2026 ASEE Annual Conference & Exposition

This Empirical Research Paper (Full Paper) presents a comprehensive analysis of how neurodivergent individuals and communities express their lived experiences, challenges, and often-overlooked strengths across major social media platforms including X (formerly Twitter) and Reddit. Social media serves as an important emancipatory space where neurodivergent users construct self-defined identities, resist deficit-based language, and highlight strengths such as creativity, attention to detail, and visual-spatial reasoning, which are frequently undervalued in traditional educational and professional environments. Grounded in inclusive pedagogy, this study aims to connect digital discourse with educational inclusivity and awareness. The large-scale and unstructured nature of such discourse demands analytical frameworks capable of capturing both thematic depth and linguistic nuance. To address this need, we collected and analyzed over 90,000 posts from X and Reddit, applying a hybrid natural language processing pipeline. We used text representation learning methods such as TF-IDF, BERT, and OpenAI embeddings to convert posts into numerical formats, then applied clustering techniques to group similar conversations and examine how emancipatory language appeared in each cluster. To complement clustering, we used topic modeling methods such as Latent Dirichlet Allocation (LDA) and Top2Vec to uncover key recurring topics. Finally, GPT-based models were used to enhance interpretability, generate clear summaries, and identify deeper themes shared across the neurodivergent community online.

Our findings reveal distinct platform-specific communication patterns, with X content being concise and advocacy-oriented, emphasizing topics such as diagnosis, stigma, and self-acceptance, whereas Reddit discussions are substantially longer, introspective, and centered on detailed sharing of coping mechanisms, academic challenges, and mental health support. Across both platforms, several recurring themes emerge, including executive dysfunction, rejection sensitivity, sensory regulation, burnout, and the pursuit of accessible education and employment environments. These insights show that neurodivergent users, including students, leverage digital platforms not merely for self-expression but also for collective knowledge-building and emotional co-regulation. This study demonstrates a scalable framework for social media mining that blends statistical, semantic, and generative AI methods, amplifying the voices of neurodivergent individuals/students and providing actionable insights to help educators, policymakers, and designers to develop more inclusive learning environments.