An important point of this paper:
Until recently, economists spent little time or effort trying to understand the cognitive processes which led to observed economic behavior, relying instead, either implicitly or explicitly, on the assumption articulated by [Milton] Friedman that people could easily learn optimal behavior through trial and error.
As if to say, those who trumpet economic game theory, the so-called free market, have little more than jingo in their arsenal.
A more important finding in this paper:
The second contribution of the paper is to show that even when the goal is to learn only this simple approximation, pure trial-and-error learning requires an enormous amount of experience to allow consumers to distinguish good rules from bad ones – far more experience than any one consumer would have over the course of a single lifetime.
In other words, not one of us will live long enough to find out if relying upon ‘utility theory’ and ‘free market theory’ is workable.
Their final thought has me chuckling. We may have already walked off a cliff:
Elucidating the circumstances under which a process of social learning can be expected to lead the population to reasonably optimal behavior will be an interesting task for future work.
pdf here, “Individual Learning About Consumption” by Christopher Carroll and Todd Allen at John Hopkins, written for the journal Macroeconomic Dynamics.