Objectives: The purpose of the current study was to investigate interactions between neural responses to social and monetary rewards in ASD in a whole brain analysis, as well as to examine group differences specific to each reward type. This will help to elucidate whether deficits in reward processing in ASD are specific to social rewards.
Methods: 28 right-handed male participants with ASD and 26 age and IQ matched controls were recruited to the study. 21 participants from each group were included in the final analysis (age; ASD = 17.73 (3.39); CON = 17.00 (3.37): IQ; ASD = 109.38 (15.94); CON = 110.00 (12.53). Participants performed an adapted version of the Social and Monetary Incentive Delay Tasks where accuracy was fixed to ~70%. Three levels of reward were possible, no reward (blurred face/coin), small reward (face small smile/20 cent) and larger reward (face big smile/1euro). Slow responses to small/large cues were presented with no reward feedback. fMRI preprocessing and analysis was carried out in SPM8. Contrasts were generated to examine the effects of postive feedback (small/large reward) modelled over a baseline of correct responses with no reward feedback. An ANOVA was carried to investigate group (ASD/CON) by reward type (Social /Monetary) interactions. In addition two two sampled t-tests were used to exmine group differences specific to each reward type (uncorrected p<0.001; extent threshold 10; apriori hypotheses in the striatum was corrected using a bilateral caudate SVC p<.05 FWE).
Results: As hypothesised the results revealed a significant interaction between group and reward type in the striatum. The ASD group showed a reduction in activity from baseline in response to social rewards but an increase in response monetary rewards in the left dorsal caudate. Controls, on the other hand, showed increased activity from baseline for social rewards but no increase in activity for monetary rewards in this region. In addition, task specific group comparisons indicated that the ASD group showed reduced activity in the caudate for social rewards but no significant difference in this region for monetary rewards. There were no significant group differences for monetary rewards that survived correction for multiple comparisons in reward related neural circuitry.
Conclusions: The results suggest that the dorsal striatum may be maladaptive in ASD such that it is hyporesponsive to social stimuli whilst remaining responsive to other salient stimuli such as monetary rewards. This contrasts with previous studies which have shown reduced activity in the ventral striatum in response to monetary rewards in ASD. This may be accounted for by the different roles of the ventral and dorsal striatum in reward processing.