The robust beauty of ordinary information.

Abstract

Heuristics embodying limited information search and noncompensatory processing of information can yield robust performance relative to computationally more complex models. One criticism raised against heuristics is the argument that complexity is hidden in the calculation of the cue order used to make predictions. We discuss ways to order cues that do not entail individual learning. Then we propose and test the thesis that when orders are learned individually, people's necessarily limited knowledge will curtail computational complexity while also achieving robustness. Using computer simulations, we compare the performance of the take-the-best heuristic-with dichotomized or undichotomized cues-to benchmarks such as the naïve Bayes algorithm across 19 environments. Even with minute sizes of training sets, take-the-best using undichotomized cues excels. For 10 environments, we probe people's intuitions about the direction of the correlation between cues and criterion. On the basis of these intuitions, in most of the environments take-the-best achieves the level of performance that would be expected from learning cue orders from 50% of the objects in the environments. Thus, ordinary information about cues-either gleaned from small training sets or intuited-can support robust performance without requiring Herculean computations.

The Revenge of Ecological Rationality: Strategy-Selection by Meta-Induction Within Changing Environments

“…Our major result was that a scale transformation from the real-valued into the binary-comparative format worsened the success of TTB compared to weighting methods. This result agrees with the finding reported in Katsikopoulos et al (2010Katsikopoulos et al ( , p. 1265.…”

“…This explains why complex strategies, if based on small samples, frequently perform worse than TTB and other frugal prediction methods (cf. Brighton and Gigerenzer 2012, p. 36, 44f;Katsikopoulos et al 2010).…”

“…Within the adaptive rationality research program, these questions have not been asked, so far, presumably because cues were 4 Exceptions to (a) are Hoffrage et al (2000), Hogarth and Karelaia (2005) and Katsikopoulos et al (2010), who study prediction tasks based on continuous-valued cues. Exceptions to (b) are Dieckmann and Todd (2012), and Rieskamp and Otto (2006), who study prediction tasks in the course of online learning.…”

Why Do Simple Heuristics Perform Well in Choices with Binary Attributes?

“…Some work has been done on this (Hogarth and Karelaia 2005, Gigerenzer and Brighton 2009, Katsikopoulos et al 2010), but as of today, we still do not know how to make the synthesis.…”

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

“…Equal weights, where a is set equal to a vector of ones, is perhaps the most well-known (Dawes, 1979;Dawes & Corrigan, 1974;Einhorn & Hogarth, 1975;Schmidt, 1971;Wainer, 1976), although see Keren and Newman (1978) for a set of unfavorable conditions for equal weighting. Single-variable weighting schemes, where only one entry of a is nonzero, such as those based on decision models such as Take The Best (Gigerenzer, Todd, & the ABC Research Group, 1999) and the Recognition Heuristic (Goldstein & Gigerenzer, 2002) have also been shown to have superior predictive power over OLS estimation under various conditions (Czerlinski, Gigerenzer, & Goldstein, 1999;Davis-Stober, Dana, & Budescu, 2010a;Gigerenzer & Goldstein, 1996;Katsikopoulos, Schooler, & Hertwig, 2010). Bobko, Roth, and Buster (2007) and Goldstein and Gigerenzer (2009) provide recent reviews and comparisons of fixed weighting models as they compare to various prediction techniques, including OLS.…”