Heuristics are Tools for Uncertainty

Abstract

Heuristics are commonly viewed in behavioral economics as inferior strategies resulting from agents' cognitive limitations. Uncertainty is generally reduced to a form of risk, quantifiable in some probabilistic format. We challenge both conceptualizations and connect heuristics and uncertainty in a functional way: When uncertainty does not lend itself to risk calculations, heuristics can fare better than complex, optimization-based strategies if they satisfy the criteria for being ecological rational. This insight emerges from merging Knightian uncertainty with the study of fast-and-frugal heuristics. For many decision theorists, uncertainty is an undesirable characteristic of a situation, yet in the world of business it is considered a necessary condition for profit. In this article, we argue for complementing the study of decision making under risk using probability theory with a systematic study of decision making under uncertainty using formal models of heuristics. In doing so, we can better understand decision making in the real world and why and when simple heuristics are successful.

Veteran‐civilian career identity conflict: What is human resource development's role?

“…For this, we use Schlossberg's (1981, 1984) adult transition theory—hereinafter Schlossberg's theory—to offer a heuristic (cf. Mousavi & Gigerenzer, 2017) model to support veterans during military‐to‐civilian career transition—especially those with identity conflict.…”

Heuristics as conceptual lens for understanding and studying the usage of bibliometrics in research evaluation

“…Those at the other extremity are called large (or uncertain) worlds (see Binmore 2007Binmore , 2009Savage 1954) in the literature on fast-andfrugal heuristics. 7 Following Gigerenzer (e.g., 2014;Hafenbrädl et al 2016;Mousavi and Gigerenzer 2017), in large, uncertain worlds, decision makers may not recognize and know all their options, with some options resting forever 'in the dark'. Decision makers may also not be aware of all potential consequences of the options at hand.…”

“…For example, we have used those very notions to question the meaningfulness of research evaluation exercises, as well as to discuss research evaluation in terms of larger historical and societal trends (Marewski and Bornmann 2019). 8 Indeed, there is a large literature (and a good amount of debate) about the scope of classic rational models and their (e.g., statistical) relatives (see e.g., Mousavi and Gigerenzer 2017;Shanks and Lagnado 2000;Todd and Gigerenzer 2000). For instance, Gigerenzer and Marewski (2015) point out that some authors (e.g., Lindley 1983) do believe that Bayesian statistics can be applied to all kinds of uncertainty, including single (e.g., isolated) occurrences: to paraphrase the line of reasoning, after all, one can specify subjec-The conceptual lens of the fast-and-frugal heuristics framework suggests that most real-world decision tasks come, in essence, as situations of uncertainty: large worlds that may (or may not) entail elements of risk, but that even if they include some known risks, still remain fundamentally uncertain (Gigerenzer 2014;Hafenbrädl et al 2016;Gigerenzer 2014, 2017).…”

“…Moreover, different heuristics and more complex strategies need to be pitted against each other in model comparisons in different environments, and not be tested in solitariness (e.g., Gigerenzer and Brighton 2009;Gigerenzer and Gaissmaier 2011;Marewski and Mehlhorn 2011). An example of isolated 'tests' is the pitting of a model of a heuristic against a verbally-specified null hypothesis rather than against other formal models of decision processes (e.g., Mousavi and Gigerenzer 2017). 14…”

“…They frequently operate in environments with criteria (outcomes) that are difficult to predict (e.g., decision makers in science who try to select the next breakthrough research projects). In these uncertain environments, "less information can be more" (e.g., Marewski et al 2010a, p. 111) when it comes to making inferences-that is, judgments about something that is unknown, be it because the criterion to be inferred lies in the future or because it is simply not accessible at present (see e.g., Gigerenzer and Brighton 2009;Gigerenzer and Gaissmaier 2011;Goldstein and Gigerenzer 2009;Hafenbrädl et al 2016;Marewski et al 2010a;Mousavi and Gigerenzer 2017).…”