Choices under manipulation: Cognitive processes and behavioral biases of older adults under perceived exposure to dark patterns in online shopping

With the convergence of population aging and digitalization, more older adults are engaged in online shopping. However, the prevalence of dark patterns on digital platforms is associated with unintended consumption behaviors. Guided by the Extended Stimulus-Organism-Response (E-S-O-R) model, this study developed an integrative framework incorporating three mediators-cognitive overload, decision fatigue, and trust bias-and examined the moderating role of digital literacy. A cross-sectional self-report survey of 408 older adults aged 60 and above with exposure to online shopping environments was conducted, and structural equation modeling (SEM) was applied for path and mediation analyses; dark patterns were operationalized as perceived exposure to selected transaction-related manipulative interface tactics. The results show that perceived exposure to dark patterns was significantly and positively associated with cognitive overload, decision fatigue, and trust bias, all of which positively predicted unintended consumption behavior, with decision fatigue exerting the strongest effect. Moreover, digital literacy negatively moderated the path from trust bias to unintended consumption, reducing the impact of biased trust. These findings highlight the mechanisms through which dark patterns impair older adults' decision-making and suggest practical implications for platform design optimization, targeted consumer protection policies, and tiered digital literacy education to mitigate irrational consumption risks in digital contexts.