The double empathy problem frames communication difficulties between neurodivergent and neurotypical individuals as arising from mutual misunderstanding, yet most interventions focus on autistic individuals. We present NeuroWise, a multi-agent LLM-based coaching system that supports neurotypical users through stress visualization, interpretation of internal experiences, and contextual guidance. In a between-subjects study (N=30), NeuroWise was rated as helpful by all participants and showed a significant condition-time effect on deficit-based attributions (p=0.02): NeuroWise users reduced deficit framing, while baseline users shifted toward blaming autistic "deficits" after difficult interactions. NeuroWise users also completed conversations more efficiently (37% fewer turns, p=0.03). These findings suggest that AI-based interpretation can support attributional change by helping users recognize communication challenges as mutual.
Figures & tables
Figure 1. Top: Conceptual comparison of Baseline Chatbot versus NeuroWise. Without support, the double empathy barrier perpetuates deficit-based framing; NeuroWise bridges this gap through three agents. Bottom: Example conversation with Alex showing the Baseline (left) and NeuroWise (right) with the support panel displaying stress level, interpretation, and coaching suggestions. Two-part figure. Top: Conceptual diagrams comparing Baseline Chatbot (showing barrier between Autistic Agent and Human User) with NeuroWise Chat Platform (showing bridge connection and three support agents). Bottom: Screenshot mockups of conversations with Alex about routine disruption, with Baseline showing only the chat and NeuroWise showing the chat plus a Support panel with stress level indicator, interpreter explanation, and coach suggestions.
Figure 2. Condition × Time interaction for deficit framing attitudes. NeuroWise participants showed slight improvement while Baseline participants regressed toward deficit framing after difficult conversations. Bar chart showing pre- and post-conversation deficit framing scores by condition. NeuroWise shows a decrease from 3.77 to 3.13 (delta = -0.63). Baseline shows an increase from 3.33 to 3.63 (delta = +0.30). The interaction is statistically significant (p = .020, delta = -0.49).
Autism Spectrum Disorder (ASD) affects more than 75 million people worldwide. However, scalable support for practicing everyday conversation is scarce: Low-cost activities such as story reading yield limited improvement. At the same time, effective role-play therapy demands expensive, in-person sessions with specialists. SocialWise bridges this gap through a browser-based application that pairs LLM conversational agents with a therapeutic retrieval augmented generation (RAG) knowledge base. Users select a scenario (e.g., ordering food, joining a group), interact by text or voice, and receive instant, structured feedback on tone, engagement, and alternative phrasing. The SocialWise prototype, implemented with Streamlit, LangChain, and ChromaDB, runs on any computer with internet access, and demonstrates how recent advances in LLM can provide evidence-based, on-demand communication coaching for individuals with ASD.
Empathy is central to human connection, yet people often struggle to express it effectively. In blinded evaluations, large language models (LLMs) generate responses that are often judged more empathic than human-written ones. Yet when a response is attributed to AI, recipients feel less heard than when comparable responses are attributed to a human. We built a conversation platform in which participants are asked to offer empathic support to an LLM expressing realistic troubles and conducted a randomized experiment collecting 33,938 messages spanning 2,904 text-based conversations between 968 participants and their LLM conversational partners. We find participants report feeling empathy but systematically fail to express it, but an LLM coaching intervention offering personalized feedback on effective empathic communication significantly boosts it without homogenizing participants' responses. Moreover, we derive a data-driven taxonomy of idiomatic empathic expressions in naturalistic dialogues across personal and workplace trouble scenarios. These results advance the scientific understanding of how empathy is expressed and demonstrate a scalable, AI-based intervention for scaffolding and cultivating it.
Aakriti Kumar, Nalin Poungpeth, Diyi Yang +2
Kellogg School of Management, Northwestern University. · Northwestern Institute on Complex Systems, Northwestern University. · Ryan Institute on Complexity, Northwestern University. +3
Evaluating persona-aligned empathy in LLM-based dialogue agents remains challenging. User states are latent, feedback is sparse and difficult to verify in situ, and seemingly supportive turns can still accumulate into trajectories that drift from persona-specific needs. We introduce EMPA, a process-oriented framework that evaluates persona-aligned support as sustained intervention rather than isolated replies. EMPA distills real interactions into controllable, psychologically grounded scenarios, couples them with an open-ended multi-agent sandbox that exposes strategic adaptation and failure modes, and scores trajectories in a latent psychological space by directional alignment, cumulative impact, and stability. The resulting signals and metrics support reproducible comparison and optimization of long-horizon empathic behavior, and they extend to other agent settings shaped by latent dynamics and weak, hard-to-verify feedback.