³ÉÈËÍ·Ìõ

SCHÃœTT, H., Kipnis, A. D., Diedrichsen, J., & Kriegeskorte, N. (23 August 2023). Statistical inference on representational geometries.

Schütt and colleagues introduce a new method for statistical inference on representational geometries based on a cross-validated two-factor bootstrap that allows for generalization across both participants and stimuli while allowing the fitting of flexible models. In a series of elegant simulations and empirical analyses on existing datasets, the authors validate the method statistically. The work provides a fundamental and compelling advance for the analysis of representational geometries.