One (possible/maybe) outcome in colder air (when training indoors in particular) is maybe slightly better results.
This is due to the amount of heat management the body has to do - if we’re producing ‘n’ watts at the pedals then we have to produce something like x2/x3 watts (can’t remember the multiple) and then we have to disperse that energy. One of the coaches write an article on this somewhere and if I can find it I’ll post it. And it may have been an article about other aspects than this subject precisely.
What I’ve never seen is any tested evidence at the ‘cold’ end of the spectrum … as I know performance drops once we can’t handle the heat build up.
So maybe once we get below a ‘certain point’ it makes no difference. Or maybe it’s better.
And then when we reach frozen bodies and we can’t physically turn the pedals, well I guess it goes all wrong at that extreme too.