Replies to message #573

The article is paywalled, but as far as I can tell, they did fMRI on a large number of people and gathered "connectivity profiles", basically it...

Yes. They used large datatsets of MRI data. "Our stDNN model was trained on a large sample (N ~ 1,000) of rsfMRI data from the Human Connectome...

Let's just say I'm highly sceptical.

The paper is paywalled and I cannot get hold of it. In principle though, I don't trust that the classification by AI reflects a real difference...