Yes, it does make a difference. Everything else does too, so the important variable here is sample size.
With a sample of 1, or 4 or even 10 instruments, it can be hard to learn to hear the contribution of the fingerboard.
However, get upwards of 100 instruments (or, in my case, pushing 1000 that I have built in full or in part, with mostly my hands
assembling/setting up the lion's share of those), and you can start to understand the effects of each part.
For example - we used to build a model with wenge fingerboards. We built dozens like that. Then we built one with ebony by client request,
and heard the difference. Later, we would move to ebony as stock for that model. The change we'd heard the first time reappeared, and was
present on all models. Same thing with maple - whenever we would build one, a particular component of the sound would appear, which simply wasn't there
with other woods. Bear in mind, these models had otherwise VERY similar specs (body wood, thickness, weight, neck joint, pickups, preamp, fretwire, scale, strings, etc.).
This is precisely why the production guitar shop is such an ideal laboratory for this kind of experiment. We even used the same amp to test them all during setup.
It's easy for anyone to hear differences between instruments. The challenge comes in learning, over time and repeated exposure,
to identify where they originate. Yes, you can EQ in treble to a darker-sounding instrument. It will not ever sound quite the same as a natively bright instrument, though - at least, not to those who have learned to hear the difference.
Regarding the colorful descriptions of acoustic back/side woods above - I'll only say that everything matters there as well, and there are many many more variables at play with acoustics. In that world, the general rule is that a lousy builder can make anything sound lousy, even wonderful material.
(Or, in the case of a factory guitar - they can overbuild to the point where the woods' particular properties are less significant in their contribution.)
However, a good builder can exploit the little differences between back/side woods to very meaningful effect.
What's important is to remember that a properly controlled experiment simply cannot be done with a small sample. For example, the case described above
(identifying a maple vs. rosewood board from a recording alone) is basically impossible because it doesn't factor in the multitude of possible variables.
If the question were to isolate the differences between recordings of 50 maple fretboards and 50 rosewoods, then maybe it would become more legitimate as an experiment.
Cheers,
Martin