After each response was centered across features, a fixed daily finger movement model gave positive mean alignment in six ECoG patients and six fMRI adults

We tested whether daily finger movement distances predict neural finger response distances. A separate public study of 22 adults set a fixed model from absolute MCP joint velocities. Other studies supplied ECoG voltage and 7 T fMRI response features. Development used two patients and one MRI adult. A local code and model lock preceded new scores in six other patients and six other adults. Each response was centered across features. Mean overall alignment was 0.57721 and 0.84164. Their 95% person t intervals were [0.12676,1.02766] and [0.76782,0.91546]. The shared person mean test passed Holm correction for two claims, p=0.021619. The stronger test removed additive finger terms and linear finger order. Its means were 0.01553 and 0.71556, with intervals [-0.51046,0.54153] and [0.50229,0.92882]. That shared test failed, p=0.471217. The overall ECoG interval included zero when feature means were kept. The overall two-claim sign-flip check also failed, p=0.0625. Five of six patients and all MRI adults had positive overall scores. All negative people and fixed checks remain. Equal task weight and object-use phases kept the broad result pattern. The external target was stable across separate behavioral groups. A second calculation matched all 297 scores within 3.8e-15. A fresh source replay matched saved tables and arrays. The contribution is a fixed external-model benchmark with a conditional broad relation and a failed stronger shared claim. Separate cohorts, clinical sampling, unequal tasks and sensing, source processing and six test people per modality limit scope. No common neural cause, detailed shared code or decoder gain is established. The package gives code, new model statistics, source pins and terms, all results, figures and repeat steps. Original neural arrays stay external.

human neuroscience · ECoG · fMRI · finger movement · representational geometry · external prediction · negative result · reproducibility