What is economic rationality?

What is rationality?

• I'm particularly interested in different definitions from different fields and philosophies. Would you please preface your answer with the perspective from which you're answering? For example, "In psychology rationality means..." or "Foucault defines rationality as..." etc

• Answer:

  I come from a behavioral economics background.  "Rational" is one of those words that can be a big headache, because there are simply far more concepts to which it is colloquially applied than there are nuances of the term. From my perspective, theoretic rationality means optimizing according to an objective function.  However, I think that the word is almost entirely ill-used in the fields of behavioral economics and psychology.  It is a theoretical concept, and has no meaningful place in experimental behavioral work.  An action can only be judged as rational if all information and all objective functions are perfectly defined.  This is only the case in "small-world" closed systems such as theoretical economic models.  It does not apply to open systems, or systems (such as the "real world") that are sufficiently complex as to approximate open. It is very trendy right now in my field to call people irrational, and to cite behaviors that support this.  However, that term must not be considered formal.  In the real world, complexity is such that we can never truly know someone's objective function, nor their information processing capabilities.  Thus, if behavior deviates from expectation, we cannot say if their objective function differs from expectation, or whether they are irrational with respect to their objective function. (Nor can we ever apply the term "rational" to the objective function itself.) Some behaviorists have noted this, and have done experiments to show  show "irrationality" by showing experimentally that people will break core axioms of rationality, such as the transitivity of preferences, or independence from random, seemingly unrelated stimuli.  However, this can still not truly be taken as irrational, given the limits on cognitive constraints.  Consider that when people are cognitively constrained, they cannot process all information, and must "decide how to decide."  Of course, this becomes iterative -- they must "decide how to decide how to decide..." and, if every iteration has even a small cost of attention, this becomes an unsolvable problem.  (An optimization procedure cannot optimize itself).  Thus many scholars believe that it has been "rational" (and by this I specifically mean evolutionarily advantageous) to develop heuristics -- quick decision rules that may not optimize the outcome of any particular decision, but that provide net positive return when applied consistently over many complex decision spaces.  When "large world" heuristics are considered, even the strongest examples of cognitive bias cannot be considered evidence for irrationality. In summary, I think that a gross number of ideas are misunderstood and a lot of headaches are cause due to this single word being used for dozens of theoretically distinct phenomena.   Although attaching my definition to the word is not critical, providing clarity to distinguish among concepts that are important elements of social science but that blur together under this single umbrella term, is. I have a great wish that several more words would be added to the English language to fill the needed gap and reduce confusion.