I've never seen any study that would help answer your intriguing question as to the age at which a child shifts from concrete interpretations of proverbs toward experience-based resymbolization. Over the next few weeks I will try to locate some references on the subject, including the original paper on proverb interpretation that piqued the interest of my colleagues so many years ago.
I've tested more than 2,000 adults with these three proverbs, and have come to the conclusion that interpretations are based on deeply held convictions about (in the case of these proverbs) attachment, shame/guilt, and distress-anguish. I had the pleasure of working in therapy some years ago with a psychotic psychologist who was an expert in the Rorschach test, but who assured me that no matter how hard she tried to make purposeful alterations in her responses when she herself had been tested, what came out of her was a fair representation of her inner life. The interesting point about Otto, as I mentioned above, is that even though any reasonable clinician might have expected him to respond to "rolling stone" in a way that showed that he eschewed intimacy, he actually responded in a way that indicated far more potential for intimacy than predicted on the basis of his presentation of self. Later, much later, when I began to reassemble the data of this case in terms of what I learned after it was "finished," I considered the possibility that he feigned madness/inadequacy in the arena of relatedness as well as in the degree of fear that seemed operative during his "acute illness."
For the record, I don't believe there is any such thing as a "latency" period of psychosexual development, and I don't think castration fears were operative in the triggering situation.