Non-suicidal self-injury (NSSI) has been regarded as a dysfunctional coping strategy for managing intensely difficult feelings and is a growing area of concern in clinical and non-clinical populations alike. Individuals who engage in NSSI often report significant interpersonal difficulties, with studies showing that they have impaired social interactions. Attending to the emotional expressions of others is important for appropriate social functioning, and subtly mimicking the expressions of others is an unconscious behaviour that encourages empathy and interpersonal bonding. Differences in emotion recognition and reflection can impact social behaviour, yet little research has assessed how individuals with a history of NSSI (HNSSI) process facial expressions of emotion. In this thesis, the results of three studies designed to investigate potential differences in emotion recognition and emotion mimicry in individuals with a history of NSSI compared to controls are reported. Results from Study 1 revealed that when presented with colourful, dynamic morphing stimuli showing emotional expressions, HNSSI participants were able to correctly categorize negative expressions like anger, disgust, sadness, and the ambiguous emotion of surprise at a lower stimulus intensity compared to controls; They also correctly categorized fear with greater accuracy, though at the same intensity as controls. However, in Study 2, when static, greyscale images were obscured with varying levels of fractal noise, HNSSI individuals did not show superior signal-proportion thresholds. These results may suggest that higher-order elements of visual and cognitive processing are evoked by more realistic social stimuli. In the third study, HNSSI participants were found to produce significantly less electrical activity in the corrugator muscle in response to viewing angry stimuli, and significantly less of an expected relaxation in muscle activity in response to viewing happy stimuli. Complementing these results, it was found that endorsing social influence as a motivator for engaging in NSSI was associated with less mimicry, whereas endorsing emotion regulation as a motivator was associated with greater incongruent muscle response when viewing happy faces. These observed differences in facial mimicry between HNSSI and controls may be related to some of the observed relationship difficulties experienced by this group. Results from this research may help us to better understand NSSI behaviour, as well as help to inform and further develop therapies intended to address it.