Somewhat to my surprise, this turned out to be relatively simple. Questions like this (confirming a negative question) are often difficult to search for on the web. But both comments approached the search solution in the same way... by converting the question into a positive form.
An important part of having a robust search strategy is knowing what general approaches are likely to work out. I think it's pretty safe to say the asking for a positive is generally safer than searching for a negative form. The query:
[ synthesia consistency ] or [ synesthesia consistent ]
will get you to papers (or articles) that are about how consistent the experience is. That's a good move.
Of course, you then have to suss out where the confirmation of difference between perceptions are, but neither respondent seemed to have much trouble with that! (I know that Katie solved the problem in less than 2 minutes; I was way impressed.)
Another take-away message from this problem is that it's often difficult to accurately assess the difficulty of a search challenge. I know that I've mis-estimated any number of times... both ways. Problems I thought were tough turned out to be trivial; problems I thought would be simple turned out to be intractable.
Do you have any tough / intractable search problems? Send them to me! Maybe we'll use them for a future Search Challenge. (Or if they're less that intractable, they'll go into the AGoogleADay bucket!)