fwiw..... from http://eec.ucdavis.edu/ACEEE/1994-96/1996/VOL10/085.PDF
this is just a tiny excerpt from the intro; read the study to get the rest of the story, but i have also read in other places that the r value of fg batts is somewhat temperature dependent. (in my case, it was reported that the colder it gets, the more the r value drops. not good.) how much of a change in the r value, and whether it is of practical significance, i don't know. i am not interested in reading more on fg insulation because i hate the stuff and will never use it again regardless of its merits or lack thereof. i've read and handled it enough. p.e.o.s. "Building energy models usually employ a constant, room-temperature-measured value for the thermal resistance of ﬁberglass roof insulation. In summer, however, the mean temperature of roof insulation can rise signiﬁcantly above room temperature, lowering the insulation’s thermal resistance by 10% to 20%."
Here is another link, showing the increase in thermal conductivity as the temp goes up.