Final Report Summary - SIZE MATTERS (Size Matters: investigating the link between affective and perceptual body representations using multisensory illusions and brain imaging)
Negative feelings towards our bodies are widespread in our modern society and can have harmful consequences for our mental and physical health being associated with eating disorders such as anorexia nervosa. Although anyone can feel dissatisfied with their body, females tend to feel more dissatisfied and represent a larger number of eating disorder patients in the clinic compared to males. One of the main causes of body dissatisfaction is thought to be comparisons between how we perceive our own body to look and unattainable social ideals, which are often more pronounced for females. However, although links between how we perceive and how we feel about our body seem intuitive, until now evidence for a direct relationship between the two has been absent, much less how such a relationship may link to disordered eating and societal pressures. Therefore, in this project we aimed to identify behavioural and neural links between body perception and body affection and to determine how such a relationship may contribute to eating disorder vulnerability in today’s modern world.
The main paradigm used within the project was the full-body illusion. Touching a mannequin body viewed through a virtual reality head-set from a natural first-person perspective in synchrony with touches delivered to the actual body elicits feelings of ownership over the mannequin. Changing the appearance of the mannequin and can in turn modify the perceived size and shape of the actual body. Asynchronous touch abolishes the illusion and provides a control condition for the experiments. During the first year of the project we completed two behavioural experiments and a neuroimaging experiment to identify behavioural and neural links between body perception and body satisfaction. In the first study we demonstrated that illusory modulation of body size has a direct effect on reported body satisfaction and were the first to use the full-body illusion to examine eating disorder psychopathology. (Preston and Ehrsson, 2014, PLoS One). We modified the appearance of the mannequin by digitally stretching and contracting the image. Body satisfaction and perceived body size were measured before and after the illusion. It was found that a narrower (slimmer) mannequin decreased perceived body width and increased body satisfaction. Widening/ stretching the image, however, had no overall effect. Instead, changes in body satisfaction owning a wider body were related to non-clinical levels of eating disorder psychopathology, with higher psychopathology corresponding to greater increases in body satisfaction. Reasons for this seemingly counterintuitive finding were thought to relate to the experimental stimuli in that a widened mannequin does not create a true socially undesirable (fat) body shape. Therefore, for subsequent experiments I developed ecologically valid stimuli by creating pre-recorded videos of real obese and slim male and female models to use instead of images of mannequins. For the videos the timing of the touch was controlled by audio cues only heard by the experimenter. For synchronous trials the experimenter was cued to touch the participant at exactly the same time as the model was touched in the video. For asynchronous trials, these audio cues were delayed.
In a second behavioural study (Preston and Ehrsson, in preparation) we found that illusory obesity induced by a full-body illusion over an obese body decreased body satisfaction. Evidence for this came from explicit reports of body satisfaction using a questionnaire and, in a separate experiment (conducted in year two of the project), implicit measures of body satisfaction using an implicit association task. Implicit association tasks measure unconscious associations between different categories. In order to infer body satisfaction, associations between self/other and attractive/unattractive categories were measured. Higher body satisfaction equates to stronger associations between self and attractive, and other and unattractive. Weaker associations between these categories (and correspondingly higher associations between self and unattractive, and other and attractive) indicate lower body satisfaction. Both explicit and implicit measures demonstrated that females (compared to males) and individuals with higher eating disorder psychopathology have stronger emotional responses to illusory changes in body size.
Our first fMRI experiment examined brain activity as a result of illusory changes in body size by inducing feelings of ownership over slim and obese bodies (Preston and Ehrsson, in preparation). Findings from this experiment implicated the anterior cingulate cortex and the anterior insular cortex, regions of the brain previously associated with anorexia nervosa, with negative feelings towards the body and eating disorder psychopathology in healthy participants. The same experiment also uncovered evidence for a direct neural link between these body satisfaction brain regions and the intraparietal sulcus - an area already known to be involved in body perception - thus suggesting that information about changes in body size from the parietal cortex reaches the anterior insula and cingulate cortices and there modifies feelings towards the body. We also found reduced cingulate activity in females and those with higher non-clinical levels eating disorder psychopathology, which may represent mechanisms underlying eating disorder vulnerability.
During the second year of the project we examined how body perception is influenced by social context. A second brain imaging study sought to modulate body satisfaction and the associated brain regions in females using images of socially ideal female bodies. These images were taken from current fashion magazines and were presented to female participants interspersed periods of a full-body illusion over a healthy weight female body. Control images were used in separate blocks consisting of magazine images of attractive home interiors. Analysis of the brain imaging data is still underway, but behavioural data from the scanning sessions confirms a negative effect of the fashion magazine images on body satisfaction. There is also evidence of a positive effect of ownership over the healthy weight body depending on level of eating disorder psychopathology, but only during trials in which participants were also exposed to the ideal body images. This suggests an interaction between the social context and body perception, which may be contribute to the development of disordered eating. Such findings can help us understand why some people develop eating disorders and others don’t within a similar social environment.
Finally, we investigated how visual perspective of the body can modulate central body representations. We were able to successfully demonstrate that the full-body illusion could be elicited over a mannequin body viewed in a mirror, both in terms of subjective reports of ownership as well as objective physiological responses when threatening the mannequin (Preston et al., under review, Scientific Reports). This finding validates these methods for further experiments modulating body size from different visual perspectives and also add to our knowledge in the current debate on visual perspective, mirror reflections, and self/body perception. However, adapting this paradigm for neuroimaging experiments and translation to ecologically valid stimuli of real obese and slim models proved challenging. Therefore, this aspect was not fully explored within the duration of the project. However, piloting conducted will inform future studies designed to tackle this question.
Collectively the results of the current project extensively add to our knowledge of the fundamental principles of human central body representations, as well as how they might contribute to disordered eating behaviours. Therefore, such findings can have significant impact facilitating further academic research and also future development of prevention and treatment strategies for eating disorders.
The main paradigm used within the project was the full-body illusion. Touching a mannequin body viewed through a virtual reality head-set from a natural first-person perspective in synchrony with touches delivered to the actual body elicits feelings of ownership over the mannequin. Changing the appearance of the mannequin and can in turn modify the perceived size and shape of the actual body. Asynchronous touch abolishes the illusion and provides a control condition for the experiments. During the first year of the project we completed two behavioural experiments and a neuroimaging experiment to identify behavioural and neural links between body perception and body satisfaction. In the first study we demonstrated that illusory modulation of body size has a direct effect on reported body satisfaction and were the first to use the full-body illusion to examine eating disorder psychopathology. (Preston and Ehrsson, 2014, PLoS One). We modified the appearance of the mannequin by digitally stretching and contracting the image. Body satisfaction and perceived body size were measured before and after the illusion. It was found that a narrower (slimmer) mannequin decreased perceived body width and increased body satisfaction. Widening/ stretching the image, however, had no overall effect. Instead, changes in body satisfaction owning a wider body were related to non-clinical levels of eating disorder psychopathology, with higher psychopathology corresponding to greater increases in body satisfaction. Reasons for this seemingly counterintuitive finding were thought to relate to the experimental stimuli in that a widened mannequin does not create a true socially undesirable (fat) body shape. Therefore, for subsequent experiments I developed ecologically valid stimuli by creating pre-recorded videos of real obese and slim male and female models to use instead of images of mannequins. For the videos the timing of the touch was controlled by audio cues only heard by the experimenter. For synchronous trials the experimenter was cued to touch the participant at exactly the same time as the model was touched in the video. For asynchronous trials, these audio cues were delayed.
In a second behavioural study (Preston and Ehrsson, in preparation) we found that illusory obesity induced by a full-body illusion over an obese body decreased body satisfaction. Evidence for this came from explicit reports of body satisfaction using a questionnaire and, in a separate experiment (conducted in year two of the project), implicit measures of body satisfaction using an implicit association task. Implicit association tasks measure unconscious associations between different categories. In order to infer body satisfaction, associations between self/other and attractive/unattractive categories were measured. Higher body satisfaction equates to stronger associations between self and attractive, and other and unattractive. Weaker associations between these categories (and correspondingly higher associations between self and unattractive, and other and attractive) indicate lower body satisfaction. Both explicit and implicit measures demonstrated that females (compared to males) and individuals with higher eating disorder psychopathology have stronger emotional responses to illusory changes in body size.
Our first fMRI experiment examined brain activity as a result of illusory changes in body size by inducing feelings of ownership over slim and obese bodies (Preston and Ehrsson, in preparation). Findings from this experiment implicated the anterior cingulate cortex and the anterior insular cortex, regions of the brain previously associated with anorexia nervosa, with negative feelings towards the body and eating disorder psychopathology in healthy participants. The same experiment also uncovered evidence for a direct neural link between these body satisfaction brain regions and the intraparietal sulcus - an area already known to be involved in body perception - thus suggesting that information about changes in body size from the parietal cortex reaches the anterior insula and cingulate cortices and there modifies feelings towards the body. We also found reduced cingulate activity in females and those with higher non-clinical levels eating disorder psychopathology, which may represent mechanisms underlying eating disorder vulnerability.
During the second year of the project we examined how body perception is influenced by social context. A second brain imaging study sought to modulate body satisfaction and the associated brain regions in females using images of socially ideal female bodies. These images were taken from current fashion magazines and were presented to female participants interspersed periods of a full-body illusion over a healthy weight female body. Control images were used in separate blocks consisting of magazine images of attractive home interiors. Analysis of the brain imaging data is still underway, but behavioural data from the scanning sessions confirms a negative effect of the fashion magazine images on body satisfaction. There is also evidence of a positive effect of ownership over the healthy weight body depending on level of eating disorder psychopathology, but only during trials in which participants were also exposed to the ideal body images. This suggests an interaction between the social context and body perception, which may be contribute to the development of disordered eating. Such findings can help us understand why some people develop eating disorders and others don’t within a similar social environment.
Finally, we investigated how visual perspective of the body can modulate central body representations. We were able to successfully demonstrate that the full-body illusion could be elicited over a mannequin body viewed in a mirror, both in terms of subjective reports of ownership as well as objective physiological responses when threatening the mannequin (Preston et al., under review, Scientific Reports). This finding validates these methods for further experiments modulating body size from different visual perspectives and also add to our knowledge in the current debate on visual perspective, mirror reflections, and self/body perception. However, adapting this paradigm for neuroimaging experiments and translation to ecologically valid stimuli of real obese and slim models proved challenging. Therefore, this aspect was not fully explored within the duration of the project. However, piloting conducted will inform future studies designed to tackle this question.
Collectively the results of the current project extensively add to our knowledge of the fundamental principles of human central body representations, as well as how they might contribute to disordered eating behaviours. Therefore, such findings can have significant impact facilitating further academic research and also future development of prevention and treatment strategies for eating disorders.