You're saying the study has no grounding in how brains work? I'd think a more reasonable conclusion would be that the neuroscientists involved have no grounding in how artificial neural networks work.

It seems the whole point is to bring in additional details of how brains work, that the think may be relevant to artificial NNs.