Work

Is Representational Similarity Analysits Reliable? A Comparison with Regression.

Year: 2025

Type: preprint

Abstract: Representational Similarity Analysis (RSA) is a popular method for analyzing neuroimaging and behavioral data. Here we evaluate the accuracy and reliability of RSA in the context of model selection, and compare it to that of regression. Although RSA offers flexibility in handling high-dimensional, cross-modal, and cross-species data, its reliance on a transformation of raw data into similarity structures may result in the loss of critical stimulus-response information. Across extensive simulation studies and empirical analyses, we show that RSA leads to lower model selection accuracy, regardless of sample size, noise level, feature dimensionality, or multicollinearity, relative to regression. While principal component analysis and feature reweighting mitigate RSA's deficits driven by multicollinearity, regression remains superior in accurately distinguishing between models. Empirical data and a follow-up fMRI simulation further support these conclusions. Our findings suggest that researchers should carefully consider which approach to use: RSA is less effective than linear regression for model selection and fitting when direct stimulus-response mappings are available.

Source: PubMed

Authors: Chuanji Gao, Gang Chen, Svetlana V Shinkareva, Rutvik H Desai

Institutions: Nanjing Normal University, National Institute of Mental Health, University of South Carolina

Language: English

Topic: Statistical Methods and Applications

Subfield: Statistics and Probability

Field: Mathematics

Domain: Physical Sciences

Open Access status: green

Funders: National Institutes of Health, National Institute on Deafness and Other Communication Disorders

Awards: Semantic Systems, Scientific and Statistical Computing Core

OpenAlex ID: W7118104683

Cites:

Cited by:

Related to: