EUVIMEDSkopri EUVIMED
Open access · CC BY 4.0

Large Language Models Estimate Fine‐Grained Human Color–Concept Associations

Kushin Mukherjee, Ankit Mohapatra, Timothy T Rogers, Karen B Schloss

Cognitive Science · 2026

DOI: 10.1111/cogs.70219 ↗

At a glance

Journal article in Cognitive Science, 2026. Written by Kushin Mukherjee and 3 co-authors. Institution: Department of Psychology, Stanford University. Major MEDLINE topics: Color Perception, Concept Formation, Language, Large Language Models. Cited 2 times so far (Crossref); 94 references. The full text is freely available (open access, CC BY 4.0). Funded by: National Science Foundation.

AstrattSubjectsDettalji biblijografiċiCiteRelated studies

Abstract (original language)

People reliably associate the meanings of both abstract and concrete words with colors distributed over color space, a phenomenon that influences aspects of visual cognition ranging from object recognition to interpreting information visualizations. Prior research has hypothesized that color–concept associations arise from the cross‐modal statistical structure of experience, but it remains unclear whether natural environments contain such structure or whether learning systems can discover it without strong prior constraints. To address these questions, we investigated whether GPT‐4, a multimodal large language model, can estimate color–concept association ratings that approximate those made by people. We tested 71 colors spanning perceptual color space and a variety of concepts varying in abstractness. GPT‐4 ratings correlated strongly with human ratings across a range of prompting strategies, outperforming prior state‐of‐the‐art methods for automatically estimating color–concept associations from images. In an empirical study assessing people's ability to interpret the meanings of colors in information visualizations, palettes generated from GPT‐4's rating data were not only interpretable but, in some cases, more effective than those based on human ratings. Taken together, our results suggest that high‐order covariance between language and perception, present in web‐scale data, provide sufficient information to learn color–concept associations without initial constraints, and that machine‐derived associations can support the optimization of information visualizations for visual communication.

Subjects

MeSH terms (MEDLINE)

Color PerceptionConcept FormationLanguageLarge Language ModelsFemaleHumansColorGenerative Artificial Intelligence

Author keywords

Color semanticsColor‐concept associationsVision‐language models

Authors and institutions

  • Department of Psychology, Stanford University.
  • Wisconsin Institute for Discovery, University of Wisconsin-Madison.
  • Department of Computer Science, University of Wisconsin-Madison.
  • Department of Psychology, University of Wisconsin-Madison.

Funding

  • National Science Foundation

Cite

APA

Mukherjee, K., Mohapatra, A., Rogers, T. T., & Schloss, K. B. (2026). Large Language Models Estimate Fine‐Grained Human Color–Concept Associations. Cognitive Science. https://doi.org/10.1111/cogs.70219

Vancouver

Mukherjee K, Mohapatra A, Rogers TT, Schloss KB. Large Language Models Estimate Fine‐Grained Human Color–Concept Associations. Cognitive Science. 2026. doi:10.1111/cogs.70219

BibTeX
@article{Mukherjee2026,
  author = {Mukherjee, Kushin and Mohapatra, Ankit and Rogers, Timothy T and Schloss, Karen B},
  title = {Large Language Models Estimate Fine‐Grained Human Color–Concept Associations},
  journal = {Cognitive Science},
  year = {2026},
  doi = {10.1111/cogs.70219}
}
Start for free →