Taught by: Heiko Schütt, NYU (October 25, 2022)
Video: Tutorial on Statistical Inference on Representational Geometries
Description: Representational similarity analysis (RSA) is a popular method for comparing representations when a mapping between them is not available. One important comparison RSA is used for is between neuronal measurements and models of brain computation like deep neural networks. RSA is a two-step process. First, a matrix of pairwise dissimilarities between conditions is computed. This matrix is then a summary of the representational geometry, which can be compared directly between different representations as it has the same dimensions. In the first half of this tutorial, I will go through some recent advancements for RSA that improve the reliability and statistical accuracy of RSA substantially: First, I will explain the reasoning for cross-validated distance measures for computing the dissimilarity matrix and for whitened similarity measures to compare them to each other. Then, I will explain why simultaneous generalization to new subjects and new stimuli is hard and a solution based on bootstrapping. Finally, I will explain the necessary cross-validation-based extensions for flexible models. In the second half of this tutorial, I will give a guide on how to run these analyses using our new rsatoolbox in Python by going through demo notebooks that illustrate the functionality.
Additional Resources:
Relevant papers:
- Schütt et al., 2021: “Statistical Inference on Representational Geometries.” arXiv:2112.09200.
- Walther et al., 2016: “Reliability of Dissimilarity Measures for Multi-voxel Pattern Analysis.” Neuroimage 137: 188–200.
- Diedrichsen et al., 2021: “Comparing Representational Geometries Using Whitened Unbiased-Distance-Matrix Similarity.” arXiv:2007.02789.
GitHub Repositories:
