Don't attach too much weight to this. First, it's just two phonemes, and very similar ones at that. Second, many different results have been obtained in similar

Don't attach too much weight to this. First, it's just two phonemes, and very similar ones at that. Second, many different results have been obtained in similar studies. Third, "differently" is a rather trivial word: anything that gets associated with these sounds may or may not "light up" in an fMRI experiment. It's not even sure the effect from that paper are relevant to language processing. In this case, it might be something as simple as familiarity. Preciously little is known about language processing.