The sight of one's own body: Could qEEG help predict the treatment response in anorexia nervosa?

Abstract

Aims of the study: The study aims to identify the differences in brain activity between participants with anorexia nervosa and healthy control using visual stimulus conditions combined with the quantitative dense-array EEG recording analysis method called Brain Activation Sequences (BAS).

Materials and methods: 23 participants with anorexia nervosa and 21 healthy controls were presented with visual stimuli, including the subject's facial expressions and body images. The 128-channel EEG data were processed using BAS and displayed as activity in up to 66 brain regions. Subsequent cluster analysis was used to identify groups of participants exhibiting area-specific activation patterns.

Results: Cluster analysis identified three distinct groups: one including all healthy controls (HC) and two consisting of all participants with anorexia (AN-I with 19 participants and AN-II with four participants). The AN-I and AN-II groups differed in their response to treatment. Comparisons of HC vs. AN confirmed the dominance of the right cerebral hemisphere in participants with anorexia nervosa in two of the three reported conditions. The facial expressions condition, specifically the facial reaction expressing disgust, indicates the existence of a social attentional bias toward faces, whereas emotions remained undetected in participants. High limbic activity, medial frontal gyrus involvement, low fusiform cortex activity, and milder visual cortex activity in healthy controls compared to participants indicate that the facial expression stimulus is perceived by healthy subjects primarily as an emotion, not as the face itself. In the body image condition, participants showed higher activity in the fusiform gyrus and right insula, indicating activation of the brain's "fear network."

Conclusion: The study describes a specific pattern of brain activation in response to facial expression of disgust and body images that likely contributes to social-cognitive and behavioral impairments in anorexia. In addition, the substantial difference in the pattern of brain activation within the participants with AN and its association with treatment resistance deserves special attention because of its potential to develop a clinically useful prediction tool and identify potential targets for, for example, neuromodulatory treatments and/or individualized psychotherapy.

Keywords: anorexia nervosa; disgust; facial expressions; perception; qEEG; treatment response.

Figure 1

Examples of presented stimuli. (A) Facial expressions; (B) Body images. The red rectangle highlights the stimuli whose results are described in this paper.

Figure 2

Electrical sources at the selected time as a check of the consistency of the measured data. Top-down Anger-Fear-Disgust-Smile, during maximal activation using the LAURA method. 2,400 voxels representing gray matter in the Montreal Neurological Institute average MRI (2,394 plus six thalamic sources) are registered in Tailarach space with a typical MRI image of the subject. Three orthogonal dipole moments are calculated for each voxel with linear inverse estimation. The color palette, which is kept constant for this series of images, reflects the effective value of the three dipole moments of each voxel.

Figure 3

The first six sites that were most active in the Disgust condition were sequenced. The purple color represents the lateral sites, and the white color refers to the medial areas. Connecting lines without arrows show the same level of activity. HC Disgust refers to the HC group, AN-I Disgust refers to the AN-I group, and the smaller AN-II Disgust image refers to the AN-II group in results (Table 2).

Figure 4

Sequenced the first six most active sites in the Anamorphic condition. Purple numbers represent lateral sites; white refers to medial sites. Connecting lines without arrows show the same level of activity at a given time. The HC corresponds to group HC, AN-I refers to group AN-I, and the smaller figure AN-II refers to group AN-II in Table 2. The upper group of figures describes the activity elicited by the Real body image, and the lower group the response to the distorted, Desired body shape.