Serendipity by Design: Evaluating the Impact of Cross-domain Mappings on Human and LLM Creativity

2026-03-19

AI Summary

This study systematically investigates differences between human and large language model (LLM) creative generation mechanisms under cross-domain analogical intervention. Combining controlled psychological experiments with prompt engineering, the research guided both human participants and LLMs to generate novel functionalities for everyday products using either distant analogical sources (e.g., octopus, cactus) or user needs as inspiration. Results reveal that humans significantly benefit from cross-domain mapping, whereas LLMs exhibit higher baseline originality but show insensitivity to such analogical prompting. Notably, both groups demonstrate enhanced creativity when the semantic distance between the analogical source and target domain is greater. This work provides the first empirical evidence of semantic distance as a moderating factor in cross-domain ideation effectiveness, offering foundational insights into the similarities and differences between human and artificial creativity.

Technology Category

• Cognitive Modeling & Cognitive Systems: Computational Creativity
• Application Domains: Humanities & Computational Social Science
• Humans and AI: Interaction Techniques and Devices

Application Category

• Economics, Online Markets and Human Computation: Humans versus LLMs for data annotation and labeling
• User Modeling, Personalization and Recommendation: Large Language Models (LLM) for user modeling and recommendation
• Semantics and Knowledge: Data modeling to support human-machine intelligence, including LLMs agents, intelligent system behavior, explanations, and user-friendly interactions

Abstract

Are large language models (LLMs) creative in the same way humans are, and can the same interventions increase creativity in both? We evaluate a promising but largely untested intervention for creativity: forcing creators to draw an analogy from a random, remote source domain (''cross-domain mapping''). Human participants and LLMs generated novel features for ten daily products (e.g., backpack, TV) under two prompts: (i) cross-domain mapping, which required translating a property from a randomly assigned source (e.g., octopus, cactus, GPS), and (ii) user-need, which required proposing innovations targeting unmet user needs. We show that humans reliably benefit from randomly assigned cross-domain mappings, while LLMs, on average, generate more original ideas than humans and do not show a statistically significant effect of cross-domain mappings. However, in both systems, the impact of cross-domain mapping increases when the inspiration source becomes more semantically distant from the target. Our results highlight both the role of remote association in creative ideation and systematic differences in how humans and LLMs respond to the same intervention for creativity.

Problem

• creativity
• large language models
• cross-domain mapping
• human-AI comparison
• serendipity

Innovation

• cross-domain mapping
• creativity
• large language models
• remote association
• human-AI comparison