected by ASD induced changes. Overall, our findings were in line with the social motivation theory of ASD, which suggests that ASD phenomenology is significantly associated with deficiencies in social reward mechanisms. These results strengthen speculation that developmental trajectories in ASD may be altered in those who develop ASD by early appearing reward deficiencies, such that social interactions are not reinforced typically. This may impact cognitive and social development. If Social motivation deficits precede fully syndrome ontogeny, interventions that boost social motivation, rather than aiming at the improvement of specific social skills, might be the most effective strategy.