Emotional dialogue research includes two influential strategy traditions. Empathetic dialogue prioritizes understanding a speaker's emotional experience. Emotional support conversation selects and sequences support for the seeker's current needs. Sustained use introduces a further goal. Effective support should sustain users'capacities for emotion regulation, coping, self-endorsed decisions, and social connection across the interaction lifecycle. We propose capability-sustaining emotional dialogue (CSED) as a longitudinal research paradigm that aligns supportive strategy with this goal and organizes data, models, system design, evaluation, and governance around repeated use, non-use, transition, and termination. A targeted literature-and-corpus audit motivates this position. In a PRISMA-ScR-guided sample, 95% of 60 system-building papers pursue relief-oriented goals. None evaluates capability or longitudinal outcomes, and only 1 considers dependency, autonomy, or termination risk. In 300 ESConv supporter turns, capability-relevant functions appear in 43.0%, while generic suggestions account for 22.0%, compared with 4.0% reappraisal, 6.7% self-efficacy support, and 0.3% boundary behavior. We release a protocol for extending the audit to model behavior. An illustrative process model connects latent user capability to six design commitments, four evaluation timescales, and lifecycle constraints. The resulting agenda makes CSED testable across data, policy design, training, evaluation, and governance.
EmoTrace, a multi-turn dialogue corpus generation framework centered on modeling seekers' emotional trajectories, is proposed, which outperforms existing approaches in terms of emotional richness and empathy quality.
Large Language Models (LLMs) have substantially advanced persona-based dialogue agents for emotion-sensitive role simulation in healthcare, education, counseling, customer service, and interactive storytelling. However, two related lines of work leave a key gap. Persona-based dialogue systems often encode emotions as static traits or surface-level stylistic cues, and affective dialogue research has largely focused on empathetic response generation toward users rather than modeling the agent persona's own evolving emotional state. As a result, trigger-driven emotional evolution within a character remains underexplored. To address this limitation, we draw inspiration from the Component Process Model (CPM), a psychological theory that views emotion as a dynamic process shaped by the appraisal of external events. We propose CPM-MultiAgent, a CPM-grounded emotion evolution multi-agent framework for supporting emotional changes in persona-based dialogue. Instead of treating a character's emotion as a fixed attribute, CPM-MultiAgent represents it as a latent state that is continuously reshaped by dialogue triggers. Through affective trigger extraction, CPM-based collaborative appraisal, and emotion state updating, the framework enables more emotionally consistent role simulation in multi-turn interactions.Experiments with baseline comparisons, ablation studies, human evaluation, and case analyses demonstrate that CPM-MultiAgent effectively models dynamic emotional evolution in emotionally sensitive role-simulation settings.
Jingyao Cai, Shuaijun Liu, Abdul Rehman et al.· 0 citations
The advent of Large Language Models has accelerated interest in empathetic conversational agents. Despite a surge in empirical research, artificial empathy remains deeply fragmented, often serving as a catch-all term for diverse interactional phenomena. Addressing this conceptual gap, we systematically review 89 empirical studies to map how human-machine empathy is operationalized. Our synthesis reveals that empathy is highly situated and driven by functional goals, like health and well-being, transactional service, social interaction, and learning support. Within these contexts, we classify affective responsiveness by its directional flow, detailing how agents project, elicit, or mediate empathy. We structure the literature into a cohesive framework spanning linguistic, paralinguistic, identity, and architectural strategies. Furthermore, our methodological evaluation reveals a reliance on adapted clinical metrics, a scarcity of longitudinal studies, and a disproportionate focus on text-based over voice-based interfaces. Ultimately, this review equips researchers and practitioners with an actionable foundation for designing, measuring, and implementing contextually appropriate and empathetic agents.
Supriya Khadka, Smit Desai· International Conference on...· 0 citations
AI-supported deliberation platforms increasingly support large-scale participation in public policymaking. Yet their design reflects strong rationalistic and emotion reductionist bias: while they structure arguments, cluster opinions, visualize disagreement or organize discussions, they largely ignore the emotional dynamics through which participants interpret claims, react to disagreement and sustain engagement. This blind spot is particularly problematic in discussions surrounding wicked public issues, where emotions are not peripheral but integral to how actors evaluate arguments and orient themselves toward collective consensus-oriented decisions. At the same time, recent advances in Generative Artificial Intelligence (GenAI) introduce new possibilities for interacting with emotional signals in digital communication. Beyond content processing, GenAI systems can detect, interpret, and generate context-sensitive emotional expressions; this opens the possibility of GenAI-mediated emotional support in deliberative environments. This study explores how such systems could be designed. Following an echeloned Design Science Research approach, the paper focuses on the initial stages of a broader design project aimed at developing a GenAI mediator for consensus-oriented online deliberation. The problem space is examined through an analysis of 3 existing deliberation platforms and a semisystematic literature review of 25 papers on emotional dynamics in consensus-oriented discussions. Using a three-step thematic synthesis, the study derives design knowledge, articulated through design requirements, for emotion-aware GenAI mediation. The results of this study consist of two outcomes. First, the analysis formulates a problem statement that identifies a gap in current deliberation platforms: while they organize informational exchanges, emotional dynamics remain unmodeled. Second, the study derives six design requirements that define the design space a solution must satisfy: emotional awareness, emotional regulation, emotional inclusive design, emotional stability, emotional conflict transformation, and emotional articulation. These findings contribute to an initial body of design knowledge that bridges research on emotions in deliberation with emerging capabilities of GenAI; they lay the groundwork for designing emotion-aware GenAI-mediated deliberations.
Antoine Danthine, Anthony Simonofski· EGOV-CeDEM-ePart 2026· 0 citations
Vision-driven organizations rely on abstract purposes to guide action under uncertainty, particularly during second-foundation growth and AI/DX transformation. While such abstraction enables adaptability, it also creates structural ambiguity, often resulting in misalignment between declared vision and everyday practice.This study employs a qualitative action research approach based on long-term participatory engagement in three Japanese organizational contexts.The findings reveal a previously under-theorized mechanism: vision–practice misalignment is informally managed through emotional translation labor. Empathetic professionals translate abstract intent into operational meaning, absorb ambiguity, and maintain psychological safety. This invisible labor functions as an implicit subsystem that stabilizes performance in the short term while concentrating emotional load on specific individuals, increasing vulnerability to burnout and exit.A leadership reflexivity gap is also identified, in which emotional feedback is not effectively integrated into decision-making. Under high-context cultural conditions, this dynamic is amplified as misalignment is internally absorbed rather than explicitly addressed.These dynamics are conceptualized as the Organizational Emotional UX Breakdown Model, reframing emotional load as a hidden system-level constraint. To move from breakdown to structural resolution, the study proposes an Organizational Untangling framework grounded in Human Systems Integration (HSI), which redistributes emotional load through visualization, decision rationale externalization, reflexive feedback loops, and cross-functional integration. AI is positioned as an emotional mediator that supports sense-making and reduces dependency on individual-based emotional processing.By positioning emotional processes as organizational infrastructure, this study extends HSI and offers a design-oriented approach to enhancing organizational resilience.
As adolescents face mounting academic pressure, Social and Emotional Learning (SEL), though widely implemented, often slips into an “obedience trap”: one that emphasizes compliance over genuine student autonomy. Grounded in Self-Determination Theory (SDT) and a proposed four-tier “inside-out” model, this paper utilizes a narrative synthesis to critically examine four specific gratitude-based interventions: journaling, meditation, respiratory exercises, and emotional verbalization. The analysis shows that when these practices are offered as voluntary choices instead of top-down requirements, they effectively strengthen students’ self-regulation and psychological safety. Overall, this study holds that fostering students’ intrinsic motivation and emotional literacy requires abandoning coercive classroom management strategies and adopting instructional approaches that support learner autonomy. This research further indicates that such practices can only exert positive effects when they are delivered as engaging educational resources rather than compulsory classroom management tools. This paper provides practical pedagogical implications for the implementation of autonomy-supportive social-emotional learning. In addition, it identifies the lack of long-term empirical evidence and suggests conducting long-term longitudinal studies to verify the sustained impacts of relevant practices on students’ academic performance and personal development in regular secondary school settings.