Multimodal UX evaluation: contrasting the SUS questionnaire with physiological activation inference

Data were collected in a single-session laboratory protocol in which each participant wore an EmotiBit wearable while completing a baseline rest period, three representative portal navigation tasks, and a post-session questionnaire. Recordings were exported from the EmotiBit firmware in standard CSV and JSON format immediately after each session. The cohort comprises 10 undergraduate university students: 5 women and 5 men, aged 19 to 25 years (self-reported). Participants were enrolled in technology- or statistics-related undergraduate programs and were familiar with institutional web services to varying degrees, as captured in the self-report file. Sessions were conducted between May and June 2026 under a protocol designed to evaluate whether global usability scores align with moment-to-moment physiological responses during concrete portal tasks. This deposit contains raw exported signals, protocol timing markers, session metadata, and self-report responses only. No derived features, model outputs, or analysis scripts are included.