# Evaluating alignment between humans and neural network representations in image-based learning tasks

arXiv: https://arxiv.org/abs/2306.09377 · HTML: https://mlsanity.com/paper/2306.09377/

## Key facts

- Cited by 6 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 15 Jun 2023.

## Built on by (most cited first, 6 of 6)

- [Post-training makes large language models less human-like](https://mlsanity.com/paper/2605.07632/) (2026)
- [The Lipreading Gap: Do VSR Models Perceive Visual Speech Like Human Lipreaders?](https://mlsanity.com/paper/2606.07435/) (2026)
- [Cross-modal Associations in Vision and Language Models: Revisiting the Bouba-Kiki Effect](https://mlsanity.com/paper/2507.10013/) (2025)
- [On the dynamic evolution of CLIP texture-shape bias and its relationship to human alignment and model robustness](https://mlsanity.com/paper/2508.09814/) (2025)
- [Investigating Fine- and Coarse-grained Structural Correspondences Between Deep Neural Networks and Human Object Image Similarity Judgments Using Unsupervised Alignment](https://mlsanity.com/paper/2505.16419/) (2025)
- [Scale-Invariance Drives Convergence in AI and Brain Representations](https://mlsanity.com/paper/2506.12117/) (2025)

## Abstract (arXiv)

Humans represent scenes and objects in rich feature spaces, carrying information that allows us to generalise about category memberships and abstract functions with few examples. What determines whether a neural network model generalises like a human? We tested how well the representations of 86 pretrained neural network models mapped to human learning trajectories across two tasks where humans had to learn continuous relationships and categories of natural images. In these tasks, both human participants and neural networks successfully identified the relevant stimulus features within a few trials, demonstrating effective generalisation. We found that while training dataset size was a core determinant of alignment with human choices, contrastive training with multi-modal data (text and imagery) was a common feature of currently publicly available models that predicted human generalisation. Intrinsic dimensionality of representations had different effects on alignment for different model types. Lastly, we tested three sets of human-aligned representations and found no consistent improvements in predictive accuracy compared to the baselines. In conclusion, pretrained neural networks can serve to extract representations for cognitive models, as they appear to capture some fundamental aspects of cognition that are transferable across tasks. Both our paradigms and modelling approach offer a novel way to quantify alignment between neural networks and humans and extend cognitive science into more naturalistic domains.