Objectives: The aims of the current study were threefold. First, to investigate whether there was evidence of atypical mu frequency activity during imitation in children with autism, compared to neurotypical controls. Second, of interest was whether intensive behavioural intervention that targeted imitative behaviours in children with autism altered response in the mu frequency range during imitation compared to children with autism who had not been taught to imitate. Third, the relationships between mu activity (and its task dependent attenuation) and imitative ability, and mu activity and the more general symptoms of autism, were explored.
Methods: A total of 37 children aged between 2.5 and 8 years (24 male, 13 female) participated, constituting 14 neuropsychiatrically and medically healthy children (control group), twelve children with ASD who had undertaken at least one month of intensive behavioural intervention (treatment group), and eleven children with ASD who had not commenced any behavioural intervention prior to their involvement in the study (non-treatment group). Low functioning subjects were included. Measures of autism symptom severity and intelligence were collected from all subjects. EEG recordings were obtained over the motor cortices using a bipolar montage placed at C3 and Cz (reference). Continuous EEG recording was conducted during an extended assessment of imitation ability following baseline recording.
Results: No significant attenuation of mu frequency activity was observed in either groups of individuals with autism. Only patterns of mu frequency activity in the autism- no treatment group were significantly different to that of typically functioning individuals, providing some suggestion that there was a tendency towards more typical mu frequency activity following intervention. Interestingly, additional analyses revealed that symptom severity was a stronger predictor of mu frequency attenuation and baseline mu power than was imitation. Further, the symptoms of autism found to be most strongly related to mu frequency activity during imitation were those associated with social functioning.
Conclusions: Targeting imitative behaviour in intensive early intervention may impact on neural circuitry underlying the development of this skill. Regarding low baseline mu power seen in individuals with ASD, at present, little is known about what baseline mu power signals, however, it is possible that it is a reflection of the connectivity in a network important for imitative tasks and that the attenuation seen during imitative behaviour (though to reflect neuronal desynchronisation) is a secondary indicator of how well connected this network is. Discrepancies between these and previous findings may be due to the inclusion of low functioning subjects in this study.