Filling the void digitally: Compensatory expectations and social support in compulsive food ordering within super-app ecosystems

Compulsive food ordering is an emerging digital risk, but the cognitive mechanisms linking psychosocial deficits to behavior are not yet fully understood. This study integrates the I-PACE model with Compensatory Consumption Theory to develop and empirically validate a theory-driven predictive framework, assessed through PLS-SEM and out-of-sample cross-validation in a super-app environment. Findings show that “compensatory expectation” is the main driver, fully mediating the effects of loneliness and self-control on compulsive food ordering. This indicates that specific compensation-seeking beliefs, rather than simple cue-reactivity, drive compulsive ordering. Perceived social support serves as a key resource, reducing risks by alleviating loneliness and supporting self-regulation. Importance-performance map analysis suggests that cognitive restructuring should be prioritized over restriction, providing a data-driven approach for digital responsibility and platform well-being. • Loneliness is linked to compulsive food ordering on delivery apps • Compensatory expectations help explain compulsive food ordering • Perceived social support is linked to lower loneliness • Higher self-control is linked to weaker compensatory expectations • The model showed meaningful out-of-sample predictive performance