Moreover, it does not settle whether the mind is an epiphenomenon of the physical brain or whether the physical brain is the mechanism of mind (like a transmitter/receiver). I know of no clinical test which could falsify either worldview.
RE js1138's statement: If memories have to be reconstructed everytime we think of them, I would think that this would make thinking an enormously inefficient process. Yet in general, Nature is "parsimonious" -- inefficiency is not rewarded and is often penalized.
I'd also like to point out that the way an experiment is designed -- especially regarding its basic assumptions which may not be explicitly clarified -- can have an effect on the conclusions that can be reached by means of that particular experiment. FWIW
Plus I visit the old "filing cabinet" all the time, whenever I need a piece of information useful to solving a problem I'm working on. I literally can execute a "file search" and get a timely response with the memories I need in order to reason and analyze current problems. Based on such experiences, I strongly doubt that the "filing cabinet model" is "mythical."