From this study I don't think it's even clear why or under what conditions higher REM latency is better, or that increasing it in an individual can necessarily

From this study I don't think it's even clear why or under what conditions higher REM latency is better, or that increasing it in an individual can necessarily be considered an improvement. (I also don't think that's what the study was for; I think they were mostly trying to validate the methods against established results.)