A Simultaneous EEG, TMS and Eye-Tracking Study Investigating Mirror Neuron System Activity in Adults with Autism Spectrum Disorder When Inferring Intentions.

Background: Autism spectrum disorder (ASD) is associated with difficulties inferring the internal states of others such as their intentions, beliefs or mental states. These are known as mentalizing tasks. Neuroimaging studies have found mirror neuron system (MNS) activity during these tasks and it has been hypothesised that MNS dysfunction may underlie mentalizing difficulties associated with ASD. However, MNS activation during mentalizing tasks may be simply due to the use of stimuli depicting biological motion and evidence regarding MNS dysfunction in adults with ASD is limited. This study investigated whether MNS activity was higher when mentalizing than in a control task when the same stimuli were used, and whether MNS functioning was atypical in adults with ASD.

Objectives:  We used EEG and single-pulse TMS techniques to identify potential differences in MNS activity in high-functioning adults with ASD when inferring intentions from hand actions.

Methods:  Thirteen adults with ASD and thirty control participants (15 low AQ, 15 high AQ) watched videos of actors performing hand actions. Participants were either asked to infer the actors’ intentions (mentalizing task) or the success of the action (control task). TMS-induced motor evoked potentials (MEPs) and mu suppression measured by EEG were both used as indices of MNS activity. Eye-tracking data were also collected in order to identify any differences in fixation patterns associated with ASD.

Results:  The EEG data showed lower levels of mu suppression during the mentalizing task across all groups. TMS-induced MEPs were largest when participants watched actions in which the intentions of the actors were not fulfilled (clumsy/accidental actions) even when participants were not directly asked to infer intentions. No differences in MNS activity associated with ASD were found. Eye-tracking data showed participants with low AQ scores looked at the head of actors more often and for longer during the mentalizing task but adults with high levels of autistic traits both with and without a diagnosis did not.

Conclusions:  Our eye-tracking data imply that adults with high levels of autistic traits (with and without a diagnosis) do not alter their fixation patterns across mentalizing and non-mentalizing tasks in the same way as adults with low autistic traits. Despite this, no differences in abilities to infer the intentions of others or MNS activity were found associated with ASD. The TMS data suggest the MNS has a role in coding the consistency of a person’s actions with their initial intentions, consistent with the predictive coding theory (Kilner, Friston & Frith, 2007). However, the EEG data provide evidence against MNS involvement in mentalizing tasks. Therefore TMS-induced MEPs and mu suppression appear to measure different aspects of MNS activity, consistent with previous studies (e.g. Lepage, Saint-Amour & Théoret, 2008).

Keywords: Transcranial magnetic stimulation (TMS), electroencephalography (EEG), mentalizing, eye-tracking