ABSTRACT
This chapter describes the rationality debate and proposes models of bounded rationality. It shows empirical evidence that humans – under constrains of time and computational costs – choose simple heuristics for decision that are both fast and frugal, without falling too far behind the Bayesian benchmarks. Herbert Simon postulated in the early fifties of the past century that humans are not rational, but boundedly rational. This can mean both that the aspiration levels of human decision are much lower than those proposed by “unbounded” rationality and also that the algorithms for decision are simpler and humbler than the Bayesian paradigm suggests. Humans have to make decisions under constraints of limited time, limited knowledge and limited computational resources. The sense of agency results from the observed feedback of our capacity to filter out and plan our mental and physical engagements, a second order kind of control structure.
