In the past we have always done the impact RT/dry and then conditioned the specimens. When impacting structural parts for cert testing it has had to be RT/AR (as received) if they're too big to oven-dry.
Most of the damage (cracks) in the resin is propagated by mode I and mode II (and of course mixed mode I/II). If the critical fracture toughness is increased when hot/wet this would be probably be suffucient justification for the procedure, along with the fact that the vast majority of field damage won't be done in extreme conditions (often -50°C and +100°C).
Most matrix-type epoxies show increasing toughness with temperature. (Googling toughness temperature epoxy will give a few hits.)
Ideally, the material system would be checked for modes I and II toughness at different temperatures (including ones <RT) both dry and wet. However, testing this thorough is rare (I don't know of any, myself). I guess you could check the actual CAI strength for all different conditions to find the most critical, but that would be a tad expensive.
