Yes, I know I'm resurrecting an old thread, but in the interest of avoiding a bunch of threads that basically talk about the same thing, I will bring this zombie back to life.
Any the hoo...
Guess what I just found?
I just found a very interesting test done by the Department of Mechanics and Vibroacoustics, AGH University of Science and Technology Cracow, Poland and they used a testing procedure which included building test guitar setups - not normal guitars, but simplified versions using sapele, rosewood, plywood, and pine. They also used a mechanized plucking apparatus to ensure consistency of plucking the guitar setup. They recorded samples and analyzed the waveforms. Then they played those to test subjects to see if they heard a difference.
One point of the study was to determine if traditional tonewoods, which are becoming scarce, could be replaced by different woods to improve sustainability. Does the wood really matter to the sound?
The bottom line?
The varying test guitar setups did produce waveforms that differed from one another and those differences could be heard by the majority of test subjects, though some samples were harder to distinguish than others.
See link below for the study.
https://journals.pan.pl/Content/121810/PDF/aoa.2021.138150.pdf?handler=pdf
As for the mechanism of how this could happen? The study doesn't address this, but different materials will absorb energy from the strings or reflect vibrations back to the strings differently from one another. In the end, it all comes down to how those materials could influence the string's vibration within the pickup's magnetic field because that is all the pickups really pick up.
Now, I do notice that the materials used in the test were quite different - this was done on purpose to maximize the possibility of detecting a difference. That makes sense and I understand the rationale. However, the more similar one guitar is to another in terms of materials, the less difference there would be on sound. As stated above, the test did note that some materials were harder to distinguish than others. This point was taken to suggest that some traditional woods could be replaced by others that are more sustainable.
One interesting thing is that the tests were performed with both single coil and humbucking pickups. Listeners were better able to discern differences when listening to samples recorded with humbuckers.
Additionally, both musicians and non-musicians comprised the listeners. Overall, the musicians scored higher than non-musicians in discerning a difference; though, the non-musicians still did rather well.
It should be noted that construction methodologies are an additional variable which add a whole new level of complexity beyond just the wood. Plus, there is the question of how audible this would be in the context of a mix. In the test, the subjects used headphones comparing recordings. They weren't listening to these samples in a mix with a bunch of instruments.
Anyway, interesting stuff.