Autism is a disorder manifest in impairment of complex social behaviours, which are likely to stem from impairment of basic mechanisms of social learning such as imitation. We have recently developed a novel method for objective, quantifiable measurement of manual (Culmer, 2009) as well as facial (Williams, in press) imitation fidelity outside of the scanning environment.

The manual task requires the comparison of kinematic parameters of actions between model and observers recorded through the use of touchscreen technology. The high levels of accuracy and precision that are possible in these measures have enabled us to show that relatively small differences between typical individuals correlate with BOLD signal differences during simple imitation in brain regions serving social cognitive functions that include insula, ventromedial prefrontal cortex and frontal pole (Braadbaart, 2012). Our third piece of pilot work to prepare for this project has been the development of touchscreen technology adapted for the MRI scanning environment.

The facial imitation task is developed in collaboration with the Perception Lab at the University of St Andrews. It requires manipulated facial expressions as novel stimuli to imitate; participant's imitative attempts are quantified through blind rating. Another behavioural measure taken into account is empathy. These measures are then correlated to BOLD signal differences in response to the facial imitation stimuli viewed in the scanner, to determine which cognitive processes are involved in facial imitation ability and how it interacts with empathy.

This project’s goal is to incorporate our objective measures of imitation fidelity into an fMRI experiment to compare directly the neural correlates of social learning between a group of young people with autism and a group of control participants. The project offers a route to developing clinically applicable measures of neurodevelopmental impairment, validated using functional neuroimaging.