Trustworthiness Judgments in Autism: An fMRI Study

Background: Previous studies have shown that individuals with autism spectrum disorder (ASD) do not judge faces for trustworthiness in the same way that neurotypical individuals do. Adolphs, Sears, & Piven (1998 & 2001) reported that individuals with ASD rated the least trustworthy faces as more trustworthy than neurotypical participants did. Interestingly, they found this same effect in individuals with amygdala brain damage. However, when participants rated written biographies, the authors found no difference between neurotypical participants and participants with ASD or amygdala damage patients, suggesting these atypical trustworthiness judgments are limited to faces. Thus, there appears to be a dissociation in the ability of individuals with ASD to judge faces or context for trustworthiness.

Objectives: This fMRI study investigated the neural activity of high functioning individuals with ASD and neurotypical individuals while they made trustworthiness judgments based on context information and face processing.

Methods: Participants are high-functioning adults with ASD and age and IQ matched neurotypical participants. Participants viewed a face on a computer for 3 seconds and had to answer yes or no to the question, “Would you trust this person?” They then read a story about the person and again saw the face for 3 seconds. Then they decided a second time if they would trust that person. Half the faces were trustworthy and half were untrustworthy, as judged by neurotypical participants in a preliminary ratings task. In half the stories, the reported actions of the character are positive (trustworthy) and in the other half, the actions of the character are negative (untrustworthy). The participants performed the task in a 3T Siemens Allegra functional magnetic resonance imaging scanner.

Results: The behavioral, functional imaging, and functional connectivity data from approximately 15 adults with ASD and 15 matched neurotypical adults will be presented.

Conclusions: It is anticipated that participants with ASD will make use of written context, but not visual information from faces, to judge trustworthiness, while neurotypical participants will use both types of information. It is predicted that this task will recruit activation in the Theory of Mind (ToM) network, including prefrontal cortex, right temporal parietal junction, the superior temporal sulcus, the fusiform gyrus, and the amygdala. It is also predicted that the individuals with ASD will recruit these areas to a different extent than the neurotypical participants. Furthermore, it is expected that neurotypical participants will activate the amygdala more when viewing untrustworthy faces as compared to trustworthy faces, but participants with ASD will not show this difference. Finally, we expect to find differences between our two groups in the functional connectivity between frontal areas and posterior areas while performing this task. These results will provide insight into the abilities of individuals with ASD to make trustworthiness judgments based on visual features of faces and context information. Furthermore, these results will illuminate the neural correlates underlying the differences between individuals with ASD and neurotypical individuals while performing trustworthiness judgments.