Emotion Attribution to Self and Other In Children and Adolescents with and without Autism: An fMRI Study

Background:

The medial prefrontal cortex (mPFC) appears to be preferentially tuned for self-relevant information. In typical children and adolescents the mPFC seems to be even more engaged during self-referential processing compared to adults (Pfeifer et al., 2007). Recent studies indicate that in autism the mPFC seems not to distinguish between self-relevant or other-relevant information, which is associated with symptom severity (Lombardo et al.,2010, Kennedy & Courchesne, 2008). In autism, it remains largely unknown how self- and other-referential processing networks change over development and how these networks are implicated in adaptive social functioning, empathy and self-referential cognition.

Objectives:

Using fMRI, we are measuring mPFC activity during affective judgments in children with autism, while accounting for individual differences in adaptive social functioning, empathy and self-referential cognition.

Methods:

Children with autism (n=4, collection ongoing) and matched controls (n = 25), were instructed to perform an emotion attribution task, in which a series of emotional pictures were presented. The children were required to make judgments under three conditions. In the Self condition, they evaluated whether the picture made them feel good or bad. In the Other condition, they evaluated whether the person or people in the picture felt good or bad. In the Location condition, they evaluated whether the picture was taken inside or outside. The Social Responsiveness Scale (SRS), the Empathy Quotient (EQ), the Interpersonal Reactivity Index (IRI), the Toronto Alexithymia Scale (TAS-20) and a Self-Referenced Memory Task were used to investigate social functioning, empathy and self-referential cognition. Eye-tracking was used to monitor looking patterns.

Results:

Two independently defined regions of interest in the mPFC (one defined by functional connectivity to a precuneus seed, the other based on an Activation Likelihood Estimation computation) were used for further quantitative tests. In typical controls, both regions exhibited the pattern of Self > Other > Location. Activation in these areas increased with age, while increasing activation was associated with less social impairment. The data collection and analysis of the matched group of children with autism is ongoing. We hypothesize that in children with autism the mPFC will show no significant difference in modulation to the Self condition compared to the Other condition consistent with previous findings in adult samples.

Conclusions:

The findings of this cross-sectional study suggest an increasing self-selectivity in the mPFC during typical development in respect to affective judgements. This increased self-selectivity seems to be socially adaptive in children and adolescents and might represent an increase in self-awareness. Conversely, in autism hypothesized reduced self-selectivity in the mPFC in respect to affective judgements could represent a lack of self-awareness associated with abnormal social functioning and deficits in empathy and self-referential cognition. In addition heightened self-referential processing might seem socially adaptive especially at particular points in development (i.e. adolescence), while it could act at the same time as a vulnerability factor for psychiatric disorders such as depression and anxiety disorders. The identification of specific neural correlates in the mPFC of behavioural symptoms in autism could help elucidate the mechanisms of this heterogeneous disorder.