# Pilot study: PORTION-O-MAT—a mixed reality solution for investigating perceptual and behavioural abnormalities during food portioning in adolescents with anorexia nervosa

## TL;DR

This pilot study explores how mixed reality can help understand how adolescents with anorexia nervosa perceive and choose food portions, finding they select smaller portions than healthy individuals.

## Contribution

The study introduces a novel mixed-reality method to investigate portion size estimation in anorexia nervosa patients with ecological validity.

## Key findings

- Anorexia nervosa patients selected significantly smaller portion sizes than healthy adults, especially for high-calorie foods.
- No significant differences were found in decision-making time or uncertainty indicators between the groups.
- The mixed-reality approach effectively simulated meal selection and showed altered food perception in anorexia patients.

## Abstract

Anorexia nervosa (AN) is a severe eating disorder characterized by perceptual distortions and restrictive eating behaviours. This pilot study examines portion size estimation in adolescent AN patients using a mixed-reality (MR) approach. The objective is to evaluate the potential of this method for the assessment and treatment of AN, with a particular focus on its ecological validity and its applicability for investigating portion size estimation.

A total of 30 female participants were recruited: 15 adolescent AN patients and 15 healthy adults as pretest. Participants engaged in a simulated meal assembly task within an MR environment, adjusting portion sizes of virtual food components to match a “typical” meal size (100%). Decision-making patterns and self-reported eating disorder symptoms were recorded. Statistical analyses included descriptive statistics, group comparisons and correlation analysis to examine associations between clinical variables and portion sizes, decision-making time and other decision parameters.

AN patients consistently selected significantly smaller portion sizes than healthy adults, particularly for high-calorie foods. No significant differences were observed in decision-making time or uncertainty indicators.

The findings support the hypothesis that AN patients exhibit altered food perception in the sense that they tend to overestimate the size of visually presented food portions. The MR approach proved effective in simulating meal selection, Future studies should include larger and more diverse samples and incorporate real food intake to further validate these results.

The online version contains supplementary material available at 10.1007/s40519-025-01797-2.