Abstract:
Understanding emotions from facial expressions is an essential component of social interaction. Mental simulation, a neural mimicry mechanism, has been proposed as the neural process that allows action and emotion understanding. Single-cell and brain imaging research indicates that a category of motor neurons, called mirror neurons, that are activated during both action execution and action observation may be the neural substrate for this simulation. Sensorimotor cortex plays a key role in mental simulation. Electroencephalography (EEG) studies in the field of action recognition have investigated sensorimotor oscillations in the alpha (8 - 13 Hz) and beta (14 - 25 Hz) bands, called the mu rhythm, to improve our understanding of the shared neural circuitry associated with action execution and observation. EEG mu studies which reported suppression in the mu rhythm during action execution, observation and imagery have associated mu modulation with the activity of the mirror neuron network (MNN). Recent findings from the MNN research indicate that action simulation extends beyond crude motor mimicry, and reflects higher order cognitive processing, including understanding other minds. The current thesis provides an in-depth investigation of the relation of the mu rhythm to understanding emotions from faces. The thesis reports six studies on the activity of the mu rhythm during observation of different types of facial movements, and the link between mu modulation, empathic ability and autism traits in typical individuals and individuals with autism. Study 1 showed that mu suppression (8 - 13 Hz) was greater while viewing dynamic images depicting mouth opening compared to non-biological movement, but there was no difference in mu suppression levels between kaleidoscope and happy or sad face images. This result was in the opposite direction to the initial hypothesis. We concluded that easily recognisable happy or sad faces do not activate the mirroring mechanism to a similar extent as unusual/ambiguous movement. Study 2 followed up on these unexpected results, and investigated mu modulation associated with categorising happy, sad and neutral static images of faces as happy or sad. As predicted, greater mu suppression was observed in the neutral compared to happy or sad face conditions. These findings suggest that unusualness/ambiguity of social information might evoke greater sensorimotor activity compared to usual/unambiguous stimuli to aid the meaning-making process. In both studies, occipital alpha activity was controlled for in order to ensure that mu rather than alpha modulation was captured. EEG data from these two experiments were used in the following three studies. Study 3 investigated hemispheric laterality of mu modulation, and the link between lateralised mu modulation and self-reported empathic ability. There was no effect of hemisphere on mu modulation, and no relation between lateralised mu modulation and empathy, except a marginal correlation between empathy and right-hemisphere mu suppression in the dynamic happy face condition. Study 4 showed that the results from the alpha component of the mu rhythm could not be extended to the beta component. Study 5 was a pilot study in which the mu modulation values of individuals with an autism spectrum disorder (autism) was compared to the group average of the neurotypicals. Findings from the present analysis and some past research suggest that the broken mirror neuron hypothesis of autism or other traditional cognitive theories of the condition cannot address the heterogeneity in autism or the dimensionality of psychopathology. Study 6 provide further support for a dimensional approach to mental health by demonstrating that greater severity of self-reported autism traits was associated with lower self-reported empathic ability. We apply the predictive coding account of brain function to emotion understanding, and discuss the relevance of our findings to a relational conceptualization of brain function and mental health.