Abstract
Heuristics are simple strategies of judgment and problem solving that reach a decision while ignoring part of the information that is available. Herbert Simon's account of bounded rationality supplied the foundation, and two research programmes later developed the idea in opposing directions. The heuristics-and-biases programme of Amos Tversky and Daniel Kahneman treated heuristics as economical and usually effective but as a source of systematic, predictable error, whereas the ecological-rationality programme of Gerd Gigerenzer and colleagues asked when simple strategies match or outperform complex ones. Several classic claims have since been revised, among them the hot hand fallacy, and the dual-process architecture often used to explain heuristics remains contested. This article reviews both programmes, the debate between them, the conditions under which intuition can be trusted, and the current research front.
Keywords: heuristics, bounded rationality, heuristics and biases, ecological rationality, decision making
A heuristic is what ordinary language calls a rule of thumb: a simple procedure that reaches a judgment by setting aside part of the available information instead of weighing all of it (Gigerenzer & Gaissmaier, 2011). One influential framework holds that what heuristics share is the reduction of effort, and that a small number of effort-reduction principles accounts for the many heuristics described in the literature (Shah & Oppenheimer, 2008).
Two traditions have shaped the field. The first, begun by Tversky and Kahneman, identified heuristics largely through the systematic errors they produce (Tversky & Kahneman, 1974). The second, associated with Gigerenzer and colleagues, studies the environments in which simple heuristics are adaptive and the conditions under which using less information yields better inferences (Gigerenzer & Gaissmaier, 2011). The sections below trace the idea from bounded rationality through both programmes, set out the debate between them and its relation to dual-process theories, and close with the conditions for trustworthy intuition and the open questions that remain.
Key Takeaways
- One framework reads every heuristic, whatever its form, as a strategy for cutting down the effort a judgement takes.
- Tversky and Kahneman described three heuristics, representativeness, availability and adjustment from an anchor, as economical and usually effective yet liable to systematic, predictable errors; the standing of that programme is contested.
- In computer simulations, Take The Best matched or beat more complex rivals, an existence proof rather than human data; less-is-more advantages can arise, for instance, when predictability is low and samples are small.
- The hot hand fallacy has been revised: the original analyses carried a streak selection bias, and correcting for it reversed the conclusion in a reanalysis.
- The validity of skilled intuition depends on a predictable environment and an opportunity to learn, and subjective confidence is an unreliable guide to accuracy.
What Heuristics Are
MeSH classifies heuristics under problem solving, and the concept sits at the junction of decision making and judgment. In the field's own terms, a heuristic is a strategy that deliberately leaves some information out of account. Gigerenzer and Gaissmaier's review separated two questions about such strategies: a descriptive one, about which heuristics people actually use, and a prescriptive one, about when a given heuristic ought to be used; it also argued for stating heuristics as formal models of inference (Gigerenzer & Gaissmaier, 2011).
A complementary definition comes from a theoretical analysis that framed every heuristic as a means of cutting effort and reduced the catalogue to a handful of effort-reduction principles (Shah & Oppenheimer, 2008). The two definitions are compatible. Both make ignoring information the defining feature; they differ in whether that omission is described mainly as a saving of effort or as a strategy whose value depends on the environment. Figure 1 sketches how the two research programmes read the same shortcut.
Figure 1
Two Views of the Same Shortcut
Bounded Rationality and Satisficing
The modern study of heuristics begins with bounded rationality. Its starting point was Simon's 1955 behavioral model of rational choice (Simon, 1955). A year later he argued that the standard models of economic rationality postulate a much larger capacity for computation in the organism than do psychology's models of adaptive behavior, and he related choice to adaptive behavior and to learning theories; he also conceded that the data for comparing the models with actual behavior were scanty (Simon, 1956).
Simon's alternative to optimization was satisficing, a mode of choice organised around an aspiration level. Some 60 years on, a cross-disciplinary literature survey traced how satisficing had developed along two separate traditions since 1955 and brought them together (Artinger et al., 2022). The same survey drew a sharp line between risk and uncertainty: under risk, satisficing was typically inferior to optimization, whereas under uncertainty it could be highly effective, and the authors set out the conditions under which it can be rational (Artinger et al., 2022). Bounded rationality is therefore best described as supported, with the rationality of satisficing itself limited to those conditions (Artinger et al., 2022).
The Heuristics-and-Biases Programme
In the early 1970s Kahneman and Tversky turned the idea of limited computation into an account of everyday probability judgment. Their 1972 paper treated subjective probability as a judgment of representativeness (Kahneman & Tversky, 1972). In a subsequent series of experiments with both naive and sophisticated subjects, category and numerical predictions followed how representative an outcome seemed, while prior probabilities were largely disregarded (Kahneman & Tversky, 1973). The representativeness heuristic had a companion in the availability heuristic, which the same authors presented as a heuristic for judging frequency and probability (Tversky & Kahneman, 1973).
The 1974 synthesis in Science named three heuristics for judgment under uncertainty: representativeness, availability, and adjustment from an anchor (Tversky & Kahneman, 1974). Its central claim had two halves. The heuristics were economical and usually effective, yet they led to errors that were systematic and predictable, which the authors called biases (Tversky & Kahneman, 1974). The programme's standing is now contested: a later critical review argued that the field had developed a bias bias, detecting biases that were often absent (Gigerenzer, 2018).
The best known demonstration of the programme concerned the conjunction fallacy, a violation of the conjunction rule of probability. Across judgment experiments in several contexts, both lay people and experts committed it, and the authors traced the error to the representativeness and availability heuristics (Tversky & Kahneman, 1983). The same paper discussed alternative interpretations and explored attempts at debiasing (Tversky & Kahneman, 1983).
Each of the three original heuristics has since been revised. For availability, three between-subjects experiments with German university and college students asked participants to recall either 6 or 12 examples of their own assertive behavior; the experienced ease of recall moderated how much the recalled content affected their assertiveness ratings (Schwarz et al., 1991). The evidence has clear limits: the key interaction in Experiment 1 was only marginal, although it was significant in Experiment 2, the diagnosticity manipulation in Experiment 2 did not reach significance, and assertiveness was measured only by self-assessment in student samples (Schwarz et al., 1991). The effort-reduction framework offers a second revision, re-examining existing heuristics in terms of effort reduction (Shah & Oppenheimer, 2008). A 2012 model comparison across two studies with student samples pitted availability against the affect heuristic in risk judgment: availability-by-recall gave the best fit overall, although the weight of affect relative to availability was greater when people judged the value of a statistical life or rated perceived risk than when they estimated frequencies, and there was some evidence that the two mechanisms combine (Pachur et al., 2012). Availability is accordingly classed as revised (Pachur et al., 2012).
Anchoring and adjustment is likewise revised. Two sets of experiments, Studies 1a and 1b and Studies 2a to 2c, examined anchors that people generate themselves; adjustment from such anchors tended to halt at the first value that seemed plausible, and more effortful search reduced the shortfall (Epley & Gilovich, 2006). The authors noted that the causes of insufficient adjustment had previously been unclear (Epley & Gilovich, 2006). A rival account, tested in three experiments, treats the anchoring effect as a case of semantic priming, which its authors called selective accessibility (Strack & Mussweiler, 1997). The same studies also showed when the effect changes: a contrast effect appeared when the primed information did not fit the absolute judgment and the two targets differed enough, and an implausible anchor let the comparison be made without drawing on knowledge about the target (Strack & Mussweiler, 1997). A later computational model, framed in terms of bounded rationality, recast anchoring as resource-rational adjustment, the rational use of limited resources, and showed that this one account could bring together ten anchoring phenomena (Lieder et al., 2018). The broader anchoring literature has been the subject of its own review (Furnham & Boo, 2011).
Representativeness, finally, was revised into a more general account. Kahneman and Frederick revisited it as a case of attribute substitution in intuitive judgment (Kahneman & Frederick, 2002). Kahneman's later theoretical review explained intuitive judgment, framing effects and attribute substitution through a single notion of accessibility, linking the programme to prospect theory; it added that accessible rules allow intuitive judgments to be corrected on occasion (Kahneman, 2003). Representativeness is therefore classed as revised. The heuristic has also been turned into a formal model of stereotypes, built on the representative types of a group relative to a reference group and examined with laboratory and field data; on this model stereotypes keep a kernel of truth, distort most when the groups are similar, and depend on context (Bordalo et al., 2016).
A related proposal, the affect heuristic, is the subject of a paper of that name by Slovic and colleagues (Slovic et al., 2007); in the 2012 comparison above it ranked behind availability-by-recall overall, so its standing is contested (Pachur et al., 2012).
Neglect of base rates, one of the 1973 findings, showed that predictions tracked representativeness while prior probability was set aside (Kahneman & Tversky, 1973). An experimental analysis of several thousand problem solutions from naive participants found that frequency formats improved Bayesian inference relative to probability formats, although even then only up to 50% of solutions were Bayesian and participants also used non-Bayesian algorithms (Gigerenzer & Hoffrage, 1995). The demonstration below works through a base-rate problem in frequency form.
Count, Do Not Guess
Base Rates in Natural Frequencies
Imagine 1,000 people tested for a condition. Move the slider to change how common the condition is, and see how many of the positive results are real.
Fast-and-Frugal Heuristics
The ecological-rationality programme reversed the question. Instead of asking where heuristics fail, it asked where simple strategies succeed. In a competition among algorithms run by computer simulation, Take The Best, a form of one-reason decision making, matched or outperformed more demanding rivals, including multiple regression (Gigerenzer & Goldstein, 1996). The authors set such fast and frugal algorithms against the ideal of a Laplacean Demon. The result is an existence proof only: it came from simulation, not from human data (Gigerenzer & Goldstein, 1996).
The recognition heuristic extended the approach to cases of missing knowledge. Goldstein and Gigerenzer presented it as an adaptive and frugal inference strategy, specified the conditions under which it succeeds, and linked it to a less-is-more effect (Goldstein & Gigerenzer, 2002). A later narrative review of theory and tests concluded that recognition is adaptive and holds a special status among cues, but it also mapped boundary conditions on its use, listed open questions, recorded a continuing debate over the underlying processes, and discussed the heuristic's relation to availability and fluency (Pachur et al., 2011). The recognition heuristic's standing is contested (Pachur et al., 2011).
More generally, the programme claims less-is-more effects, in which a simpler strategy outperforms a more complex one. In Gigerenzer and Gaissmaier's review, setting aside some of the available information sometimes made judgments more accurate, for instance in settings of low predictability and with small samples, although the authors conceded that the field still had no systematic theory of heuristics (Gigerenzer & Gaissmaier, 2011). On that basis the fast-and-frugal account is best described as contested (Gigerenzer & Gaissmaier, 2011). Critics have disputed the psychological plausibility of both probabilistic mental models and the fast-and-frugal heuristics, setting their assumptions against existing data (Dougherty et al., 2008). Earlier tests of the recognition heuristic had confounded recognition with other cues, and when two studies controlled for those confounds, they found no support for its use (Oppenheimer, 2003). Its applied arm has argued, in a review, that fast-and-frugal heuristics can serve as decision aids and improve real decisions made under uncertainty (Hafenbrädl et al., 2016).
A recent theory paper widened the scope to embodied heuristics, taking the gaze heuristic as its case and placing such strategies within an adaptive toolbox (Gigerenzer, 2021). The paper is programmatic rather than empirical, and it notes that no modelling programmes for embodied heuristics yet exist (Gigerenzer, 2021). The demonstration below runs Take The Best on a small set of cues.
One Good Reason
Take The Best Versus Tallying
Which option scores higher on some criterion? Toggle the cues. Take The Best stops at the first cue that tells the options apart; tallying counts every cue.
| Cue | Option A | Option B |
|---|---|---|
| Cue 1 (most valid) | ||
| Cue 2 | ||
| Cue 3 | ||
| Cue 4 (least valid) |
The Rationality Debate
The two programmes disagree less about whether people use heuristics than about how to read the errors. The conjunction fallacy is the standard case. Tversky and Kahneman attributed it to representativeness and availability in lay and expert judges alike (Tversky & Kahneman, 1983). In a series of studies built on the Linda problem, Hertwig and Gigerenzer argued instead that the word probability is polysemous, that participants draw semantic and pragmatic inferences about its meaning, and that content-blind norms therefore partly misread such inferences as violations of the conjunction rule; the studies also examined frequency judgments and extensional cues (Hertwig & Gigerenzer, 1999). Kahneman and Tversky answered Gigerenzer's critique directly, arguing that judgments of frequency are biased too, not only subjective probabilities, and so defending the heuristics-and-biases account of cognitive illusions (Kahneman & Tversky, 1996).
The wider critique is the bias bias. A critical review of behavioral economics argued that biases are often spotted even when there are none, that the evidence that they impose real costs is weak, and that some alleged biases rest on confusing the statistics of small samples with those of large ones (Gigerenzer, 2018). Nudges and rational choice theory were among the topics it addressed.
The hot hand shows how a classic result can move. The 1985 study of basketball shooting treated belief in a hot hand as a misperception of random sequences (Gilovich et al., 1985). Miller and Sanjurjo combined a theoretical proof with a reanalysis and identified a streak selection bias in the way such sequences had been analysed; once it was corrected, the evidence pointed the other way (Miller & Sanjurjo, 2018). The size of the bias depends on the data, shrinking as sequences grow longer and growing as the streak of interest lengthens (Miller & Sanjurjo, 2018). The hot hand fallacy is therefore classed as revised.
Replicability bears on both sides of the debate. A multi-site replication project tested 13 effects in 36 independent samples totalling 6,344 participants, and 10 of the effects replicated, with the outcome depending on the effect (Klein et al., 2014). The project covered a small sample of effects and found little difference between laboratory and online data collection (Klein et al., 2014). The demonstration below computes the streak selection bias exactly for short sequences of coin flips.
A Bias in the Measure
Streak Selection Bias in Short Sequences
Every flip is a fair coin, so a head follows a head half the time. Yet averaging, sequence by sequence, the share of heads after a head gives less than one half.
Heuristics and Dual-Process Theories
Heuristic judgment is often assigned to the fast system in a dual process theory. Kruglanski and Gigerenzer objected to that assignment, arguing that intuitive and deliberate judgments draw on common rules; they named rule selection as the important open question and proposed a two-step process for it, and they held that deliberation is not generally more accurate than intuition (Kruglanski & Gigerenzer, 2011). Evans and Stanovich answered five main lines of argument against such theories, distinguishing Type 1 processes, which are autonomous and supply default responses, from Type 2 processes, which engage working memory and support hypothetical thinking (Evans & Stanovich, 2013). They accepted that some criticisms were valid against some theories, judged the criticisms overstated overall, and argued that critics had wrongly assumed a generic version that applies to all (Evans & Stanovich, 2013).
A more recent theoretical critique and evidence review looked back on roughly 50 years of fast-and-slow models and concluded that two of their core assumptions are poorly founded: the exclusivity feature, which assigns a response to one type of processing or the other, and the switch feature, which governs the hand-off from intuitive to deliberate processing (De Neys, 2023). The alternative architecture it proposed was only sketched (De Neys, 2023). Dual-process theories are therefore contested (De Neys, 2023).
When Intuition Can Be Trusted
The two traditions converged on a practical question about intuition. In a joint theory paper, Kahneman, from heuristics and biases, and Klein, from naturalistic decision making, compared their views of professional intuition and agreed that the validity of intuitive expertise depends on a predictable environment and on an opportunity to learn its regularities; they also agreed that subjective confidence is an unreliable indicator of accuracy (Kahneman & Klein, 2009). Earlier, Kahneman had argued that documenting the biases of intuition is consistent with seeing intuitive thinking as, on the whole, skilled and successful (Kahneman, 2003). Table 1 summarises the standing of each account discussed on this page.
Table 1
Major Accounts of Heuristics and Their Standing
| Theory | Core claim | Status | Key source |
|---|---|---|---|
| Bounded rationality | Models of economic rationality postulate more computation than psychology's models of adaptive behavior; satisficing is rational only under certain conditions | Supported (Artinger et al., 2022) | Simon, 1956 |
| Heuristics-and-biases program | Three heuristics are economical and usually effective but produce systematic, predictable errors | Contested (Gigerenzer, 2018) | Tversky & Kahneman, 1974 |
| Fast-and-frugal heuristics | Simple heuristics can be adaptive; less-is-more effects found, for instance, under low predictability and small samples | Contested (Gigerenzer & Gaissmaier, 2011) | Gigerenzer & Goldstein, 1996 |
| Recognition heuristic | Recognition is an adaptive, frugal cue whose success requires specified conditions | Contested (Pachur et al., 2011) | Goldstein & Gigerenzer, 2002 |
| Availability heuristic | Frequency and probability judged by availability; ease of retrieval moderates the impact of recalled content | Revised (Schwarz et al., 1991) | Tversky & Kahneman, 1973 |
| Anchoring and adjustment | Adjustment from self-generated anchors tends to stop once a plausible value is reached | Revised (Epley & Gilovich, 2006) | Epley & Gilovich, 2006 |
| Representativeness heuristic | Predictions follow representativeness and neglect prior probability; later recast through attribute substitution | Revised (Kahneman, 2003) | Kahneman & Tversky, 1973 |
| Hot hand fallacy | Belief in streaks misperceives random sequences; a streak selection bias reverses the evidence | Revised (Miller & Sanjurjo, 2018) | Gilovich et al., 1985 |
| Dual-process theories | Autonomous Type 1 and working-memory-based Type 2 processes; exclusivity and switch assumptions disputed | Contested (De Neys, 2023) | Evans & Stanovich, 2013 |
| Resource-rational analysis | A proposed unifying framework combining cognitive constraints with optimality principles | Proposed framework (Lieder & Griffiths, 2020) | Lieder & Griffiths, 2020 |
| Affect heuristic | Affect informs risk judgments; its weight relative to availability was greater when judging the value of a statistical life or perceived risk than when judging frequency, but it fit less well overall than availability-by-recall (student samples) | Contested (Pachur et al., 2012) | Slovic et al., 2007 |
Note. The Status column shows how each account stands today, and its citation points to the work from which that judgment is drawn.
Worked Example
The two calculations below are illustrative arithmetic and come from no study.
Base rates. Suppose a condition has a base rate of 1% in a group of 1,000 people, so that 10 people have it. A test detects the condition 80% of the time, which yields 8 true positives among those 10. The test also gives a false alarm 10% of the time among the 990 people without the condition, which yields 99 false positives. Of the 107 people who test positive, only 8 have the condition, so the probability that a positive result is correct is 8 / 107, or about 7.5%. The BaseRateDemo reproduces this calculation in frequency form.
Streaks. Consider every possible sequence of 4 flips of a fair coin; there are 16. In 14 of them a head appears somewhere before the final flip, so the share of heads that immediately follow a head can be computed for each of those sequences. Averaging that share across the 14 sequences, each sequence counted equally, gives 17/42, or about 0.405, rather than 0.5, even though every flip is independent. Selecting the flips that follow a streak within a short sequence thus produces a downward bias of the kind identified in the hot hand literature (Miller & Sanjurjo, 2018). The HotHandDemo computes this value exactly.
Discussion
The two programmes agree on more than their rhetoric suggests. Both treat heuristics as often useful: Tversky and Kahneman called them economical and usually effective (Tversky & Kahneman, 1974), and Gigerenzer and Gaissmaier described simple heuristics as adaptive (Gigerenzer & Gaissmaier, 2011). They part company over what to do with the failures. The heuristics-and-biases programme uses systematic error as its window onto the underlying judgment, whereas the ecological-rationality programme asks how well a strategy fits its environment and treats some apparent errors as products of content-blind norms (Hertwig & Gigerenzer, 1999).
Much has moved since 1974. Availability is now discussed in terms of ease of retrieval and effort reduction, anchoring in terms of where adjustment stops, representativeness in terms of attribute substitution, and the hot hand in terms of a selection bias in the original analyses; each of these accounts is accordingly classed as revised in Table 1. Several questions remain open. A systematic theory of heuristics is still lacking (Gigerenzer & Gaissmaier, 2011), the processes behind the recognition heuristic remain disputed (Pachur et al., 2011), and the dual-process architecture into which heuristics are usually placed has itself been questioned (De Neys, 2023). Settling these questions would require models specific enough to predict, in advance, which strategy a person will use in a given environment and how accurate it will be.
Current Directions
Recent work has pursued integration as much as debate. The two traditions of satisficing research have been drawn together in a single survey that specifies when satisficing is and is not rational (Artinger et al., 2022). Resource-rational analysis proposes a unifying modelling framework that treats cognition as the optimal use of limited computational resources, combining cognitive constraints with optimality principles and cognitive architectures (Lieder & Griffiths, 2020). It is a theoretical proposal whose authors describe it as a promising framework (Lieder & Griffiths, 2020).
Other lines continue the older disputes. The architecture of fast and slow thinking has been challenged at its foundations (De Neys, 2023), the bias bias critique has questioned how often biases are real and costly (Gigerenzer, 2018), and the reanalysis of the hot hand has shown how a selection bias in the analysis could sustain a textbook fallacy (Miller & Sanjurjo, 2018). The ecological-rationality programme has meanwhile turned outward, toward embodied heuristics in a programmatic proposal (Gigerenzer, 2021) and toward fast-and-frugal tools for applied decisions (Hafenbrädl et al., 2016).
Common Misconceptions
- Heuristics are just errors in thinking.
- Even the programme best known for documenting biases described heuristics as economical and usually effective (Tversky & Kahneman, 1974). The rival programme treats simple heuristics as adaptive strategies (Gigerenzer & Gaissmaier, 2011).
- More information always leads to better decisions.
- Less-is-more effects show that a simpler strategy can do better, for instance when predictability is low and samples are small (Gigerenzer & Gaissmaier, 2011).
- The hot hand is a settled fallacy.
- A streak selection bias in the standard analysis of short sequences was later identified, and a reanalysis that corrected for it reversed the original conclusion (Miller & Sanjurjo, 2018).
Glossary
- Affect heuristic.
- Use of affect to inform risk judgments, the subject of a 2007 paper of that name by Slovic and colleagues; contested, since in a 2012 comparison it fit risk judgments less well overall than availability-by-recall.
- Anchoring and adjustment.
- A judgment strategy that starts from an anchor value and adjusts from it; for self-generated anchors, adjustment tends to stop at the first plausible value.
- Attribute substitution.
- The account by which Kahneman and Frederick reinterpreted representativeness in intuitive judgment, later explained by Kahneman through accessibility.
- Availability heuristic.
- A heuristic for judging frequency and probability; later work with student samples showed that the experienced ease of recall moderates the impact of what is recalled.
- Base-rate neglect.
- Setting aside prior probability in intuitive prediction in favour of how representative an outcome appears.
- Bounded rationality.
- Simon's view that models of rational choice must respect the limited computation available to real organisms.
- Conjunction fallacy.
- A violation of the conjunction rule of probability in intuitive judgment, shown by lay people and experts; its reading as an error is disputed.
- Ease of retrieval.
- The subjective experience of how easily examples come to mind, treated as information in its own right in a revision of the availability heuristic.
- Ecological rationality.
- The programme that evaluates a heuristic by how well it fits the environment in which it is used.
- Hot hand fallacy.
- The claim that belief in shooting streaks misperceives random sequences; revised after a streak selection bias was found in the original analyses.
- Less-is-more effect.
- Better performance from a simpler strategy, found for instance under low predictability and with small samples.
- Recognition heuristic.
- An adaptive, frugal inference strategy that exploits missing knowledge; its success requires specified conditions, and its underlying processes are debated.
- Representativeness heuristic.
- Judging probability or making a prediction by how representative an outcome seems, with prior probability neglected.
- Satisficing.
- Simon's alternative to optimization, organised around an aspiration level; judged rational only under certain conditions and usually inferior under risk.
- Streak selection bias.
- A bias in analyses of what follows a streak within finite sequences; it shrinks with longer sequences and grows with longer streaks.
- Take The Best.
- A one-reason decision heuristic that matched or beat more complex algorithms in computer simulation, an existence proof rather than human data.
Key Researchers
Gerd Gigerenzer
(b. 1947). Director of the Harding Center for Risk Literacy at the University of Potsdam and director emeritus at the Max Planck Institute for Human Development; with Daniel Goldstein he developed fast-and-frugal heuristics such as Take The Best. ORCID - Wikidata - Wikipedia
Thomas Gilovich
(b. 1954). Irene Blecker Rosenfeld Professor of Psychology at Cornell University; coauthor of the 1985 hot hand study and of work on insufficient adjustment from self-generated anchors. Wikipedia
Daniel G. Goldstein
(b. 1969). Cognitive psychologist at Microsoft Research New York City; with Gerd Gigerenzer he formulated the recognition heuristic. Wikidata - Wikipedia - Faculty Page
Ralph Hertwig
(b. 1963). Director of the Center for Adaptive Rationality at the Max Planck Institute for Human Development, Berlin; coauthor of a reinterpretation of the conjunction fallacy. ORCID - Wikidata - Wikipedia - Faculty Page
Daniel Kahneman
(1934–2024). Psychologist at Princeton University who, with Amos Tversky, founded the heuristics-and-biases programme and later wrote on attribute substitution and the conditions for intuitive expertise. Wikidata - Wikipedia
Herbert A. Simon
(1916–2001). Scholar at Carnegie Mellon University whose behavioral model of rational choice introduced bounded rationality and satisficing. Wikidata - Wikipedia
Paul Slovic
(b. 1938). Professor Emeritus of Psychology at the University of Oregon, affiliated with Decision Research; coauthor of work on the affect heuristic. Wikidata - Wikipedia
Amos Tversky
(1937–1996). Psychologist at Stanford University, earlier at the Hebrew University of Jerusalem, who with Daniel Kahneman described the availability, representativeness and anchoring heuristics and the conjunction fallacy. Wikidata - Wikipedia
Frequently Asked Questions
What is a heuristic?
A heuristic is a simple strategy, often called a rule of thumb, that reaches a judgment or decision while ignoring part of the available information (Gigerenzer & Gaissmaier, 2011).
Who introduced the idea of bounded rationality?
Herbert Simon developed it; in 1956 he argued that the models of rational behavior used in economics postulate a much larger capacity for computation in the organism than the models of adaptive behavior used in psychology, while noting that the comparison data were scanty (Simon, 1956).
What are the three classic heuristics?
Tversky and Kahneman named representativeness, availability, and adjustment from an anchor as the three heuristics used in judgment under uncertainty (Tversky & Kahneman, 1974).
Are heuristics bad for decision making?
Not in general. Tversky and Kahneman described heuristics as economical and usually effective, even though they can lead to systematic and predictable errors (Tversky & Kahneman, 1974).
What is Take The Best?
Take The Best is a one-reason decision heuristic that matched or outperformed more complex algorithms, including multiple regression, in computer simulations; the result is an existence proof and not evidence from human participants (Gigerenzer & Goldstein, 1996).
Is the hot hand a fallacy?
The question has been reopened. A streak selection bias was identified in the way short sequences had been analysed, and correcting for it reversed the earlier conclusion in a reanalysis (Miller & Sanjurjo, 2018).
How do heuristics relate to System 1 and System 2?
Heuristics are often assigned to the fast, intuitive side of dual-process models, but a recent critique argued that the exclusivity and switch assumptions of those models are poorly founded (De Neys, 2023).
When can intuition be trusted?
Kahneman and Klein agreed that the validity of intuitive expertise depends on a predictable environment and an opportunity to learn it, and that subjective confidence is not a reliable sign of accuracy (Kahneman & Klein, 2009).
References
Artinger, F. M., Gigerenzer, G., & Jacobs, P. (2022). Satisficing: Integrating two traditions. Journal of Economic Literature, 60(2), 598-635. https://doi.org/10.1257/jel.20201396
Bordalo, P., Coffman, K., Gennaioli, N., & Shleifer, A. (2016). Stereotypes. Quarterly Journal of Economics, 131(4), 1753-1794. https://doi.org/10.1093/qje/qjw029
De Neys, W. (2023). Advancing theorizing about fast-and-slow thinking. Behavioral and Brain Sciences, 46, e111. https://doi.org/10.1017/S0140525X2200142X
Dougherty, M. R., Franco-Watkins, A. M., & Thomas, R. (2008). Psychological plausibility of the theory of probabilistic mental models and the fast and frugal heuristics. Psychological Review, 115(1), 199-213. https://doi.org/10.1037/0033-295x.115.1.199
Epley, N., & Gilovich, T. (2006). The anchoring-and-adjustment heuristic: Why the adjustments are insufficient. Psychological Science, 17(4), 311-318. https://doi.org/10.1111/j.1467-9280.2006.01704.x
Evans, J. St. B. T., & Stanovich, K. E. (2013). Dual-process theories of higher cognition: Advancing the debate. Perspectives on Psychological Science, 8(3), 223-241. https://doi.org/10.1177/1745691612460685
Furnham, A., & Boo, H. C. (2011). A literature review of the anchoring effect. The Journal of Socio-Economics, 40(1), 35-42. https://doi.org/10.1016/j.socec.2010.10.008
Gigerenzer, G. (2018). The bias bias in behavioral economics. Review of Behavioral Economics, 5(3-4), 303-336. https://doi.org/10.1561/105.00000092
Gigerenzer, G. (2021). Embodied heuristics. Frontiers in Psychology, 12, 711289. https://doi.org/10.3389/fpsyg.2021.711289
Gigerenzer, G., & Gaissmaier, W. (2011). Heuristic decision making. Annual Review of Psychology, 62, 451-482. https://doi.org/10.1146/annurev-psych-120709-145346
Gigerenzer, G., & Goldstein, D. G. (1996). Reasoning the fast and frugal way: Models of bounded rationality. Psychological Review, 103(4), 650-669. https://doi.org/10.1037/0033-295X.103.4.650
Gigerenzer, G., & Hoffrage, U. (1995). How to improve Bayesian reasoning without instruction: Frequency formats. Psychological Review, 102(4), 684-704. https://doi.org/10.1037/0033-295X.102.4.684
Gilovich, T., Vallone, R., & Tversky, A. (1985). The hot hand in basketball: On the misperception of random sequences. Cognitive Psychology, 17(3), 295-314. https://doi.org/10.1016/0010-0285(85)90010-6
Goldstein, D. G., & Gigerenzer, G. (2002). Models of ecological rationality: The recognition heuristic. Psychological Review, 109(1), 75-90. https://doi.org/10.1037/0033-295X.109.1.75
Hafenbrädl, S., Waeger, D., Marewski, J. N., & Gigerenzer, G. (2016). Applied decision making with fast-and-frugal heuristics. Journal of Applied Research in Memory and Cognition, 5(2), 215-231. https://doi.org/10.1016/j.jarmac.2016.04.011
Hertwig, R., & Gigerenzer, G. (1999). The 'conjunction fallacy' revisited: How intelligent inferences look like reasoning errors. Journal of Behavioral Decision Making, 12(4), 275-305. https://doi.org/10.1002/(sici)1099-0771(199912)12:4%3C275::aid-bdm323%3E3.0.co;2-m
Kahneman, D. (2003). A perspective on judgment and choice: Mapping bounded rationality. American Psychologist, 58(9), 697-720. https://doi.org/10.1037/0003-066X.58.9.697
Kahneman, D., & Frederick, S. (2002). Representativeness revisited: Attribute substitution in intuitive judgment. In T. Gilovich, D. Griffin, & D. Kahneman (Eds.), Heuristics and biases: The psychology of intuitive judgment (pp. 49-81). Cambridge University Press. https://doi.org/10.1017/CBO9780511808098.004
Kahneman, D., & Klein, G. (2009). Conditions for intuitive expertise: A failure to disagree. American Psychologist, 64(6), 515-526. https://doi.org/10.1037/a0016755
Kahneman, D., & Tversky, A. (1972). Subjective probability: A judgment of representativeness. Cognitive Psychology, 3(3), 430-454. https://doi.org/10.1016/0010-0285(72)90016-3
Kahneman, D., & Tversky, A. (1973). On the psychology of prediction. Psychological Review, 80(4), 237-251. https://doi.org/10.1037/h0034747
Kahneman, D., & Tversky, A. (1996). On the reality of cognitive illusions. Psychological Review, 103(3), 582-591. https://doi.org/10.1037/0033-295x.103.3.582
Klein, R. A., Ratliff, K. A., Vianello, M., Adams, R. B., Jr., Bahník, Š., Bernstein, M. J., et al. (2014). Investigating variation in replicability: A "many labs" replication project. Social Psychology, 45(3), 142-152. https://doi.org/10.1027/1864-9335/a000178
Kruglanski, A. W., & Gigerenzer, G. (2011). Intuitive and deliberate judgments are based on common principles. Psychological Review, 118(1), 97-109. https://doi.org/10.1037/a0020762
Lieder, F., & Griffiths, T. L. (2020). Resource-rational analysis: Understanding human cognition as the optimal use of limited computational resources. Behavioral and Brain Sciences, 43, e1. https://doi.org/10.1017/S0140525X1900061X
Lieder, F., Griffiths, T. L., Huys, Q. J. M., & Goodman, N. D. (2018). The anchoring bias reflects rational use of cognitive resources. Psychonomic Bulletin & Review, 25(1), 322-349. https://doi.org/10.3758/s13423-017-1286-8
Miller, J. B., & Sanjurjo, A. (2018). Surprised by the hot hand fallacy? A truth in the law of small numbers. Econometrica, 86(6), 2019-2047. https://doi.org/10.3982/ECTA14943
Oppenheimer, D. M. (2003). Not so fast! (and not so frugal!): Rethinking the recognition heuristic. Cognition, 90(1), B1-B9. https://doi.org/10.1016/s0010-0277(03)00141-0
Pachur, T., Hertwig, R., & Steinmann, F. (2012). How do people judge risks: Availability heuristic, affect heuristic, or both? Journal of Experimental Psychology: Applied, 18(3), 314-330. https://doi.org/10.1037/a0028279
Pachur, T., Todd, P. M., Gigerenzer, G., Schooler, L. J., & Goldstein, D. G. (2011). The recognition heuristic: A review of theory and tests. Frontiers in Psychology, 2, 147. https://doi.org/10.3389/fpsyg.2011.00147
Schwarz, N., Bless, H., Strack, F., Klumpp, G., Rittenauer-Schatka, H., & Simons, A. (1991). Ease of retrieval as information: Another look at the availability heuristic. Journal of Personality and Social Psychology, 61(2), 195-202. https://doi.org/10.1037/0022-3514.61.2.195
Shah, A. K., & Oppenheimer, D. M. (2008). Heuristics made easy: An effort-reduction framework. Psychological Bulletin, 134(2), 207-222. https://doi.org/10.1037/0033-2909.134.2.207
Simon, H. A. (1955). A behavioral model of rational choice. The Quarterly Journal of Economics, 69(1), 99-118. https://doi.org/10.2307/1884852
Simon, H. A. (1956). Rational choice and the structure of the environment. Psychological Review, 63(2), 129-138. https://doi.org/10.1037/h0042769
Slovic, P., Finucane, M. L., Peters, E., & MacGregor, D. G. (2007). The affect heuristic. European Journal of Operational Research, 177(3), 1333-1352. https://doi.org/10.1016/j.ejor.2005.04.006
Strack, F., & Mussweiler, T. (1997). Explaining the enigmatic anchoring effect: Mechanisms of selective accessibility. Journal of Personality and Social Psychology, 73(3), 437-446. https://doi.org/10.1037/0022-3514.73.3.437
Tversky, A., & Kahneman, D. (1973). Availability: A heuristic for judging frequency and probability. Cognitive Psychology, 5(2), 207-232. https://doi.org/10.1016/0010-0285(73)90033-9
Tversky, A., & Kahneman, D. (1974). Judgment under uncertainty: Heuristics and biases. Science, 185(4157), 1124-1131. https://doi.org/10.1126/science.185.4157.1124
Tversky, A., & Kahneman, D. (1983). Extensional versus intuitive reasoning: The conjunction fallacy in probability judgment. Psychological Review, 90(4), 293-315. https://doi.org/10.1037/0033-295X.90.4.293