NeuroBridge: Using Generative AI to Bridge Cross-neurotype Communication Differences through Neurotypical Perspective-taking

2025-09-27

AI Summary

Communication barriers between autistic and neurotypical individuals stem from bidirectional mutual misunderstanding, yet existing interventions often unilaterally demand autistic adaptation to neurotypical norms—distorting authentic interaction and increasing cognitive load. To address this, we introduce the first large language model (LLM)-based cross-neurotype communication simulation platform. It employs embodied AI agents and four prototypical dialogue scenarios to generatively model autistic communication—characterized by directness and literalism—from a neurotypical perspective for the first time. A feedback-driven interactive mechanism enables users to reflect on their interpretive biases. A user study with 12 neurotypical participants demonstrated significant improvement in understanding autistic linguistic features; feedback was rated as constructive, logically coherent, non-judgmental, and largely accurate in representational fidelity. This work advances bidirectional neurodiversity awareness and AI-augmented inclusive communication design.

Technology Category

• Humans and AI: AI for Accessibility
• Multiagent Systems: Agent Communication
• Machine Learning: Large Multimodal Models (LMMs)

Application Category

• User Modeling, Personalization and Recommendation: Large Language Models (LLM) for user modeling and recommendation
• Semantics and Knowledge: Data modeling to support human-machine intelligence, including LLMs agents, intelligent system behavior, explanations, and user-friendly interactions
• Search and Retrieval-Augmented AI: Multilingual and cross-lingual Web search

Abstract

Communication challenges between autistic and neurotypical individuals stem from a mutual lack of understanding of each other's distinct, and often contrasting, communication styles. Yet, autistic individuals are expected to adapt to neurotypical norms, making interactions inauthentic and mentally exhausting for them. To help redress this imbalance, we build NeuroBridge, an online platform that utilizes large language models (LLMs) to simulate: (a) an AI character that is direct and literal, a style common among many autistic individuals, and (b) four cross-neurotype communication scenarios in a feedback-driven conversation between this character and a neurotypical user. Through NeuroBridge, neurotypical individuals gain a firsthand look at autistic communication, and reflect on their role in shaping cross-neurotype interactions. In a user study with 12 neurotypical participants, we find that NeuroBridge improved their understanding of how autistic people may interpret language differently, with all describing autism as a social difference that "needs understanding by others" after completing the simulation. Participants valued its personalized, interactive format and described AI-generated feedback as "constructive", "logical" and "non-judgmental". Most perceived the portrayal of autism in the simulation as accurate, suggesting that users may readily accept AI-generated (mis)representations of disabilities. To conclude, we discuss design implications for disability representation in AI, the need for making NeuroBridge more personalized, and LLMs' limitations in modeling complex social scenarios.

Problem

• Bridging communication gaps between autistic and neurotypical individuals
• Reducing mental exhaustion from adapting to neurotypical communication norms
• Improving neurotypical understanding of autistic communication styles

Innovation

• Using generative AI to simulate autistic communication styles
• Creating interactive scenarios for cross-neurotype perspective-taking
• Providing AI-generated feedback to improve neurotypical understanding