The work shows how generative AI can mediate engagement with poetic heritage in culturally grounded emotional-support interactions and suggests that culturally grounded content and structured guidance should anchor system design, while multimodal presentation may strengthen resonance and engagement.
Abstract
Digital systems increasingly mediate emotional support, yet their interactions often remain culturally generic. Accordingly, we examine how a poetic tradition can be operationalized as a culturally grounded interactive medium and how generative AI can support such engagement. The resulting interaction design framework translates staged literature-based support and tradition-specific poetic aesthetics into guidance for digital system design. Poemithy instantiates the framework as a multimodal, LLM-enabled multi-agent system for guided reflection through classical Chinese poetry. A controlled between-subjects study with 50 participants compared text-only and multimodal versions. Both conditions showed medium-to-large within-session improvements in affect, anxiety, and emotion regulation, while between-condition tests detected no differences in these changes. Among secondary post-session user-experience measures, the clearest observed differences favored multimodality in perceived attunement, perceived task success, and engagement; usability and hedonic quality were descriptively higher, while workload did not differ detectably. Post-only cultural ratings were descriptively favorable in both conditions for cultural identification, poetry-engagement and dissemination intentions, and perceived cultural enrichment. Together, the findings suggest that culturally grounded content and structured guidance should anchor system design, while multimodal presentation may strengthen resonance and engagement. More broadly, the work shows how generative AI can mediate engagement with poetic heritage in culturally grounded emotional-support interactions.
Live music is not only an artistic expression but also a social encounter in which performers and listeners negotiate meaning. While popular concerts often embrace this exchange through collective gestures and technological mediation, classical recitals usually preserve a strictly passive audience role. This work investigates how digital interaction can bridge this gap. We designed FeelIt, a mobile and web-based application that transforms spectators’ emotional responses into visual projections, thereby integrating the public into the concert atmosphere without altering the music itself. The system was evaluated during an international piano seminar, where engagement was assessed through direct observation, the User Engagement Scale (UES), and qualitative accounts from both listeners and performers. Audience members highlighted the novelty, aesthetic appeal, and rewarding nature of the tool, though some reported difficulties related to interface design and reaction visibility. Musicians, in turn, emphasized that the visual feedback generated a stronger sense of dialogue with the public and intensified their emotional connection during performance. Beyond reporting empirical evidence, this work contributes by (i) introducing and validating a non-intrusive mechanism for audience participation in classical concerts, (ii) offering insights into how interactive technologies can enrich experiences without compromising artistic integrity, and (iii) providing practical design guidelines for integrating spectators’ voices into traditionally passive cultural events.
Genildo Gomes, Fernando Nogueira, João Gustavo Kienen et al.· Journal of the Brazilian Com...· 0 citations
This study addresses the need to understand human engagement with empathic Generative Artificial Intelligence (GenAI) digital companions (DC), as these systems increasingly fulfil social and emotional roles in daily life. Motivated by the rapid adoption of DC technologies and their potential to reshape human–AI relationships, the study applies Assemblage Theory to explore DC within broader socio‐material systems. Through a mixed‐methods design, 40 participants interacted with a Carl Sagan‐inspired avatar while biometric and qualitative data were collected to map engagement dynamics. Results show that greater user autonomy and participatory linguistic styles significantly enhance engagement, whereas authoritative language reduces it. Findings reveal how context, agency, and language shape bonds between humans and AI, emphasizing the importance of designing DCs that empower users and create authentic connections. This research advances theoretical and practical understanding of devices that promise to transform future social interactions.
Aitor Ezker‐Galech, Raquel Chocarro, Mónica Cortiñas et al.· Psychology & Marketing· 0 citations
Guided meditation is predominantly delivered through audio-only media or in-person sessions, yet social robots may offer a richer form of interaction through physical embodiment. To investigate this, we implemented a loving-kindness meditation on the Furhat robot and compared user responses to the robot-guided meditation to an equivalent audio-only condition. Social presence was measured on a quantitative survey, alongside qualitative post-experiment interviews. Results showed significantly higher social presence scores in the embodied condition, with participants reporting higher levels of engagement and companionship. While embodiment did not increase perceived authenticity, participants were comfortable with the robot in the structure of a human-designed meditation. These findings suggest that physical embodiment can meaningfully enhance meditation experiences through social presence, as long as the design avoids intrusiveness or false human equivalence.
Ellie Green, Andrew Blair, Mary Ellen Foster· International Conference on...· 0 citations
In the context of global cultural consumption upgrading and the rise of the emotional economy, traditional museum cultural and creative products face growing challenges in engaging contemporary audiences. Current academic research predominantly emphasizes the visual translation of cultural symbols, yet remains insufficiently attentive to users’deeper emotional engagement and affective connections. To address this issue, a strategic shift toward "emotional empowerment" has become essential. This study draws upon Donald Norman’s three-level theory of emotional design and adopts the Palace Museum’s cultural and creative products as a representative case study. It investigates how strategic design interventions facilitate users’ emotional progression—from initial sensory experience to sustained cultural identification. The study proposes an integrated emotional design framework structured across three interrelated levels: instinctive, behavioral, and reflective. At the instinctive level, visual appeal is enhanced through contemporary reinterpretation of traditional motifs; at the behavioral level, usability and interactive engagement are strengthened to support functional satisfaction; and at the reflective level, cultural metaphors and narrative depth are leveraged to cultivate users’ sense of belonging and cultural resonance. Collectively, this framework establishes a coherent, sequential pathway：“visual attraction—deepened experiential engagement—spiritual resonance”，thereby offering systematic, theory-informed design guidance for museum cultural products to evolve from object-centered storytelling toward meaningful, emotionally grounded user connections.
This study examined keroncong music as a culturally grounded therapeutic communication medium within mental health programmes for older adults in Cirebon. Through a qualitative case study approach, the participants’ experiences during keroncong-based activities were explored via in-depth interviews, observations, and document analysis. Data were analysed using Miles and Huberman’s interactive model. The findings indicate that keroncong facilitates socio-emotional well-being through three interrelated mechanisms: autobiographical memory activation, emotional regulation, and social interaction. Familiar musical elements stimulated reminiscence and collective storytelling, while regular participation was associated with increased emotional stability, enhanced social engagement, and reduced feelings of loneliness. In addition, keroncong functioned as a non-verbal communication medium that supported interpersonal connection within a culturally familiar context. This study conceptualises keroncong as a culturally grounded therapeutic communication medium that operates through non-verbal, collective, and memory-based interaction. A conceptual framework titled ‘Therapeutic Communication Based on Keroncong Music’ is proposed, highlighting its potential as a culturally relevant approach in elderly mental health programmes.
In the fast-paced and high-pressure conditions of contemporary life, chronic stress can aggravate anxiety and depression, increasing the need for gentle, everyday, non-pharmacological ways to support mind–body regulation. Grounded in Traditional Chinese Medicine, this study translates Five-Tone Therapy linking the pentatonic scale ( gong, shang, jue, zhi, yu) with five internal organs into an operable digital, music-based healing experience. Using the guqin’s aesthetic form as the central medium, we develop an interactive therapeutic system from an embodied cognition perspective to facilitate emotional regulation.The study adopts embodied cognition theory and a design research approach using user journey mapping. Key nodes of the therapeutic process, affective rhythms, and system response logic are identified and organized into five sequential stages: entering the application, completing a mind–body assessment, receiving system-recommended therapeutic strategies, conducting self-directed therapy, and reviewing therapy records and longitudinal trends. The paper focuses on personas, namely university students, working professionals, and retired older adults, and designs based on the common pain points shared by all these groups.Based on these findings, we prototype a multimodal system comprising a guqin-inspired therapeutic speaker, a companion mobile app, and tangible props representing the five organs. After a baseline assessment, users trigger corresponding Five-Tone playback by touching a prop with a built-in photosensitive sensor, synchronizing speaker lighting with app visuals to form a multisensory feedback loop. The outcomes offer practical insights for embedding cultural elements in therapeutic product design and for leveraging embodied interaction to enhance user experience and therapeutic effectiveness.
A new method, called CW-Net, translates the reasoning process of an autonomous vehicle’s AI system into understandable concepts that explain its behavior.
The professor of physics and inaugural director of the NSF AI Institute for Artificial Intelligence and Fundamental Interactions will lead LNS and continue his research in particle physics.
MIT News · Artificial Intelligence· news.mit.eduJun 30, 2026
Computer scientist Phillip Isola cuts through the hype to explain how AI agents work and what the future might hold for this rapidly advancing technology.
