Social Navigation Support for Neurodivergent University Students

Project description

University students regularly face high-stakes social situations - group work conflict, ambiguous supervisor feedback, requesting extensions, technical interviews - that can be especially challenging for those with autism, ADHD, anxiety, or other neurodivergent profiles. Existing tools are largely designed for children and focused on deficit correction rather than context-specific navigation support. This project aims to develop a research prototype using AI-powered conversational practice to help neurodivergent CS/IT students build confidence in these situations. Suits students with interests in HCI, accessibility, CS education, or AI systems design. Possible areas for exploration include:

• Participatory scenario design: what situations do neurodivergent CS/IT students find hardest, and what makes AI feedback useful?
• Voice-based rehearsal: does speaking a response aloud change the rehearsal experience, and what feedback is meaningful at that level?
• Role-play character calibration: how can we tune an AI-played character (supervisor, teammate, interviewer) to be appropriately challenging without being distressing?
• LLM prompt architecture for ND-appropriate feedback: how should prompts be structured to produce feedback that is direct and neurodiversity-informed?
• Progress tracking without performance anxiety: how do you track progress without replicating graded assessment dynamics?

Note: You need to register interest in projects from different supervisors (not a number of projects with the one supervisor).
You must also contact each supervisor directly to discuss both the project details and your suitability to undertake the project.