A body image disturbance is considered to be a key characteristic of patients suffering from an eating disorder (ED) such as anorexia nervosa (AN) and bulimia nervosa (BN). It is, however, a complex phenomenon whose inherent multidimensionality (neurophysiological, cognitive and emotional aspects) is a serious challenge for researchers. Two independent components of body image disturbance have been distinguished: a distorted body perception (‘body distortion’) and a negative cognitive evaluation (‘body dissatisfaction’) (Gardner and Garfinkel, 1981 and Gardner, 1996; Cash and Deagle, 1997). With respect to neuropsychology, several studies have tried to address questions as to whether there is a localized body image neural module, whether it corresponds to the somatosensory areas of the brain, and whether it is separate from the cerebral locations of our assembled knowledge and beliefs about our bodies (Pruzinsky and Cash, 2002). Body awareness is a complex and dynamic process including sensory modalities, affective experiences and cognitive interpretations. Still, it is not clear how the brain integrates the relevant information to compute something we call ‘body image’. Despite several limitations due to methodological difficulties, neuroimaging studies – mainly functional techniques – are improving the understanding of ED, particularly with respect to the brain areas that might be implicated (Chowdhury and Lask, 2001 and Frank et al., 2004). Besides the hypoperfusion (Kuruoglu et al., 1998, Naruo et al., 2000, Takano et al., 2001, Râstam et al., 2001 and Chowdhury et al., 2003) and hypometabolism found in AN and BN in resting conditions (Delvenne et al., 1999), task activation experiments have shown areas of brain activation that might be specific to the pathophysiology of ED. Food imagination, visual presentations and eating high caloric food provoke anxiety and increase activation in areas related to emotional restorage such as the amygdale (Nozoe et al., 1995, Naruo et al., 2001, Gordon et al., 2001 and Santel et al., 2006), and in areas of executive function, decision-making, error monitoring and reward expectancy, as the anterior cingulated cortex and medial prefrontal (Uher et al., 2004). More research is needed, however, to clarify the nature of those changes. At present, very few studies have investigated the rCBF changes in the recovery process of AN (Kojima et al., 2005 and Matsumoto et al., 2006). Furthermore, controversial findings regarding body image dysfunction have been reported. To be explicit, the temporomesial area has been associated with the perception of unpleasant words concerning body image (Shirao et al., 2003). In addition, neuroimaging studies using digital body images have shown right amygdala activation (fear network) in AN patients, comparable to the reactions in patients with an anxiety disorder who are recalling aversive memories (Seeger et al., 2002). Nonetheless, the same group reported a larger series of patients (Wagner et al., 2003) and had somewhat different findings: compared with healthy controls, AN patients experienced a hyperresponsiveness in the inferior parietal lobe, suggestive of a disturbance in the visuospatial processing of the own body shape. Finally, brain responses to line drawings of underweight, normal and overweight female bodies have been assessed with functional magnetic resonance (Uher et al., 2005). ED patients rated the body shapes as more aversive than healthy women, and this correlated positively with activity in the right medial apical prefrontal cortex.
A number of controversial aspects regarding the nature and impact of body image disturbances may possibly be elucidated in the future, for instance, by means of activation techniques assessing the interaction of cognitive and emotional processing in ED (Chowdhury and Lask, 2001). At the moment, we do not know of any research assessing at the same time the relationships of both the ‘perceptual distortion’ and the ‘cognitive evaluation’ components of body image disturbance with brain activity, controlling for the influence of anxiety and depressive symptoms. Considering that BN and AN patients have shown differences regarding body image disturbance, distortion as well as dissatisfaction (Cash and Deagle, 1997), it would be interesting to test whether those differences may be replicated and show a correlation with brain activity. We have designed the present study to answer the following question: Do AN patients differ from BN patients and control participants with regard to the patterns of change in regional cerebral blood flow (rCBF) from baseline to their own body image exposure and from baseline to a neutral stimulus exposure? Beyond the existing imaging literature, this study offers the comparison of the mentioned three groups. The main hypothesis was that patients with AN and BN, compared with healthy people, might reveal different regional paterns of brain activation within the above-described experiments, probably related to the somatosensory and emotion processing systems, respectively. This investigation is part of a longitudinal study also aimed at evaluating therapeutic interventions.
