I concur, given the speaker's general use case and its implications, adding some primarily voice-based material for the listening tests would be a really interesting and useful case to have some subjective data for.
Until now, I have seen many comments about lesser "Perfoming" centers, having a narrow range, but in my limited and not totally scientific "experiments" I have not found it really to matter as much as many think it will.
I may be in the minority here for sure, but my one 3 way Center, with really wide dispersion, to me, does not sound as good, possibly because voices appear more diffuse, or less pinpoint in where they are coming from.
I sense the "issues" many claim to see, actually are somewhat of an advantage.
I found when sitting far to one side, it appears LESS natural for a voice to not roll off in the upper registers.