Neuro-Inclusive Generative AI: Cognitive Barriers, Coping Strategies, and Design Patterns in Prompt-Based Creative Tools

Prompt-based creative AI systems, diffusion-based image generators in particular, open new possibilities for visual expression while also imposing cognitive demands that the literature links to exclusion for some neurodivergent users. This structured narrative review draws on scholarship from 2020 to 2026 across human-computer interaction, cognitive accessibility, neurodiversity, and generative AI to examine how prompt-mediated interaction creates barriers for people with ADHD, autism, dyslexia, and related cognitive differences. This review proposes four barrier types grounded in convergent findings from adjacent HCI and accessibility research: working-memory overload from prompt construction and parameter management; executive-function burden during iterative refinement; sensory and language-processing friction in visually dense, text-centric interfaces; and opaque feedback and hidden prompting conventions that prevent users from diagnosing why outputs diverge from intention. Direct empirical evidence specific to diffusion tools remains thin. Each proposed barrier type warrants direct empirical testing with diffusion-tool users before the framework can be treated as validated. The paper proposes a three-layer taxonomy separating barriers users encounter, coping strategies they develop, and design patterns that future interfaces should test. Accessible creative AI, the review concludes, requires moving past compliance-based thinking toward co-designed, cognitively adaptive interfaces with progressive disclosure, multimodal input, editable prompt histories, interpretable feedback, and metrics built for the prompting workflow rather than borrowed from static interfaces.