This is perhaps a bit presumptuous of me, since it is not really my field, but in this case the online evidence is overwhelming: there are few search results for "intrasubject variables" and "intersubject variables" and many for "between-subject variables" and "within-subject variables". The latter are clearly the usual ANOVA (analysis of variance) statistical terms. I don't see any point in quoting lots of examples; the information is easy to find. See here, for example:
http://onlinestatbook.com/2/chapter0/analysis_lab_anova.html


--------------------------------------------------
Note added at 1 hr (2017-02-17 14:07:40 GMT)
--------------------------------------------------

To be fair to biologists, it's not really their fault in this case; this is statistics rather than biology as such.

--------------------------------------------------
Note added at 2 hrs (2017-02-17 15:04:39 GMT)
--------------------------------------------------

I can't see that the jury's out, Lucy. When almost every professional who deals with this matter uses one pair of terms rather than another, I think the conclusion is clear. As always with online sources, it's not just how many, it's where they come from.

I think Phil's comment is beside the point. It's not a question of whether the alternative is "wrong"; it's a question of what do people in the field usually say. Only people who don't care whether informed readers are going to think "that sounds a bit odd!" are going to say it makes no difference.