Automated Creativity Assessment: A Review of Methods, Challenges, and Future Prospects

Creativity assessment has historically relied on human raters to judge the quality of ideas and products. However, this approach is exceptionally costly due to the limited number of experts, the time and resources required for training them, and the high task burden. To overcome these barriers and expedite creativity research, computational automation of creativity assessment has been explored. We provide an overview of recent developments in computational creativity assessment methods, including techniques based on artificial intelligence (such as large language models) and semantic distance. Leveraging these modern machine-learning approaches, computational creativity assessment methods have enabled automatic scoring across various tasks—from simple idea generation to complex problem-solving and artistic expressions—often matching human rater agreement. These methods also have the potential to automatically generate new creativity tests (“automated item generation”) and can be used to detect invalid responses (i.e., random/task-irrelevant ideas), thereby ensuring assessment integrity for high-stakes applications. We discuss the current challenges and potential future directions, such as improving the interpretability and explainability of automated scoring, and a complementary approach centered around evaluating mental processes leading to creativity.