A fixed test did not confirm shared face-body reversal in eight EEG and 15 MEG people for 48 images

We tested whether the face-minus-body relation for human images points against the relation for animal images in separate human studies. Two EEG people supplied development. A local plan fixed tests in eight other EEG people and 15 MEG people. The latter cohort also supplied fMRI. Each of four image groups had twelve examples. A balanced correlation-distance contrast canceled additive image distinctness. Raw and appearance-adjusted evidence was required. Mean raw contrasts were -0.051730 in EEG, +0.021727 in MEG and +0.050926 in IT fMRI. Pointwise 95% person intervals were -0.075911 to -0.027549, +0.006131 to +0.037323, and +0.028097 to +0.073755. The adjusted MEG interval included zero. The fixed EEG-MEG reversal rule failed, with Holm p=0.9971. The separate positive fMRI rule passed, with Holm p=0.0003835. Both EEG repeat halves retained negative means. Every one-image deletion kept each study's mean sign. Raw EEG and MEG ranges included zero when people and images were resampled. A later narrow-window EEG audit was uncertain before adjustment. An exact model changed the contrast sign through sensor weights without changing latent patterns. Another control gave a negative adjusted score from a human-face-only coordinate with raw score zero. Neither model was fitted to the human outcomes. The result limits a shared reversal claim. It does not establish a shared positive neural effect or a causal modality difference. Unequal windows, tasks, cohorts, spatial sampling, source processing and unmatched image properties limit scope. The package gives code, permitted EEG derivatives, exact source versions and terms, local plans, all outcomes, figures, source checks and repeat steps.

EEG · MEG · fMRI · visual representation · face and body images · negative result · cross-modal test · sensor weights · reproducibility