Mike Oaksford

A Rational Analysis of the Selection Task as Optimal Data Selection

Human reasoning in hypothesis-testing tasks like P. C. Wasons (1968) selection task has been depicted as prone to systematic biases. However, performance on this task has been assessed against a now outmoded falsificationist philosophy of science. Therefore, the experimental data is reassessed in the light of a Bayesian model of optimal data selection in inductive hypothesis testing. The model provides a rational analysis (J. R. Anderson, 1990) of the selection task that fits well with peoples performance on both abstract and thematic versions of the task. The model suggests that reasoning in these tasks may be rational rather than subject to systematic bias.

Probabilities and polarity biases in conditional inference.

A probabilistic computational level model of conditional inference is proposed that can explain polarity biases in conditional inference (e.g., J. St. B. T. Evans, 1993). These biases are observed when J. St. B. T. Evanss (1972) negations paradigm is used in the conditional inference task. The model assumes that negations define higher probability categories than their affirmative counterparts (M. Oaksford & K. Stenning, 1992); for example, P(not-dog) > P(dog). This identification suggests that polarity biases are really a rational effect of high-probability categories. Three experiments revealed that, consistent with this probabilistic account, when high-probability categories are used instead of negations, a high-probability conclusion effect is observed. The relationships between the probabilistic model and other phenomena and other theories in conditional reasoning are discussed.

Reasoning with conditionals containing negated constituents

Three experiments investigated matching bias in conditional reasoning tasks. Matching bias occurs when Ss ignore negations and match named items. Experiment 1 used an abstract and a thematic version of Evanss (1972) construction task. Results showed that matching may be due to an interaction between task demands and constructing contrast classes when interpreting negations. Experiment 2, which used Wasons (1968) selection task, introduced a manipulation to ease contrast-class construction. Confirmation plus falsification dominated over matching. Experiment 3 introduced two other manipulations to aid contrast-class construction with abstract material. Confirmation was facilitated, matching was suppressed, and falsification remained unchanged. These results suggest that matching occurs only when insufficient or ambiguous information prevents the intended interpretation of negations.

Rationality in an uncertain world : essays on the cognitive science of human reasoning

Part I: problems with logicism autonomy, implementation and cognitive architecture - a reply to Fodor and Pylyshyn connectionism, classical cognitive science, and experimental psychology against logicist cognitive science I - the core argument against logicist cognitive science II - objections and replies reasoning theories and bounded rationality bounded rationlity in taking risks and drawing inferences logicism and everyday reasoning - mental methods and mental logic the falsity of folk theories - implications for psychology and philosophy. Part II: the probablistic approach a rational analysis of the selection task I - optimal data selection a rational analysis of the selection task II - abstract materials a rational analysis of the selection task III - thematic materials a rational analysis of the selection task IV - implications rational explanation of the selection task information gain explains relevance, which explains the selection current developments and future directions.

Fast, frugal, and rational: How rational norms explain behavior

Much research on judgment and decision making has focussed on the adequacy of classical rationality as a description of human reasoning. But more recently it has been argued that classical rationality should also be rejected even as normative standards for human reasoning. For example, Gigerenzer and Goldstein (1996) and Gigerenzer and Todd (1999a) argue that reasoning involves “fast and frugal” algorithms which are not justified by rational norms, but which succeed in the environment. They provide three lines of argument for this view, based on: (A) the importance of the environment; (B) the existence of cognitive limitations; and (C) the fact that an algorithm with no apparent rational basis, Take-the-Best, succeeds in an judgment task (judging which of two cities is the larger, based on lists of features of each city). We reconsider (A)–(C), arguing that standard patterns of explanation in psychology and the social and biological sciences, use rational norms to explain why simple cognitive algorithms can succeed. We also present new computer simulations that compare Take-the-Best with other cognitive models (which use connectionist, exemplar-based, and decision-tree algorithms). Although Take-the-Best still performs well, it does not perform noticeably better than the other models. We conclude that these results provide no strong reason to prefer Take-the-Best over alternative cognitive models.

Ten years of the rational analysis of cognition

Rational analysis is an empirical program that attempts to explain the function and purpose of cognitive processes. This article looks back on a decade of research outlining the rational analysis methodology and how the approach relates to other work in cognitive science. We illustrate rational analysis by considering how it has been applied to memory and reasoning. From the perspective of traditional cognitive science, the cognitive system can appear to be a rather arbitrary assortment of mechanisms with equally arbitrary limitations. In contrast, rational analysis views cognition as intricately adapted to its environment and to the problems it faces.

Adaptive Non-Interventional Heuristics for Covariation Detection in Causal Induction: Model Comparison and Rational Analysis.

In this article, 41 models of covariation detection from 2 × 2 contingency tables were evaluated against past data in the literature and against data from new experiments. A new model was also included based on a limiting case of the normative phi-coefficient under an extreme rarity assumption, which has been shown to be an important factor in covariation detection (McKenzie & Mikkelsen, 2007) and data selection (Hattori, 2002; Oaksford & Chater, 1994, 2003). The results were supportive of the new model. To investigate its explanatory adequacy, a rational analysis using two computer simulations was conducted. These simulations revealed the environmental conditions and the memory restrictions under which the new model best approximates the normative model of covariation detection in these tasks. They thus demonstrated the adaptive rationality of the new model.

Optimal data selection: Revision, review, and reevaluation

Since it first appeared, there has been much research and critical discussion on the theory of optimal data selection as an explanation of Wason’s (1966, 1968) selection task (Oaksford & Chater, 1994). In this paper, this literature is reviewed, and the theory of optimal data selection is reevaluated in its light. The information gain model is first located in the current theoretical debate in the psychology of reasoning concerning dual processes in human reasoning. A model comparison exercise is then presented that compares a revised version of the model with its theoretical competitors. Tests of the novel predictions of the model are then reviewed. This section also reviews experiments claimed not to be consistent with optimal data selection. Finally, theoretical criticisms of optimal data selection are discussed. It is argued either that the revised model accounts for them or that they do not stand up under analysis. It is concluded that some version of the optimal data selection model still provides the best account of the selection task. Consequently, the conclusion of Oaksford and Chater’s (1994) original rational analysis (Anderson, 1990), that people’s hypothesis-testing behavior on this task is rational and well adapted to the environment, still stands.

Theories of reasoning and the computational explanation of everyday inference

Abstract Following Marr (1982), any computational account of cognition must satisfy constraints at three explanatory levels: computational, algorithmic, and implementational. This paper focuses on the first two levels and argues that current theories of reasoning cannot provide explanations of everyday defeasible reasoning, at either level. At the algorithmic level, current theories are not computationally tractable: they do not “scale-up” to everyday defeasible inference. In addition, at the computational level, they cannot specify why people behave as they do both on laboratory reasoning tasks and in everyday life (Anderson, 1990). In current theories, logic provides the computational-level theory, where such a theory is evident at all. But logic is not a descriptively adequate computational-level theory for many reasoning tasks. It is argued that better computational-level theories can be developed using a probabilistic framework. This approach is illustrated using Oaksford and Chaters (1994) probabil...

Optimal Data Selection in the Reduced Array Selection Task (RAST)

The predictions of M. Oaksford and N. Chaters (see record 1995-08271-001) optimal data selection (ODS) model for the reduced array selection task (RAST) were tested in 4 experiments. Participants tested a hypothesis, if p then q, by selecting cards showing q or not q instances. In Experiment 1, where selections were made from different sized stacks of q and not q cards, as P(q) increased, not q card selections rose, and q card selections fell, as predicted. Experiment 2 controlled for the possibility that stack height influenced responses; these results were also consistent with ODS. Experiment 3, which controlled further for this possibility, replicated Experiment 1. Experiment 4 addressed a final issue concerning the medium P(q) condition by concentrating on initial card selections; the results were again consistent with ODS. Although generally consistent with the ODS model, these experiments also suggest some interesting revisions.