A fixed test did not confirm larger visual responses from half the exposure in five ECoG patients and six fMRI adults

We tested whether two short image pulses give a larger visual response than one continuous pulse with twice the image-on time. Separate public studies supplied ECoG broadband power and fMRI response estimates. The fixed comparison was two 133 ms pulses with a 533 ms gap versus one 533 ms pulse. Three ECoG patients supplied development. Tests used five other patients and six fMRI adults. Only one of five test patients had a positive contrast. The mean scaled difference was -0.198, with a 95% person bootstrap interval of -0.566 to +0.177. Two fMRI people had records in both source experiments, but their identity links were missing. Across all 72 links, the mean was -0.196 to -0.022 and the outer interval was -0.424 to +0.185. The shared one-sided sign p value was 0.96875. The primary claim failed. A planned descriptive check showed recovery at fixed exposure in both measurements. It does not confirm a benefit from half the exposure. Controls for baselines, response windows, spatial weights and image types remain. This is a new endpoint in reused cohorts, not a new acquisition. Finite electrical response tails, source selection, vascular readout and different stimuli limit scope. No new positive cross-modal effect or causal mechanism is claimed. The package gives a direct response-order benchmark, source code, permitted compact inputs, exact source records, all results, figures and repeat steps.

visual cortex · ECoG · fMRI · temporal summation · adaptation · response order · negative result · subject dependence