Smart Textile Interfaces for Immersive Museum Experiences using AR/AI

Thesis Abstract

The proliferation of smart textiles and augmented/immersive technologies has opened new avenues for reimagining museum pedagogy, where garment-based interfaces can blend tactile, visual, and contextual data to augment visitor interaction with cultural artifacts. This study addresses the gap in empirical evidence on how integrated AR/AI-enabled textile interfaces influence user engagement, learning outcomes, and accessibility in museum contexts. Specifically, it investigates how smart fabric sleeves embedded with conductive threads, micro-LEDs, and haptic actuators, when synchronized with a mobile AR application and an AI-driven content mediator, can (i) enhance interpretive depth of exhibits, (ii) support inclusivity for diverse visitors, and (iii) provide measurable data for curatorial decision-making. The aim is to develop a scalable prototype and evaluate its effects on embodied learning and visitor satisfaction in a real-world gallery setting. The objectives are (1) to design and implement a wearable textile interface integrated with AR visualization and AI-based content adaptation; (2) to assess the impact of the interface on perceived engagement, recall, and experiential learning using a mixed-methods approach; (3) to examine inclusivity outcomes across age, prior expertise, and accessibility needs; (4) to analyze interaction patterns, session flow, and gesture-based controls using log-data and video analytics; and (5) to develop a design framework and recommendations for future curatorial practice. The study adopts a pragmatic research design combining experimental and case-study elements to balance control and ecological validity. The population comprises museum visitors at three gallery spaces, with a target sample of 180 participants across two exhibit clusters. A stratified random sample will select 90 participants for the experimental condition and 90 for the control condition, with an embedded qualitative subsample of 24 participants for in-depth interviews and 12 curatorial staff for focus groups. Instruments include a wearable textile prototype evaluation checklist, AR content interaction logs, pre/post knowledge assessments, the User Engagement Inventory, and semi-structured interview guides. Validity and reliability are ensured through pilot testing (n=20), triangulation of quantitative and qualitative data, and inter-rater reliability coding (Cohen’s kappa > 0.80) for observational analyses. Data analysis employs a mixed-methods sequence. Quantitative analysis will use repeated-measures ANOVA to compare learning gains and engagement scores between experimental and control groups, supplemented by multiple regression to identify predictors such as prior museum experience and accessibility needs. Descriptive statistics will profile interaction patterns from AR/AI logs, while time-series analyses will examine engagement trajectories across exhibit segments. Qualitative data will undergo thematic analysis guided by Braun and Clarke’s framework, with coding verified by a second coder to reach thematic saturation. A conceptual model integrating constructivist learning theory with situated cognition and embodied interaction will frame interpretation, with the Technology Acceptance Model adapted to wearable AR contexts to interpret user receptivity and perceived usefulness. Expected findings suggest that smart textile interfaces with synchronized AR overlays and AI-driven content personalization yield statistically significant improvements in knowledge retention, engagement, and positive affect, relative to conventional exhibits. It is anticipated that haptic cues and adaptive narratives will reduce cognitive load for novice museum-goers while offering richer interpretive pathways for expert visitors. The study also expects differential effects across demographic groups, highlighting the importance of accessibility-aware design and adjustable interaction modalities. The contribution to knowledge includes (i) an empirically validated, scalable blueprint for wearable-AR museum experiences, (ii) a design framework linking textile engineering, AR visualization, and AI-driven content mediation, and (iii) practical guidelines for curators on data-informed exhibit configuration and visitor segmentation. Conclusions emphasize that integrated smart textiles can meaningfully enhance embodied learning in museum settings when complemented by carefully calibrated AR content and inclusive interaction design. Recommendations address iterative prototyping cycles, long-term deployment strategies, ethical considerations for data privacy, and standards for interoperability across devices and exhibits to support broader adoption in cultural heritage institutions.