AHFE Open AccessFrom Memory to Medium: Investigating How Generative AI–Based Dream Recreation Supports Emotional Restoration and Self-Continuity
Abstract
In the digital age, young individuals face both emotional stress and fragmented self-narratives. Positive dreams may contain clues for repairing self-identity and enhancing life continuity, yet their fleeting nature makes them difficult to integrate. Current technological solutions fail to effectively support users in visually reconstructing meaning from dreams.This study proposes the Narrative Self-Integration (NSI) framework, using generative AI to guide users from passive recorders to active self-narrative editors. Developed through qualitative interviews (N=16) and validated through an experimental study (N=17), the framework demonstrates significant enhancement of positive emotions and sense of self-continuity.Theoretically, this work repositions generative AI as a constructive tool for self-narrative coherence. Practically, it provides a concrete pathway for next-generation digital mental health interventions.
Keywords: Generative Artificial Intelligence, Dream Interaction, Sense Of Self-continuity, Emotional Repair, Narrative Integration, User-centered Design, Digital Mental Health
DOI: 10.54941/ahfe1007341
Cite this paper
More from this volume
- An Embodied Interaction System for Five-Tone Music Therapy: A Guqin-Inspired Multimodal Design
- Beyond Function: An Analysis of Affective Design Factors in Japanese Mechanical Watches with High Auction Prices
- Environment Providing Necessary Information to Users Using Multiple IoT Avatars
- i-EyFuze: An Eye-Shaped eHMI in Autonomous Vehicles that Provides Intentions for Pedestrians
- Voice-Based Human Relaxation Assessment Using Autoencoder-Driven Anomaly Detection of Calm Speech
- Feasibility study of estimating visuospatial cognition and mental states using eye movement and brain activity during domain-specific tasks
- Deep Learning of Latent Gaze Representations for Cognitive Ability and Mental State Estimation
- Lightweight Driver Drowsiness Detection Model Using MediaPipe Blendshapes and a Dual-Attention Hierarchical BiLSTM
- Estimating 3D Ground Reaction Forces During Gait Using a Deep Learning Model with IMU and Plantar Pressure Data
- Integrating SOR and TAM Models to Explore Consumer Emotions and Preferences in Fur Fashion Design
- Effect of Changing Task Sequence on Physical Workload in Agricultural Operations
- Influence of Social Appearance Attributes of Cyber Driving Support Agents on the Passenger Effect