I guess on the animal model research side, the gap in various metrics between artificially "cleaning" the brains of sleep-deprived mice vs mice that get to slee

I guess on the animal model research side, the gap in various metrics between artificially "cleaning" the brains of sleep-deprived mice vs mice that get to sleep would be one way of measuring some of the non-cleaning functions of sleep (e.g. memory consolidation). In some far hypothetical future human device, I think even if amplifying a "washing" function doesn't replace sleep it could still be helpful ... but outweighing the risks involved in the intervention (attaching a person to a pump?) would be a high bar. But if decades from now you were already going to put in a neuralink v20, perhaps it would seem reasonable.