Abstract / Summary
Eating behaviors are traditionally conceptualized as individual psychological phenomena, yet their structural organization remains largely unmapped in non-Western contexts. Adopting a dimensional perspective, this study used network analysis to characterize the architecture of non-clinical eating behaviors in India (N = 1,508; 48.9% male, 50.7% female; age range = 18-65 years) based on 35 well-validated biopsychosocial measures. Mixed Graphical Models revealed a highly optimized, small-world system defined by a dual-layered hierarchy. Cultural and living conditions (religion, home type) served as local behavioral anchors (highest expected influence), whereas socioeconomic factors (employment, education) and self-esteem functioned as structural bridges (highest betweenness centrality) integrating the biopsychosocial system. Notably, shape and weight concerns were highly predictable but operated as local cluster nodes rather than global integrators-directly challenging Western body-image-centric models. Limitations include cross-sectional design, precluding causal inference, and the need for ongoing validation of Western-derived measures in Indian contexts. These findings suggest that in a developing economy, broader psychosocial structural factors rather than individual factors may contribute to the maintenance of eating disorder symptoms.