Empathy describes the phenomenon of feeling and understanding another's internal states. It has been broadly conceptualised as consisting of two dimensions, one cognitive and one emotional. The cognitive dimension is top-down driven and encompasses perspective taking, while the emotional dimension is bottom-up driven and encompasses emotional contagion. In other words, cognitive empathy refers to knowing and understanding someone else's internal states, while emotional empathy refers to feeling with and for someone else. Therefore, empathy is a phenomenological representation of related, yet distinct processes. While many studies have examined the factors that influence and relate to empathy, both as a stable personality trait and momentary, state empathic responses, only a limited number of recent studies have examined the effect of social power on empathy. However, the idea that having social power may affect one's empathy has been proposed since the early days of Social Psychology, starting with the Stanford Prison Experiment. Using new tools, my research examines an age-old question: does social power have an analgesic effect on empathy for pain? Moreover, I explore the underlying mechanisms that reverse, attenuate and enhance the effect of social power on empathy for pain, including personality traits, demographic background, and neural components using a combination of surveys, behavioural tasks and electroencephalogram (EEG). I hypothesise that social power dampens empathy for pain, in that powerful individuals have reduced empathic responses when observing another in pain.

In chapter 2, I examined whether perceived social power is associated with empathy for pain. Across two studies, I elicited empathic responses in participants using a previously-validated series of photos depicting painful injections delivered to another's hand. I subsequently assessed individual differences in personal sense of power. Results from study 1 and 2 showed that high social power predicted low empathic responses.

In chapter 3, I experimentally manipulated social power using a role-play task, during which participants were randomly assigned to the role of a powerful manager or powerless subordinate prior to completing the same empathy task used in studies 1 and 2. Across behavioural studies 3 and 4, I report results on differences in empathic responses between powerful and powerless participants.

Building upon results from chapter 3, I explored the role of gender and age across studies 1 to 4 in chapter 4. I report results from a meta-analysis aggregating participants across studies and discuss the social demographic factors that drive a sense of social power. Building upon results from the meta-analysis, I examined the personality traits that moderate the effect of social power on empathy for pain in study 5 and report the results in the same chapter.

In chapter 5, I examined the effect of social power on the neural components of empathy for pain using the same empathy task and EEG. As EEG provides a high temporal resolution of neural activity on a scale of milliseconds, it provides valuable insights into the process and magnitude of empathic responses across powerful and powerless participants. To manipulate social power, participants wrote about a past memory in which they held power over someone else or vice versa, and subsequently completed the same empathy task used in studies reported in chapters 2 and 3. Moreover, participants completed a previously-used helping task that assessed their prosocial behaviour. I report results on the differences in event-related potentials (ERPs) and prosocial behaviour between powerful and powerless participants.

Finally, in chapter 6, I synthesise findings obtained across all behavioural studies and EEG experiment. I offer concluding thoughts and directions for future research, and outline potential limitations of studies conducted in this doctoral thesis.