Regions of Rationality: Maps for Bounded Agents

Abstract

An important problem faced by boundedly rational agents is to identify "regions of rationality," i.e., the areas for which simple, boundedly rational models are and are not effective. To map the contours of such regions, we derive probabilities that models identify the best of m alternatives (m > 2) characterized by k attributes (k > 1). The models include a single variable (lexicographic), variations of elimination-by-aspects, equal weighting, hybrids of the preceding, and models exploiting dominance. We compare all with multiple regression. We illustrate the theory with twenty simulated and four empirical datasets. Fits between predictions and realizations are excellent. However, the terrain mapped by our work is complex and no single model is "best." We further provide an overview by regressing the performance of the different models on factors characterizing environments. We conclude with suggestions for further research as well as implications for the concept of rationality in economics.

From intrapsychic to ecological theories in social psychology: Outlines of a functional theory approach

“…Historians of science may remember that Egon Brunswik's (, ) early visions of an environmental research program, proposed 80 years ago, were never adopted by a broader scientific community, apparently because it did not fit the mainstream of experimental psychology. However, I not only hope but also confidently believe that it will not take again such a long time until an environmental approach begins to flourish, simply because the intrapsychic program appears to be exhausted and because some of the carriers of a new environmental research program (Chater & Oaksford, ; Denrell & Le Mens, ; Gigerenzer, Hertwig, & Pachur, ; Herzog & Hertwig, ; Hogarth & Karelaia, ) are more influential than Brunswik was in the 1950s in Berkeley.…”

Heuristics as adaptive decision strategies in management

“…In a compensatory environment, a cue can be outweighed, or compensated for, by an individual cue or combination of other cues. In a series of articles, Hogarth and Karelaia (, , ) characterized the statistical properties of the environment and under what conditions a particular type of decision strategy performs best in prediction. They showed that even in an environment that is characterized by risk (as opposed to uncertainty) and where only noise is added, there is no all‐purpose tool such as linear regression that performs best across environments.…”

A Geometric Analysis of When Fixed Weighting Schemes Will Outperform Ordinary Least Squares

“…This framework is therefore well-suited as a tool of comparison for a wide range of decision models that can be described as fixed weighting schemes. There has been considerable interest on this topic (Baucells, Carrasco, & Hogarth, 2008;Davis-Stober et al, 2010a;Fasolo et al, 2007;Goldstein & Gigerenzer, 2002;Hogarth & Karelaia 2005a, 2005b, 2006a, 2006bKatsikopoulos & Martignon, 2006;Martignon & Hoffrage, 1999;Payne, Bettman, & Johnson, 1993) with most studies examining a range of different criteria.…”