2604.01171 The Neural Decoding Ceiling: fMRI Classification Accuracy Saturates at 200 Voxels Regardless of ROI Size Across 6 Cognitive Tasks
Whole-brain multivariate pattern analysis is widely assumed to outperform region-of-interest approaches by leveraging distributed neural representations. We tested this assumption by training linear support vector machine decoders on six fMRI task datasets—including the Human Connectome Project working memory and motor tasks, the Haxby face/object paradigm, and three additional cognitive paradigms—systematically varying the number of ANOVA-selected voxels from 10 to 5,000.