Keywords
Generative AI, Problem Solving, Trust

Technologies
Generative AI, ChatGPT

Overview This project investigates how software developers think and make decisions when using Generative AI tools such as ChatGPT. Using a mixed-methods design combining experimental and qualitative approaches, participants completed programming tasks either with or without AI assistance while verbalising their thoughts. The study examines how developers' cognitive models of GenAI relate to their trust, self-efficacy, reliance, and task satisfaction, with the aim of determining whether GenAI enhances learning and productivity or fosters overreliance.

My Contribution My individual contribution focuses on how GenAI assistance shapes developer decision-making in real time, with a particular emphasis on verification and evaluation behaviours, specifically, how developers assess, question, and act on AI-generated output during live coding tasks. Using Verbal Protocol Analysis of think-aloud data, this analysis aims to surface patterns in how developers engage critically with AI suggestions, and what this reveals about autonomy and cognitive dependence in AI-assisted programming.