21732
Reduced Reward Related Response to Imitation in Autism Spectrum Conditions

Thursday, May 12, 2016: 2:21 PM
Room 310 (Baltimore Convention Center)
B. Chakrabarti1, J. Neufeld1,2 and C. T. Hsu1, (1)School of Psychology & Clinical Language Sciences, University of Reading, Reading, United Kingdom, (2)Center of Neurodevelopmental Disorders, Karolinska Institutet, Stockholm, Sweden
Background:

Imitation facilitates the formation of social bonds in humans, from an early age. Parents routinely imitate their babies to build rapport. Neurotypical humans find others who imitate them to be more rewarding, and imitate those who are more rewarding to them. These phenomena suggest a link between imitation and reward. In separate EMG and fMRI studies in individuals with and without Autism Spectrum Disorders (ASD), we have previously demonstrated that autistic traits modulate imitative responses to rewarding social stimuli. These results suggest that autistic traits modulate the link from reward to imitation. However, the link from imitation to reward has not been studied in ASD. Investigating this link is vital in light of the crucial role of imitation as an instrument of social bonding. On the basis of our previous results showing an atypical reward-imitation link in individuals with high autistic traits and a clinical diagnosis of ASD, we hypothesised that being imitated may not be equally rewarding for individuals with ASD.

Objectives:

To test if individuals with ASD show a reduced reward-related neural response to being imitated.

Methods:

26 ASD and 30 neurotypical (NT) adults, matched for age, gender, and IQ, performed a 2-part task. The first part was a conditioning task outside the scanner, where participants were mimicked by one face and ‘anti-mimicked’ by another. The participants were instructed to make a facial expression (happy/sad) while watching faces on screen that would either make a congruent or an incongruent expression, 700ms after the participant initiated the instructed expression. This created the subjective impression of being mimicked/ anti-mimicked. The second part was done in a 3T fMRI scanner, where participants passively viewed neutral expressions of the conditioned faces one at a time, using an event-related design. We hypothesised that the NT group would show greater reward-related ventral striatal (VS) response to the mimicking compared to the anti-mimicking faces, when compared to the ASD group. The VS cluster was defined using an independent meta-analysis of reward processing studies (Liu et al, 2010).

Results:

Data were analysed using SPM8, using a standard processing pipeline. Contrast statistics for [mimicking>anti-mimicking faces] were extracted from the VS cluster, and analysed using multiple regression. This analyses showed a main effect of group in the left VS (beta [group] = -0.662, p = 0.0265). Specifically, the left VS response to [mimicking>antimimicking faces] was found to be significantly lower in the ASD group compared to the NT group (t = -1.69, p = 0.0486). The left VS response negatively correlated with self-reported autistic traits (as measured using the Autism Spectrum Quotient, AQ) across the whole sample (r = -0.29, p = 0.0307). No significant differences were observed in the right VS.

Conclusions:

The results support our hypothesis that in individuals with high autistic traits and those with a clinical diagnosis of ASD, being imitated is associated with lower reward-related striatal response when compared to controls. This result offers a potential explanation for why individuals with ASD find it difficult to form social bonds.