Downloads & Free Reading Options - Results

ChatGPT, powered by a large language model (LLM), has revolutionized everyday human-computer interaction (HCI) experiences since its 2022 release. While now used by millions around the world, a coherent framework for evaluating the user experience (UX) ChatGPT offers remains missing. In this rapid review of the literature, I will investigate how ChatGPT UX has been approached quantitatively so far. I will focus on the independent variables (IVs) manipulated, the dependent variables (DVs) measured, the methods used for measurement, and the relationships between IVs and DVs. Findings are expected to reveal current trends, gaps, and emerging consensus in UX assessments of ChatGPT as a representative LLM-based chatbot. I aim to advance standardization in approaches to ChatGPT UX evaluation, elevate the field of ChatGPT UX, and empower researchers and practitioners in optimizing user interactions with ChatGPT and similar LLM-based systems. Note that this is based on an earlier rapid review for voice UX: https://osf.io/423pq