The filtered self: How digital beauty standards shape social comparison and identity curation in the age of social media

Digital aesthetic sensitivity — individuals’ orientation toward visual perfection and curated self-presentation on social media — has emerged as a correlate of online impression management. However, the psychological mechanisms through which aesthetic sensitivity is associated with self-presentation behavior remain underexplored. Objective: This study proposes and tests a serial mediation model in which digital aesthetic sensitivity (DA) is indirectly associated with social media self-presentation (SMSP) through two sequential mediators: social media comparison (SMC) and false self-presentation (FSP). Data were collected from 359 active social media users via an online snowball-sampled survey. Instruments included the Digital Aesthetic Scale (α = .898), the Social Media Comparison Scale (α = .856), and the Social Media Self-Presentation Scale. Hierarchical regression analyses and structural equation modeling (SEM) with maximum likelihood estimation were employed. Digital aesthetics was significantly associated with social media comparison (β = .589, p < .001), which in turn was associated with false self-presentation (β = .370, p < .001). False self-presentation showed the strongest association with self-presentation (β = .449, p < .001). The direct effect of digital aesthetics on self-presentation became non-significant when both mediators were included, indicating that the total DA→SMSP effect is fully mediated; digital aesthetics nonetheless retained a significant direct path to false self-presentation, indicating partial mediation of that specific segment. The model explained 52.7% of variance (CFI = .958; RMSEA = .052). Women, younger adults, and students reported significantly higher aesthetic sensitivity and false self-presentation. The total association between digital aesthetics and self-presentation was carried by a sequential psychological pathway involving cognitive comparison and identity distortion, with no significant residual direct effect on self-presentation itself. Findings have potential implications for platform design and digital literacy education; given the non-clinical convenience sample, any implications for clinical screening should be regarded as exploratory and contingent on future validation in clinical populations. Limitations include cross-sectional design, convenience sampling, reliance on self-report measures, and predominant use of a single platform (Instagram, 79.9%).