A fixed test did not confirm taxonomic geometry beyond animacy and visual controls in 14 SEEG patients and five fMRI adults

We tested whether an object taxonomy predicts neural response geometry beyond coarse animacy, a visual model and additive category terms. Separate public studies supplied human SEEG and BOLD fMRI. Development used three patients. A local code lock preceded tests in 14 other patients and all five MRI adults. Different pictures estimated each half of a category pattern. The panels had 50 clinical categories and 150 MRI categories. Mean partial correlations were 0.01483 and 0.00364. Their 95% person Student intervals were [-0.01012, 0.03977] and [-0.00651, 0.01379]. The fixed shared positive test failed, p=0.21875. Raw-distance means were 0.00347 and -0.00254. All negative people and fixed controls remain. Scores were larger without the animacy control, but these checks do not replace the failed test. Category deletion changed the sign of some MRI group means. An exact tree model gave rank alignment one for orthogonal response patterns with different sizes. Thus a favorable rank score alone cannot exclude response-size effects. Separate cohorts, unequal signal filters and spatial samples, few MRI people, source processing and an assumed clinical category order limit scope. Code, source versions and terms, person results, figures, audits and repeat steps are supplied. No positive cross-modal neural effect or causal mechanism was confirmed.

human neuroscience · SEEG · fMRI · object categories · WordNet · representational geometry · animacy · negative result