# From Self-Esteem to Symptoms: A Potential Role for Difficulties Accessing Internal States and Body-Checking Behavior in Disordered Eating Patterns

## TL;DR

This study explores how low self-esteem might lead to disordered eating patterns through difficulties in understanding internal states and body-checking behaviors.

## Contribution

The study introduces a serial mediation model linking self-esteem, DAIS, body-checking, and disordered eating patterns.

## Key findings

- Lower self-esteem was associated with higher disordered eating patterns.
- Both DAIS and body-checking partially mediated the relationship between self-esteem and disordered eating.
- A serial mediation model showed a significant pathway from self-esteem to DAIS to body-checking to disordered eating.

## Abstract

Drawing on the Seeking Proxies for Internal States (SPIS) model and the concept of Difficulties in Accessing Internal States (DAIS), the present study examined the statistical associations among self-esteem, DAIS, body-checking, and disordered eating patterns (DEP). Within the SPIS framework, self-esteem is conceptualized as an evaluative internal state that may be appraised through externally observable proxies, such as body appearance. Cross-sectional data were collected from 200 adults recruited through Prolific Academic. Hayes’ PROCESS macro was used to test simple and serial mediation models examining whether DAIS and body-checking statistically account for associations between state self-esteem and DEP. Lower self-esteem was associated with higher DEP. Both DAIS and body-checking statistically accounted for portions of this association in simple mediation models. In a serial mediation model, the fully sequential pathway (self-esteem → DAIS → body-checking → DEP) remained statistically significant after accounting for shared variance among mediators. Given the cross-sectional design, these findings cannot establish temporal or causal relationships, but the observed pattern of associations is compatible with the proposed conceptual process. Additionally, our findings are based on a nonclinical sample and reflect variability in subclinical eating pathology. Implications for extending the SPIS framework to dimensional eating-related phenomena are discussed.