Women are up to twice as likely to suffer from post-traumatic stress disorder (PTSD) than men. Failure to discriminate between cues predicting threat and safety is associated with PTSD, yet sex differences in fear discrimination remain poorly understood. Here, we examined sex differences in auditory fear discrimination in rats using a combination of behavioural, computational and electrophysiological methods. In the initial behavioural study, males and naturally cycling females underwent 1-3 days of discrimination training, consisting of pairings of one tone (CS+) with shock and presentations of another tone (CS-) alone. After one day of training, females, but not males, discriminated between the CS+ and CS-. With 2-3 days of training, however, males discriminated and females generalised between the CS+ and CS-. Further testing also revealed that males successfully encode the CS- as a safety signal, whereas females do not. Using reduced computational models, we investigated how both ‘discrimination’ and ‘generalisation’ phenotypes can be generated in silico. We achieved this through a simulation of neural activity produced via ‘fear’ and ‘safety’ neural sub-populations of the basolateral amygdala (BLA) in response to CS+ and CS- cues. By using a model representation of extended fear discrimination training and retrieval, we found that generalisation between the CS+ and CS- could be produced from reduced inhibition, or increased excitation, of fear neurons. Due to their involvement in regulating learned fear, we additionally aimed to investigate the roles of the prelimbic (PL) and infralimbic (IL) cortices of themedial prefrontal cortex in fear discrimination. By concurrently recording activity from the PL, IL and BLA in awake behaving animals during retrieval of the CS+ and CS- after extended discrimination training, we examined the individual contributions and functional interactions of these regions during this learning paradigm. We found that, in males, the PL showed an increase in power at both theta (4-12 Hz) and gamma (30-120 Hz) frequencies during presentations of the CS- compared to the CS+, whereas this increase was largely absent in females. Taken together, these results indicate that, while females show fear discrimination with limited training, they generalise with extended training. We hypothesised that this generalisation in females is likely due to impaired safety learning, which may result, in part, from sex differences in the neural circuitry underlying fear discrimination.

The amount of pain we feel is not always directly related to the amount of damage our body suffers. In fact, research over the last decade has shown that the experience of pain is strongly linked with how we feel about our body, including its shape and how much we like it. Chronic pain is associated with a distorted mental representation of the body, which can have a significant impact on everyday live. Evidence suggests that multisensory illusions can modulate pain and can lead to changes in body perception. However, the factors that may contribute to previously observed analgesic effects remain unclear. This thesis aimed to systematically asses the effects of these illusions on pain and body representation in both healthy individuals and individuals with chronic pain. First, this thesis aimed to investigate the effects of multisensory illusions on body representation, sense of body ownership, and pain in healthy individuals. Chapter 2 found that multisensory illusions can alter perceived body shape, body satisfaction, without losing ownership. Chapter 3 examined whether distorting the size and shape of the virtual hand could modulate pain in healthy participants. No evidence was found for this, which could indicate that pain is not necessarily affected by virtual body ownership over a distorted hand. Having established that multisensory illusions can alter body perception and the affective experience of the body, the second aim of this thesis was to investigate the effect of multisensory illusions on body representation, body ownership, and pain in individuals with hand osteoarthritis (HOA). Chapter 4 investigated whether people with HOA performed differently on a hand laterality motor task to investigate whether people with HOA present with a disruption of the working body schema. Though no evidence was found for a general impairment on this task, the findings suggested that performance on the task was mediated by the presence of pain. Chapter 5 investigated effects of multisensory illusions on pain and pain sensitivity in individuals with HOA. The results demonstrated that viewing the body could modulate pain; and affected body representation, body ownership, and agency. Chapter 6 examined the effects of multisensory illusions on subjective and objective aspects of body perception in people with HOA of the hands, compared to healthy controls. Though no evidence was found for an analgesic effect of the illusion, results showed that participants with HOA have a disturbed experience of the size of their hand compared to healthy controls. Furthermore, the results suggest that individuals with HOA may have an abnormally high body dissatisfaction that cannot readily be altered by multisensory illusions. This thesis found mixed support for the analgesic effects of multisensory illusions on pain in HOA and concludes that the specific context in which pain occurs is important. It also highlights the many perceptual and cognitive factors that may contribute to the modulation of pain. The findings imply that future work should focus on interventions that are more portable and accessible for home use, and focus on developing research around the effects of repeated and prolonged exposure to multisensory illusions on pain, body image disturbances and body dissatisfaction.