<p dir="ltr">Instagram is among the most popular social media platforms used by young adults worldwide, yet the psychological mechanisms underlying sustained usage remain insufficiently understood. Traditional usability research in human–computer interaction has primarily focused on task-oriented qualities such as efficiency and ease of use, whereas contemporary user-experience (UX) research emphasises hedonic, emotional, and experiential dimensions of interaction. The User Engagement Scale–Short Form (UES-SF) integrates these perspectives by conceptualising engagement through four dimensions—perceived usability (PU), aesthetic appeal (AE), focused attention (FA), and reward (RW)—but relatively few studies have directly contrasted their predictive power in explaining actual social media behaviour. Using survey data from 445 university students in India (mean age ≈ 20 years), this study examined whether Instagram usage is driven more strongly by perceived usability or by emotional engagement.</p><p><br></p><p dir="ltr">Participants reported demographic characteristics, average daily Instagram usage, years of platform use, account privacy status, and responses to 12 UES-SF items. PU items were reverse-coded so that higher scores reflected better usability and lower frustration. The psychometric properties of the UES-SF were assessed using reliability analysis, composite reliability, and average variance extracted (AVE). Confirmatory factor analysis supported the four-factor structure and discriminant validity. Descriptive statistics and correlation analyses compared usability and emotional engagement profiles. Hierarchical regression analyses tested the incremental predictive power of PU versus AE, FA, and RW on standardized Instagram usage while controlling for age, gender, years on Instagram, and account privacy. In addition, competing usability-dominant, emotion-dominant, and integrated structural models were approximated using path analysis and compared using Akaike and Bayesian information criteria.</p><p><br></p><p dir="ltr">The UES-SF demonstrated acceptable reliability (Cronbach’s α = 0.81 for FA, 0.65 for PU, 0.88 for AE, and 0.77 for RW) and good convergent validity, with AVE exceeding 0.50 for three dimensions. Descriptively, reversed PU scores were highest, exceeding FA and RW. However, Instagram usage (approximately 2.9 hours per day) was only weakly correlated with PU and more strongly associated with RW. Regression analyses showed that PU did not significantly improve the prediction of usage beyond demographic controls, whereas AE, FA, and RW jointly explained additional variance. When entered simultaneously, only FA and RW remained significant predictors. Path analyses indicated that an emotion-dominant model provided the best fit. These findings suggest that while usability constitutes a necessary baseline for positive interaction, emotional engagement—particularly focused attention and perceived reward—plays a more decisive role in sustaining Instagram use among young adults.</p>
