With fixed condition means and independent repeat errors, tests supported two residual motor map directions in 15 ECoG patients and 58 fMRI adults

We tested whether hand and tongue movement maps need two repeatable spatial directions beyond affine position trends. Separate public human studies supplied cortical ECoG power and BOLD fMRI. Four distinct repeat groups enter a determinant score. Two unused groups set its scale. For one true map, the expected score is zero if repeat errors are independent and condition means are fixed. Three patients and four MRI adults supplied development. A local code lock preceded new scores in 15 eligible reserved patients and 58 reserved MRI adults. Earlier SCI work used the same cohorts. Mean scores were 0.072561 [0.032013, 0.113110] and 0.070927 [0.062417, 0.079437]. Brackets give 95% person Student limits. The shared rule passed, p=0.00180949. All negative people and fixed controls remain. In an exact one-map example, independent errors gave mean zero, but a pooled noisy determinant gave 0.258253. Errors shared across groups gave 0.444444. Thus the test needs a group-error model; it cannot identify a neural source or cause. Sparse clinical coverage, different movements and cohorts, temporal drift, stable artifacts and source filters limit scope. The contribution is a checked four-group assay and a two-study benchmark. The package gives code, source versions and terms, scalar results, figures, audits and repeat steps. Original neural inputs remain external.

motor cortex · ECoG · fMRI · measurement error · dimensionality · independent repeats