Moodialogue: Transforming Emotions into Personified, Conversational Agents

Empathy & Emotional DesignAffective Human-Computer DialogueBehavior Change & Reflection TechnologyPsychiatrists & PsychotherapistsHCI Researchers

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:
    1. Introduces emotion personification, enabling users to classify, label, and visually represent emotions.
    2. Facilitates dialogue with personified emotions, uncovering hidden causes, coexisting feelings, and contextual insights.
    3. 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.

Quick Actions

Share

Share this page

ios_share

https://hci.top/en/papers/chi/222600/2026

AdRecommended

Learn AI Coding at CodeNow

open_in_newOpen DOI Link
DOI: https://doi.org/10.1145/3772318.3790807
At a Glance

Paper Snapshot

fact_check
dataset
Source
CHI
calendar_month
Year
2026
emoji_events
Award
No award tagged
group
Authors
4 authors
sell
Subtopics
Empathy & Emotional Design, Affective Human-Computer Dialogue, Behavior Change & Reflection Technology
work
Professions
Psychiatrists & Psychotherapists, HCI Researchers
article
Content Status
Full text indexed
hub
Related Papers
2 related papers