Extracting Low-Dimensional Psychological Representations from Convolutional Neural Networks

Aditi Jha, Joshua C. Peterson, Thomas L. Griffiths

Research output: Contribution to journalArticlepeer-review

7 Scopus citations


Convolutional neural networks (CNNs) are increasingly widely used in psychology and neuroscience to predict how human minds and brains respond to visual images. Typically, CNNs represent these images using thousands of features that are learned through extensive training on image datasets. This raises a question: How many of these features are really needed to model human behavior? Here, we attempt to estimate the number of dimensions in CNN representations that are required to capture human psychological representations in two ways: (1) directly, using human similarity judgments and (2) indirectly, in the context of categorization. In both cases, we find that low-dimensional projections of CNN representations are sufficient to predict human behavior. We show that these low-dimensional representations can be easily interpreted, providing further insight into how people represent visual information. A series of control studies indicate that these findings are not due to the size of the dataset we used and may be due to a high level of redundancy in the features appearing in CNN representations.

Original languageEnglish (US)
Article numbere13226
JournalCognitive science
Issue number1
StatePublished - Jan 2023

All Science Journal Classification (ASJC) codes

  • Experimental and Cognitive Psychology
  • Artificial Intelligence
  • Cognitive Neuroscience


  • Categorization
  • Deep learning
  • Interpretability
  • Neural networks
  • Psychological representations
  • Similarity judgments


Dive into the research topics of 'Extracting Low-Dimensional Psychological Representations from Convolutional Neural Networks'. Together they form a unique fingerprint.

Cite this