Moodialogue: Transforming Emotions into Personified, Conversational Agents
Authors
Paper Title
Moodialogue: Transforming Emotions into Personified, Conversational Agents
Publication Info
- Topic area: Emotion journaling and reflection through interactive systems
- Keywords: emotion personification, conversational agents, emotional reflection, mood tracking, journaling, self-awareness, LLM-based systems, cognitive reappraisal, emotional well-being, human-computer interaction
Background and Problem
- Problem / challenge: Existing mood-tracking systems focus on recording emotions as isolated events, lacking tools to help users reinterpret emotional contexts or revisit emotions over time. This limits deeper reflection and meaning-making.
- Significance: Understanding emotions in context and over time can improve emotional awareness, regulation, and well-being. Tools that support such reflection can help users uncover hidden emotional causes and reframe experiences constructively.
- Motivation and related work: Prior research highlights the benefits of user-defined emotional expression and cognitive reappraisal but lacks systems that enable ongoing engagement with recorded emotions. Existing LLM-based journaling tools focus on single-session reflection, leaving gaps in supporting longitudinal emotional reinterpretation.
Solution
- Proposed approach: Moodialogue, a mobile journaling system that allows users to record, personify, and engage in dialogue with their emotions, transforming them into interactive agents for sustained reflection.
- Novelty:
- Introduces emotion personification, enabling users to classify, label, and visually represent emotions.
- Facilitates dialogue with personified emotions, uncovering hidden causes, coexisting feelings, and contextual insights.
- Shifts emotion recording from a one-time act to an iterative process of reflection and meaning-making.
- Procedure and key techniques:
- Users select an emotion from Plutchik’s eight basic categories, describe it in their own words, and create a visual profile.
- Personified emotions are stored as agents, allowing users to engage in dialogues that explore emotional contexts and triggers.
- Conversations are converted into journal drafts, which can be revisited and refined over time.
- The system uses a LangChain-based LLM pipeline to generate nuanced emotional personas and guide reflective dialogues.
Results
- Concrete findings:
- 226 dialogue sessions were conducted, averaging 10.32 conversational turns per session.
- 125 emotions were personified (70 positive, 55 negative), with weekly re-engagements increasing over time.
- Participants reported uncovering overlooked emotional contexts, coexisting feelings, and reframing negative emotions constructively.
- Advantage over baselines:
- Moodialogue enabled deeper emotional reflection compared to traditional journaling by integrating personification and dialogue.
- Users revisited and evolved their understanding of emotions over time, which is not supported by existing mood-tracking tools.
- Experiments / evaluation:
- Six-week field study with nine participants who frequently recorded emotions.
- Data collected through interviews, chat logs, and journal entries.
- Participants highlighted benefits such as increased emotional awareness, comfort, and insights but noted limitations in capturing multiple emotions and evolving personas.
- Limitations and future work:
- Limited to single-event-focused dialogues, restricting the capture of daily emotional diversity.
- Personas remained static, lacking mechanisms to evolve with user insights.
- Small, age-homogeneous sample limits generalizability; future work will involve larger, more diverse groups and explore therapeutic efficacy.
Summary
Moodialogue is a mobile journaling system that transforms emotion recording into an iterative process of reflection by enabling users to personify and converse with their emotions. A six-week field study demonstrated that Moodialogue helped participants uncover hidden emotional causes, recognize coexisting emotions, and reframe experiences constructively. The system’s design highlights the importance of post-recording interactions for sustained emotional reflection. Future work will address limitations such as static personas and single-event focus, aiming to support more nuanced and longitudinal emotional engagement. Moodialogue offers a promising approach to fostering emotional awareness, growth, and meaning-making.
Research Questions / Practical Problems
Question signals indexed for this paper.
- 83%
Venus and Mars on Canvas: AI-Mediated Collaborative Drawing for Romantic Relationship Insight
CHI '26· Affective Human-Computer Dialogue +3
- 67%
Mental Health Impacts of AI Companions: Triangulating Social Media Quasi-Experiments, User Perspectives, and Relational Lens
CHI '26· Affective Human-Computer Dialogue +2
Based on Jaccard similarity of research subtopics & professions (≥60%)