## The Role of Causality in Judgment Under Uncertainty

Citations: | 12 - 0 self |

### BibTeX

@MISC{Krynski_therole,

author = {Tevye R. Krynski and Joshua B. Tenenbaum},

title = { The Role of Causality in Judgment Under Uncertainty},

year = {}

}

### OpenURL

### Abstract

Leading accounts of judgment under uncertainty evaluate performance within purely statistical frameworks, holding people to the standards of classical Bayesian (Tversky & Kahneman, 1974) or frequentist (Gigerenzer & Hoffrage, 1995) norms. We argue that these frameworks have limited ability to explain the success and flexibility of people's real-world judgments, and propose an alternative normative framework based on Bayesian inferences over causal models. Deviations from traditional norms of judgment, such as "base-rate neglect", may then be explained in terms of a mismatch between the statistics given to people and the causal models they intuitively construct to support probabilistic reasoning. Four experiments show that when a clear mapping can be established from given statistics to the parameters of an intuitive causal model, people are more likely to use the statistics appropriately, and that when the classical and causal Bayesian norms differ in their prescriptions, people's judgments are more consistent with causal Bayesian norms.

### Citations

7489 |
Probabilistic Reasoning in Intelligent Systems: Networks of Plausible Inference
- Pearl
- 1988
(Show Context)
Citation Context ...ss caused, that the causes act independently to generate positive mammograms, and that both of the given causes are fairly strong. These assumptions can be captured using a noisy-OR parameterization (=-=Pearl, 1988-=-), a simple model to characterize independent generative causes that is equivalent to the parameterizations used in Cheng’s (1997) power PC theory or Rehder’s (2003) causal models for categorization. ... |

1476 |
Judgment under Uncertainty: Heuristics and Biases. Science 185:1124–31
- Tversky, Kahneman
- 1974
(Show Context)
Citation Context ...aken some framework of statistical inference to be the normative standard (Anderson, 1990; Gigerenzer & Hoffrage, 1995; McKenzie, 2003; Oaksford & Chater, 1994; Peterson & Beach, 1967; Shepard, 1987; =-=Tversky & Kahneman, 1974-=-). Statistical inference frameworks generally approach the judgment of an uncertain variable, such as whether someoneshas a disease, by considering the current data, such as the person’s symptoms, as ... |

1268 |
Causality: models, reasoning, and inference
- Pearl
- 2000
(Show Context)
Citation Context ...s with knowledge of causal relationships, resulting in more reliable judgments, with less data required than purely statistical methods. It is becoming clear from research in artificial intelligence (=-=Pearl, 2000-=-), associative learning (Cheng, 1997; Glymour, 2001; Gopnik & Glymour, 2002; Gopnik & Sobel, 2000; Waldmann, 1996), and categorization (Ahn, 1999; Rehder, 2003) that causal reasoning methods are often... |

654 | Graphical models - JORDAN |

498 |
The adaptive character of thought
- Anderson
- 1990
(Show Context)
Citation Context ...ks for Judgment Under Uncertainty Most previous accounts—whether arguing for or against human adherence to rationality—have taken some framework of statistical inference to be the normative standard (=-=Anderson, 1990-=-; Gigerenzer & Hoffrage, 1995; McKenzie, 2003; Oaksford & Chater, 1994; Peterson & Beach, 1967; Shepard, 1987; Tversky & Kahneman, 1974). Statistical inference frameworks generally approach the judgme... |

400 |
On the psychology of prediction
- Kahneman, Tversky
- 1973
(Show Context)
Citation Context ...across scenarios, yet we were able to induce use or neglect of the base rate by varying causal structure. The results also cast doubt on the hypothesis that people use a representativeness heuristic (=-=Kahneman & Tversky, 1973-=-) in social judgment tasks. A representativeness heuristic would judge P(male�over 5�7�) by the degree to which a person over 5�7� represents a typical male, ignoring the base rate of males on the tea... |

372 |
Toward a universal law of generalization for psychological science
- SHEPARD
- 1987
(Show Context)
Citation Context ...ionality—have taken some framework of statistical inference to be the normative standard (Anderson, 1990; Gigerenzer & Hoffrage, 1995; McKenzie, 2003; Oaksford & Chater, 1994; Peterson & Beach, 1967; =-=Shepard, 1987-=-; Tversky & Kahneman, 1974). Statistical inference frameworks generally approach the judgment of an uncertain variable, such as whether someoneshas a disease, by considering the current data, such as ... |

282 | How to improve Bayesian reasoning without instruction: frequency formats
- Gigerenzer, Hoffrage
- 1995
(Show Context)
Citation Context ...Under Uncertainty Most previous accounts—whether arguing for or against human adherence to rationality—have taken some framework of statistical inference to be the normative standard (Anderson, 1990; =-=Gigerenzer & Hoffrage, 1995-=-; McKenzie, 2003; Oaksford & Chater, 1994; Peterson & Beach, 1967; Shepard, 1987; Tversky & Kahneman, 1974). Statistical inference frameworks generally approach the judgment of an uncertain variable, ... |

259 |
From covariation to causation: a causal power theory
- Cheng
- 1997
(Show Context)
Citation Context ...hips, resulting in more reliable judgments, with less data required than purely statistical methods. It is becoming clear from research in artificial intelligence (Pearl, 2000), associative learning (=-=Cheng, 1997-=-; Glymour, 2001; Gopnik & Glymour, 2002; Gopnik & Sobel, 2000; Waldmann, 1996), and categorization (Ahn, 1999; Rehder, 2003) that causal reasoning methods are often better suited than purely statistic... |

211 | Are humans good intuitive statisticians after all? Rethinking some conclusions from the literature on judgment and uncertainty - Cosmides, Tooby - 1996 |

193 | Belief in the law of small numbers
- Tversky, Kahneman
- 1971
(Show Context)
Citation Context ... on people’s abilities to comprehend randomness, including such well-known phenomena as the gambler’s fallacy, the hot-hand fallacy (Gilovich, Vallone, & Tversky, 1985), and the law of small numbers (=-=Tversky & Kahneman, 1971-=-). In Experiments 1–3, as part of clarifying the causal structure of the scenario, we moved the main source of randomness from the efficacy of the mechanism to the presence or absence of other causal ... |

184 | A rational analysis of the selection task as optimal data selection
- Oaksford, Chater
- 1994
(Show Context)
Citation Context ...her arguing for or against human adherence to rationality—have taken some framework of statistical inference to be the normative standard (Anderson, 1990; Gigerenzer & Hoffrage, 1995; McKenzie, 2003; =-=Oaksford & Chater, 1994-=-; Peterson & Beach, 1967; Shepard, 1987; Tversky & Kahneman, 1974). Statistical inference frameworks generally approach the judgment of an uncertain variable, such as whether someoneshas a disease, by... |

183 | A theory of causal learning in children: causal maps and Bayes nets
- Gopnik, Glymour, et al.
- 2004
(Show Context)
Citation Context ...inference in real-world environments. Causal Bayesian networks have been proposed as tools for understanding how people intuitively learn and reason about causal systems (e.g., Glymour & Cheng, 1998; =-=Gopnik et al., 2004-=-; Griffiths & Tenenbaum, 2005; Sloman & Lagnado, 2005; Steyvers, Tenenbaum, Wagenmakers, & Blum, 2003; Tenenbaum & Griffiths, 2001, 2003; Waldmann, 2001), but their implications for more general pheno... |

124 | The hot hand in basketball: On the misperception of random sequences
- Gilovich, Vallone, et al.
- 1985
(Show Context)
Citation Context ...make a mistake 20% of the time. A number of studies have cast doubt on people’s abilities to comprehend randomness, including such well-known phenomena as the gambler’s fallacy, the hot-hand fallacy (=-=Gilovich, Vallone, & Tversky, 1985-=-), and the law of small numbers (Tversky & Kahneman, 1971). In Experiments 1–3, as part of clarifying the causal structure of the scenario, we moved the main source of randomness from the efficacy of ... |

106 | Structure and strength in causal induction
- Griffiths, Tenenbaum
- 2005
(Show Context)
Citation Context ...ld environments. Causal Bayesian networks have been proposed as tools for understanding how people intuitively learn and reason about causal systems (e.g., Glymour & Cheng, 1998; Gopnik et al., 2004; =-=Griffiths & Tenenbaum, 2005-=-; Sloman & Lagnado, 2005; Steyvers, Tenenbaum, Wagenmakers, & Blum, 2003; Tenenbaum & Griffiths, 2001, 2003; Waldmann, 2001), but their implications for more general phenomena of judgment under uncert... |

99 |
Inferring causal networks from observations and interventions
- Steyvers, Tenenbaum, et al.
- 2003
(Show Context)
Citation Context ...roposed as tools for understanding how people intuitively learn and reason about causal systems (e.g., Glymour & Cheng, 1998; Gopnik et al., 2004; Griffiths & Tenenbaum, 2005; Sloman & Lagnado, 2005; =-=Steyvers, Tenenbaum, Wagenmakers, & Blum, 2003-=-; Tenenbaum & Griffiths, 2001, 2003; Waldmann, 2001), but their implications for more general phenomena of judgment under uncertainty have not been systematically explored. We see the present article ... |

91 |
The Mind’s Arrows: Bayes Nets and Graphical Causal Models in Psychology (Cambridge
- Glymour
- 2001
(Show Context)
Citation Context ...ng in more reliable judgments, with less data required than purely statistical methods. It is becoming clear from research in artificial intelligence (Pearl, 2000), associative learning (Cheng, 1997; =-=Glymour, 2001-=-; Gopnik & Glymour, 2002; Gopnik & Sobel, 2000; Waldmann, 1996), and categorization (Ahn, 1999; Rehder, 2003) that causal reasoning methods are often better suited than purely statistical methods for ... |

78 |
Probabilistic reasoning in clinical medicine: Problems and opportunities
- Eddy
- 1982
(Show Context)
Citation Context ...cian hypothesis did not reign for long. It was not able to account for a rapidly accumulating body of experimental evidence that people reliably violate Bayesian norms (Ajzen, 1977; Bar-Hillel, 1980; =-=Eddy, 1982-=-; Lyon & Slovic, 1976; Nisbett & Borgida, 1975; Tversky & Kahneman, 1974). For example, consider the mammogram problem, a Bayesian diagnosis problem that even doctors commonly fail (Eddy, 1982). One w... |

70 |
Causal schemas in judgment under uncertainty
- Tversky, Kahneman
- 1982
(Show Context)
Citation Context ...ment by focusing on how it modulates the classic phenomenon of base-rate neglect. Early studies indicated that people were more likely to neglect base rates that lack “causal relevance” (Ajzen, 1977; =-=Tversky & Kahneman, 1980-=-), although the notion of causal relevance was never well defined. Bar-Hillel (1980) argued that the salience of the base rate determined whether people would use this information, and that causal rel... |

52 | Structure learning in human causal induction
- Tenenbaum, Griffiths
- 2001
(Show Context)
Citation Context ...tuitively learn and reason about causal systems (e.g., Glymour & Cheng, 1998; Gopnik et al., 2004; Griffiths & Tenenbaum, 2005; Sloman & Lagnado, 2005; Steyvers, Tenenbaum, Wagenmakers, & Blum, 2003; =-=Tenenbaum & Griffiths, 2001-=-, 2003; Waldmann, 2001), but their implications for more general phenomena of judgment under uncertainty have not been systematically explored. We see the present article as a first attempt in this di... |

51 | Detecting blickets: How young children use information about novel causal powers in categorization and induction
- Gopnik, Sobel
- 2000
(Show Context)
Citation Context ...s data required than purely statistical methods. It is becoming clear from research in artificial intelligence (Pearl, 2000), associative learning (Cheng, 1997; Glymour, 2001; Gopnik & Glymour, 2002; =-=Gopnik & Sobel, 2000-=-; Waldmann, 1996), and categorization (Ahn, 1999; Rehder, 2003) that causal reasoning methods are often better suited than purely statistical methods for inference in real-world environments. Causal B... |

51 |
Man as an intuitive statistician
- Peterson, Beach
- 1967
(Show Context)
Citation Context ...t human adherence to rationality—have taken some framework of statistical inference to be the normative standard (Anderson, 1990; Gigerenzer & Hoffrage, 1995; McKenzie, 2003; Oaksford & Chater, 1994; =-=Peterson & Beach, 1967-=-; Shepard, 1987; Tversky & Kahneman, 1974). Statistical inference frameworks generally approach the judgment of an uncertain variable, such as whether someoneshas a disease, by considering the current... |

49 |
The base-rate fallacy in probability judgments
- Bar-Hillel
- 1980
(Show Context)
Citation Context ...intuitive statistician hypothesis did not reign for long. It was not able to account for a rapidly accumulating body of experimental evidence that people reliably violate Bayesian norms (Ajzen, 1977; =-=Bar-Hillel, 1980-=-; Eddy, 1982; Lyon & Slovic, 1976; Nisbett & Borgida, 1975; Tversky & Kahneman, 1974). For example, consider the mammogram problem, a Bayesian diagnosis problem that even doctors commonly fail (Eddy, ... |

40 |
Categorization as causal reasoning
- Rehder
- 2003
(Show Context)
Citation Context ...from research in artificial intelligence (Pearl, 2000), associative learning (Cheng, 1997; Glymour, 2001; Gopnik & Glymour, 2002; Gopnik & Sobel, 2000; Waldmann, 1996), and categorization (Ahn, 1999; =-=Rehder, 2003-=-) that causal reasoning methods are often better suited than purely statistical methods for inference in real-world environments. Causal Bayesian networks have been proposed as tools for understanding... |

39 |
Knowledge-base causal induction
- Waldmann
- 1996
(Show Context)
Citation Context ...urely statistical methods. It is becoming clear from research in artificial intelligence (Pearl, 2000), associative learning (Cheng, 1997; Glymour, 2001; Gopnik & Glymour, 2002; Gopnik & Sobel, 2000; =-=Waldmann, 1996-=-), and categorization (Ahn, 1999; Rehder, 2003) that causal reasoning methods are often better suited than purely statistical methods for inference in real-world environments. Causal Bayesian networks... |

36 |
Intuitive theories of events and the effects of base-rate information on prediction
- Ajzen
(Show Context)
Citation Context ...P(D�¬H). The intuitive statistician hypothesis did not reign for long. It was not able to account for a rapidly accumulating body of experimental evidence that people reliably violate Bayesian norms (=-=Ajzen, 1977-=-; Bar-Hillel, 1980; Eddy, 1982; Lyon & Slovic, 1976; Nisbett & Borgida, 1975; Tversky & Kahneman, 1974). For example, consider the mammogram problem, a Bayesian diagnosis problem that even doctors com... |

35 | Ten years of the rational analysis of cognition - Chater, Oaksford - 1999 |

35 | God does not play dice: Causal determinism and children's inferences about unobserved causes - Schulz, Sommerville - 2006 |

32 |
Attribution and the psychology of prediction
- Nisbett, Borgida
(Show Context)
Citation Context ... long. It was not able to account for a rapidly accumulating body of experimental evidence that people reliably violate Bayesian norms (Ajzen, 1977; Bar-Hillel, 1980; Eddy, 1982; Lyon & Slovic, 1976; =-=Nisbett & Borgida, 1975-=-; Tversky & Kahneman, 1974). For example, consider the mammogram problem, a Bayesian diagnosis problem that even doctors commonly fail (Eddy, 1982). One well-tested version comes from Gigerenzer and H... |

31 | Teaching Bayesian reasoning in less than two hours - Sedlmeier, Gigerenzer - 2001 |

31 |
Predictive versus diagnostic causal learning: evidence from an overshadowing paradigm
- Waldmann
- 2001
(Show Context)
Citation Context ...usal systems (e.g., Glymour & Cheng, 1998; Gopnik et al., 2004; Griffiths & Tenenbaum, 2005; Sloman & Lagnado, 2005; Steyvers, Tenenbaum, Wagenmakers, & Blum, 2003; Tenenbaum & Griffiths, 2001, 2003; =-=Waldmann, 2001-=-), but their implications for more general phenomena of judgment under uncertainty have not been systematically explored. We see the present article as a first attempt in this direction, with a focus ... |

30 | Estimating causal strength: The role of structural knowledge and processing effort - Waldmann, Hagmayer - 2001 |

29 | Base rates in Bayesian inference: Signal detection analysis of the cab problem - Birnbaum - 1983 |

29 | Theory-based causal inference - Tenenbaum, Griffiths - 2002 |

22 | Causal mechanism and probability: a normative approach
- Glymour, Cheng
- 1998
(Show Context)
Citation Context ...tatistical methods for inference in real-world environments. Causal Bayesian networks have been proposed as tools for understanding how people intuitively learn and reason about causal systems (e.g., =-=Glymour & Cheng, 1998-=-; Gopnik et al., 2004; Griffiths & Tenenbaum, 2005; Sloman & Lagnado, 2005; Steyvers, Tenenbaum, Wagenmakers, & Blum, 2003; Tenenbaum & Griffiths, 2001, 2003; Waldmann, 2001), but their implications f... |

17 | Dominance of accuracy information and neglect of base rates in probability estimation - Lyon, Slovic |

17 | Cognitive processes and societal risk taking - Slovic, Fischhoff, et al. - 1976 |

16 |
Causal maps and Bayes nets: a cognitive and computational account of theoryformation
- Gopnik, Glymour
- 2002
(Show Context)
Citation Context ...able judgments, with less data required than purely statistical methods. It is becoming clear from research in artificial intelligence (Pearl, 2000), associative learning (Cheng, 1997; Glymour, 2001; =-=Gopnik & Glymour, 2002-=-; Gopnik & Sobel, 2000; Waldmann, 1996), and categorization (Ahn, 1999; Rehder, 2003) that causal reasoning methods are often better suited than purely statistical methods for inference in real-world ... |

16 | Learning, prediction and causal bayes nets - Glymour - 2003 |

13 |
Effect of causal structure on category construction
- Ahn
- 1999
(Show Context)
Citation Context ...ming clear from research in artificial intelligence (Pearl, 2000), associative learning (Cheng, 1997; Glymour, 2001; Gopnik & Glymour, 2002; Gopnik & Sobel, 2000; Waldmann, 1996), and categorization (=-=Ahn, 1999-=-; Rehder, 2003) that causal reasoning methods are often better suited than purely statistical methods for inference in real-world environments. Causal Bayesian networks have been proposed as tools for... |

12 |
Back to base rates
- Bar-Hillel
- 1990
(Show Context)
Citation Context ...ucture, as well as provide statistics that map onto that structure. For example, consider the statistic that patients are more likely to survive after being treated by Doctor B than by Doctor A (from =-=Bar-Hillel, 1990-=-). One could easily get the impression that Doctor A provides inferior care, unless one is specifically informed that Doctor A specializes in life-threatening diseases and Doctor B does not. This new ... |

10 |
On prediction and judgment
- Tversky, Kahneman
- 1972
(Show Context)
Citation Context ... be neglected even if they do fit into the model, because they may not be required in the final judgment. This is what we propose happens when people seem to neglect the base rate in the cab problem (=-=Kahneman & Tversky, 1972-=-). The cab problem was one of the earliest problems found to elicit base-rate neglect. In this problem, participants were told that a witness identified the color of a cab in a hit-and-run accident, c... |

8 |
Rational models as theories—not standards—of behavior
- McKenzie
- 2003
(Show Context)
Citation Context ...us accounts—whether arguing for or against human adherence to rationality—have taken some framework of statistical inference to be the normative standard (Anderson, 1990; Gigerenzer & Hoffrage, 1995; =-=McKenzie, 2003-=-; Oaksford & Chater, 1994; Peterson & Beach, 1967; Shepard, 1987; Tversky & Kahneman, 1974). Statistical inference frameworks generally approach the judgment of an uncertain variable, such as whether ... |

6 |
The inverse fallacy: an account of deviations from Bayes Theorem and the additivity principle
- Villejoubert, Mandel
- 2002
(Show Context)
Citation Context ...nation of the original mammogram problem suggests that people are confused by the given conditional probability P(�M�cancer) and think it means P(cancer��M). This has been called the inverse fallacy (=-=Villejoubert & Mandel, 2002-=-). We find this limited as an explanation, because people seem to exhibit the inverse fallacy only when they expect both probabilities to have roughly the same value. For instance, though people might... |

6 | Crimes analysis software: ‘pin maps’, clustering and Bayes net prediction”, Expert Systems and - Oatley, Ewart |

4 | The role of covariation vs. mechanism information in causal attribution - Ahn, Kalish, et al. - 1995 |

4 | The adaptive character of thought. Hillsdale, NJ: Erlbaum. KHEMLANI AND JOHNSON-LAIRD - Anderson - 1990 |

4 | Rational models as theories – not standards – of behaviour - McKenzie - 2003 |

3 | The role of causal models in reasoning under uncertainty - Krynski, Tenenbaum - 2003 |

3 |
The meaning and computation of causal power: A critique of Cheng
- Luhmann, Ahn
- 2005
(Show Context)
Citation Context ...indeterminism introduced through uncertainty about the presence of hidden causal variables, has recently been proposed in both the artificial intelligence (Pearl, 2000) and psychological literatures (=-=Luhmann & Ahn, 2005-=-; Schulz, Sommerville, & Gopnik, in press). Making Statistics Easy People clearly have natural abilities to interpret statistics, yet they are often poor at interpreting published statistical data. Pr... |