We favor formal models of heuristics rather than listsof loose dichotomies: a reply to Evans and Over

摘要

In their comment on Marewski et al. (goodjudgments do not require complex cognition, 2009) Evansand Over (heuristic thinking and human intelligence: acommentary on Marewski, Gaissmaier and Gigerenzer,2009) conjectured that heuristics can often lead to biasesand are not error free. This is a most surprising critique.The computational models of heuristics we have testedallow for quantitative predictions of how many errors agiven heuristic will make, and we and others have mea-sured the amount of error by analysis, computer simulation,and experiment. This is clear progress over simply givingheuristics labels, such as availability, that do not allow forquantitative comparisons of errors. Evans and Over arguethat the reason people rely on heuristics is the accuracy-effort trade-off. However, the comparison between heu-ristics and more effortful strategies, such as multipleregression, has shown that there are many situations inwhich a heuristic is more accurate with less effort. Finally,we do not see how the fast and frugal heuristics programcould benefit from a dual-process framework unless thedual-process framework is made more precise. Instead,the dual-process framework could benefit if its two "blackboxes" (Type 1 and Type 2 processes) were substitutedby computational models of both heuristics and otherprocesses.