We are currently developing Personas for our website and as part of the validation process we wanted to see if usability participants would be able to identify which persona segment they most closely resemble. We compared what they said to how they interacted with the site and so far they don't seem to be matching very well.
There could be any number of reasons why this experiment would fail but I was wondering if there was any research that could support or discredit the idea that people can accurately identify their own persona segment.