# Correcting Biased Centered Kernel Alignment Measures in Biological and Artificial Neural Networks arXiv: https://arxiv.org/abs/2405.01012 ยท HTML: https://mlsanity.com/paper/2405.01012/ ## Key facts - Cited by 18 papers in the map (data through 14 Oct 2026). - Builds on 8 of its 34 references that are in the map. - First posted on arXiv 2 May 2024. ## Builds on (most cited first, 8 of 8) - [Deep Residual Learning for Image Recognition](https://mlsanity.com/paper/1512.03385/) (2015) - [Similarity of Neural Network Representations Revisited](https://mlsanity.com/paper/1905.00414/) (2019) - [Do Wide and Deep Networks Learn the Same Things? Uncovering How Neural Network Representations Vary with Width and Depth](https://mlsanity.com/paper/2010.15327/) (2020) - [Getting aligned on representational alignment](https://mlsanity.com/paper/2310.13018/) (2023) - [Reliability of CKA as a Similarity Measure in Deep Learning](https://mlsanity.com/paper/2210.16156/) (2022) - [Geometric Understanding of Deep Learning](https://mlsanity.com/paper/1805.10451/) (2018) - [Partial Distance Correlation with Methods for Dissimilarities](https://mlsanity.com/paper/1310.2926/) (2013) - [The effect of task and training on intermediate representations in convolutional neural networks revealed with modified RV similarity analysis](https://mlsanity.com/paper/1912.02260/) (2019) ## Built on by (most cited first, 18 of 18) - [Revisiting the Platonic Representation Hypothesis: An Aristotelian View](https://mlsanity.com/paper/2602.14486/) (2026) - [Model Alignment Search](https://mlsanity.com/paper/2501.06164/) (2025) - [Estimating Neural Representation Alignment from Sparsely Sampled Inputs and Features](https://mlsanity.com/paper/2502.15104/) (2025) - [AWM: Accurate Weight-Matrix Fingerprint for Large Language Models](https://mlsanity.com/paper/2510.06738/) (2025) - [The Truth Lies Somewhere in the Middle (of the Generated Tokens)](https://mlsanity.com/paper/2605.09969/) (2026) - [Convergence Without Understanding: When Language Models Agree on Representations but Disagree on Reasoning](https://mlsanity.com/paper/2605.23315/) (2026) - [Spectral Analysis of Representational Similarity with Limited Neurons](https://mlsanity.com/paper/2502.19648/) (2025) - [Neural Correlates of Language Models Are Specific to Human Language](https://mlsanity.com/paper/2510.03156/) (2025) - [Confidence is Not Competence](https://mlsanity.com/paper/2510.24772/) (2025) - [Stimulus symmetries can confound representational similarity analyses](https://mlsanity.com/paper/2605.21324/) (2026) - [Seeing Through the Brain: New Insights from Decoding Visual Stimuli with fMRI](https://mlsanity.com/paper/2510.16196/) (2025) - [DAIT: Distillation from Vision-Language Models to Lightweight Classifiers with Adaptive Intermediate Teacher Transfer](https://mlsanity.com/paper/2603.15166/) (2026) - [Signal-to-Noise Ratio and Sample Size Govern Representational Alignment in Neural Networks](https://mlsanity.com/paper/2605.26973/) (2026) - [Resolving superposition in AI for interpretability and cross-modal alignment in patient-neuronal images](https://mlsanity.com/paper/2606.31394/) (2026) - [Virtual Encoders in Multimodal Transformers](https://mlsanity.com/paper/2609.26513/) (2026) - [Hidden Activations are not Enough I: Knowledge Matrices as Higher Representations](https://mlsanity.com/paper/2609.34166/) (2026) - [The System Prompt Illusion: How Instruction Preambles Modify Computation in Language Models](https://mlsanity.com/paper/2609.38205/) (2026) - [Measuring Learned Equivariance: An Information Theoretic Approach](https://mlsanity.com/paper/2610.17269/) (2026) ## Abstract (arXiv) Centred Kernel Alignment (CKA) has recently emerged as a popular metric to compare activations from biological and artificial neural networks (ANNs) in order to quantify the alignment between internal representations derived from stimuli sets (e.g. images, text, video) that are presented to both systems. In this paper we highlight issues that the community should take into account if using CKA as an alignment metric with neural data. Neural data are in the low-data high-dimensionality domain, which is one of the cases where (biased) CKA results in high similarity scores even for pairs of random matrices. Using fMRI and MEG data from the THINGS project, we show that if biased CKA is applied to representations of different sizes in the low-data high-dimensionality domain, they are not directly comparable due to biased CKA's sensitivity to differing feature-sample ratios and not stimuli-driven responses. This situation can arise both when comparing a pre-selected area of interest (e.g. ROI) to multiple ANN layers, as well as when determining to which ANN layer multiple regions of interest (ROIs) / sensor groups of different dimensionality are most similar. We show that biased CKA can be artificially driven to its maximum value when using independent random data of different sample-feature ratios. We further show that shuffling sample-feature pairs of real neural data does not drastically alter biased CKA similarity in comparison to unshuffled data, indicating an undesirable lack of sensitivity to stimuli-driven neural responses. Positive alignment of true stimuli-driven responses is only achieved by using debiased CKA. Lastly, we report findings that suggest biased CKA is sensitive to the inherent structure of neural data, only differing from shuffled data when debiased CKA detects stimuli-driven alignment.