Using natural language processing to support digital communication in adolescents and young adults with autism spectrum disorder.

BACKGROUND: Autism Spectrum Disorder (ASD) is characterized by social communication challenges, including difficulties interpreting figurative language, understanding conversational context, and expressing thoughts clearly. This feasibility study examined an NLP-based communication assistance tool for adolescents and young adults with ASD. The tool combined sentiment analysis, intent identification, and situational simplification to provide real-time feedback in digital communication. METHODS: Participants were 16 individuals with ASD (aged 15-28 years) and 16 neurotypical conversation partners in a within-subjects, mixed-methods design. The intervention lasted eight weeks. The tool was used in both structured tasks and naturalistic conversations. RESULTS: Preliminary data suggest associations between tool use and improved outcomes. Communication clarity increased from 3.12 to 3.89 (d = 0.74, 95% CI [0.28, 1.20], p < 0.001 after Bonferroni correction). Anxiety decreased by 2.3 points on a 7-point scale (d = 1.21, 95% CI [0.62, 1.80], p < 0.001). Confidence scores improved by 42% from baseline (d = 1.15, 95% CI [0.56, 1.74], p < 0.001). Qualitative thematic analysis (Braun & Clarke, 2006; inter-coder reliability, κ = 0.81) identified five themes: reduced communication burden, learning through use, empowerment and autonomy, context-specific value, and desire for customization. These preliminary findings suggest that NLP-driven assistive technologies warrant further investigation in controlled trials. CONCLUSION: This study demonstrates feasibility and provides preliminary evidence for NLP-based communication support, though causal conclusions cannot be drawn due to the absence of a control group, small sample size (N = 16), and short duration.