# Cognitive Psychology for Deep Neural Networks: A Shape Bias Case Study

arXiv: https://arxiv.org/abs/1706.08606 · HTML: https://mlsanity.com/paper/1706.08606/

## Key facts

- Cited by 46 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 26 Jun 2017.

## Built on by (most cited first, 30 of 46)

- [Deep Reinforcement Learning at the Edge of the Statistical Precipice](https://mlsanity.com/paper/2108.13264/) (2021)
- [Using cognitive psychology to understand GPT-3](https://mlsanity.com/paper/2206.14576/) (2022)
- [Language models show human-like content effects on reasoning tasks](https://mlsanity.com/paper/2207.07051/) (2022)
- [Can Subnetwork Structure be the Key to Out-of-Distribution Generalization?](https://mlsanity.com/paper/2106.02890/) (2021)
- [No Representation Rules Them All in Category Discovery](https://mlsanity.com/paper/2311.17055/) (2023)
- [Mechanistic Mode Connectivity](https://mlsanity.com/paper/2211.08422/) (2022)
- [POODLE: Improving Few-shot Learning via Penalizing Out-of-Distribution Samples](https://mlsanity.com/paper/2206.04679/) (2022)
- [Exploring Corruption Robustness: Inductive Biases in Vision Transformers and MLP-Mixers](https://mlsanity.com/paper/2106.13122/) (2021)
- [The Emergence of the Shape Bias Results from Communicative Efficiency](https://mlsanity.com/paper/2109.06232/) (2021)
- [ImageNet-trained CNNs are not biased towards texture: Revisiting feature reliance through controlled suppression](https://mlsanity.com/paper/2509.20234/) (2025)
- [Visual Search Asymmetry: Deep Nets and Humans Share Similar Inherent Biases](https://mlsanity.com/paper/2106.02953/) (2021)
- [Distinguishing rule- and exemplar-based generalization in learning systems](https://mlsanity.com/paper/2110.04328/) (2021)
- [Convolutional Neural Networks Rarely Learn Shape for Semantic Segmentation](https://mlsanity.com/paper/2305.06568/) (2023)
- [IFBiD: Inference-Free Bias Detection](https://mlsanity.com/paper/2109.04374/) (2021)
- [A Developmentally-Inspired Examination of Shape versus Texture Bias in Machines](https://mlsanity.com/paper/2202.08340/) (2022)
- [Understanding Segment Anything Model: SAM is Biased Towards Texture Rather than Shape](https://mlsanity.com/paper/2311.11465/) (2023)
- [Video Color Grading via Look-Up Table Generation](https://mlsanity.com/paper/2508.00548/) (2025)
- [Characterizing Verbatim Short-Term Memory in Neural Language Models](https://mlsanity.com/paper/2210.13569/) (2022)
- [Shape-Biased Domain Generalization via Shock Graph Embeddings](https://mlsanity.com/paper/2109.05671/) (2021)
- [Disentangling Abstraction from Statistical Pattern Matching in Human and Machine Learning](https://mlsanity.com/paper/2204.01437/) (2022)
- [Repaint: Improving the Generalization of Down-Stream Visual Tasks by Generating Multiple Instances of Training Examples](https://mlsanity.com/paper/2110.10366/) (2021)
- [InBiaseD: Inductive Bias Distillation to Improve Generalization and Robustness through Shape-awareness](https://mlsanity.com/paper/2206.05846/) (2022)
- [Gaussian Process Surrogate Models for Neural Networks](https://mlsanity.com/paper/2208.06028/) (2022)
- [A Survey of Dataset Refinement for Problems in Computer Vision Datasets](https://mlsanity.com/paper/2210.11717/) (2022)
- [A Procedural World Generation Framework for Systematic Evaluation of Continual Learning](https://mlsanity.com/paper/2106.02585/) (2021)
- [Towards a Psychology of Machines: Large Language Models Predict Human Memory](https://mlsanity.com/paper/2403.05152/) (2024)
- [CogLM: Tracking Cognitive Development of Large Language Models](https://mlsanity.com/paper/2408.09150/) (2024)
- [The Acquisition of Physical Knowledge in Generative Neural Networks](https://mlsanity.com/paper/2310.19943/) (2023)
- [Right for the Right Latent Factors: Debiasing Generative Models via Disentanglement](https://mlsanity.com/paper/2202.00391/) (2022)
- [Few-shot Shape Recognition by Learning Deep Shape-aware Features](https://mlsanity.com/paper/2312.01315/) (2023)

## Abstract (arXiv)

Deep neural networks (DNNs) have achieved unprecedented performance on a wide range of complex tasks, rapidly outpacing our understanding of the nature of their solutions. This has caused a recent surge of interest in methods for rendering modern neural systems more interpretable. In this work, we propose to address the interpretability problem in modern DNNs using the rich history of problem descriptions, theories and experimental methods developed by cognitive psychologists to study the human mind. To explore the potential value of these tools, we chose a well-established analysis from developmental psychology that explains how children learn word labels for objects, and applied that analysis to DNNs. Using datasets of stimuli inspired by the original cognitive psychology experiments, we find that state-of-the-art one shot learning models trained on ImageNet exhibit a similar bias to that observed in humans: they prefer to categorize objects according to shape rather than color. The magnitude of this shape bias varies greatly among architecturally identical, but differently seeded models, and even fluctuates within seeds throughout training, despite nearly equivalent classification performance. These results demonstrate the capability of tools from cognitive psychology for exposing hidden computational properties of DNNs, while concurrently providing us with a computational model for human word learning.