Investigating Creativity in Humans and Generative AI Through Circles Exercises

2025-02-11

AI Summary

Generative AI exhibits a “narrow creativity” bottleneck—akin to humans—characterized by overreliance on high-frequency, familiar, or incremental solutions within the design space, impeding breakthrough ideation. Method: We conducted an empirical study using the standardized “circle test” creativity assessment, coupled with LLM response sampling, frequency distribution analysis, and quantitative diversity metrics. Contribution/Results: This work provides the first systematic behavioral evidence that generative AI suffers from a human-like narrow creativity constraint. Advanced prompting techniques—such as chain-of-thought (CoT)—yield only marginal improvements in ideational breadth and fail to overcome this fundamental boundary. While GenAI can efficiently produce larger volumes of incremental concepts, it demonstrates no advantage over humans in generating truly novel or transformative ideas. These findings establish the first experimentally grounded benchmark and theoretical framework for designing AI-augmented creativity tools, highlighting the need to explicitly target breakthrough ideation rather than mere output volume or superficial variation.

Technology Category

• Cognitive Modeling & Cognitive Systems: Computational Creativity
• Humans and AI: Game Design — Procedural Content Generation & Storytelling
• Natural Language Processing: Generation

Application Category

• Economics, Online Markets and Human Computation: Economic ramifications for generative AI infrastructure and applications
• Social Networks and Social Media: Generative AI / large language models and their impact on social systems
• Search and Retrieval-Augmented AI: Agentic search

Abstract

Generative AI (GenAI) is transforming the creativity process. However, as presented in this paper, GenAI encounters"narrow creativity"barriers. We observe that both humans and GenAI focus on limited subsets of the design space. We investigate this phenomenon using the"Circles Exercise,"a creativity test widely used to examine the creativity of humans. Quantitative analysis reveals that humans tend to generate familiar, high-frequency ideas, while GenAI produces a larger volume of incremental innovations at a low cost. However, similar to humans, it struggles to significantly expand creative boundaries. Moreover, advanced prompting strategies, such as Chain-of-Thought (CoT) prompting, mitigate narrow creativity issues but still fall short of substantially broadening the creative scope of humans and GenAI. These findings underscore both the challenges and opportunities for advancing GenAI-powered human creativity support tools.

Problem

• Exploring creativity limitations in GenAI
• Comparing human and GenAI creative outputs
• Evaluating advanced prompting strategies' effectiveness

Innovation

• Generative AI creativity analysis
• Circles Exercise for creativity
• Chain-of-Thought prompting strategy