Steven A. Sloman

Feature-Based Induction

A connectionist model of argument strength that applies to categorical arguments involving natural categories and predicates about which subjects have few prior beliefs is proposed. An example is robins have sesamoid bones, therefore falcons have sesamoid bones. The model is based on the hypothesis that argument strength is related to the proportion of the conclusion category′s features that are shared by the premise categories. The model assumes a two-stage process. First, premises are encoded by connecting the features of premise categories to the predicate. Second, conclusions are tested by examining the degree of activation of the predicate upon presentation of the features of the conclusion category. The model accounts for 13 qualitative phenomena and shows close quantitative fits to several sets of argument strength ratings.

Base-rate respect: From ecological rationality to dual processes

The phenomenon of base-rate neglect has elicited much debate. One arena of debate concerns how people make judgments under conditions of uncertainty. Another more controversial arena concerns human rationality. In this target article, we attempt to unpack the perspectives in the literature on both kinds of issues and evaluate their ability to explain existing data and their conceptual coherence. From this evaluation we conclude that the best account of the data should be framed in terms of a dual-process model of judgment, which attributes base-rate neglect to associative judgment strategies that fail to adequately represent the set structure of the problem. Base-rate neglect is reduced when problems are presented in a format that affords accurate representation in terms of nested sets of individuals.

Similarity- versus rule-based categorization

An influential study by Rips (1989) provides the strongest evidence available that categorization cannot be reduced to similarity. In Rips’s study, subjects were presented asparse description of an object that mentioned only a value on a single dimension (e.g., “an object 3 inches in diameter”), followed by two categories (e.g., PIZZA and QUARTER), where one allowed more variability on the relevant dimension than did the other (the diameter of pizzas is more variable than that of quarters). Subjects judged the described object to be more likely to be a member of the variable category (PIZZA), but more similar to the nonvariable category (QUARTER). This dissociation between categorization and similarity strongly implies that categorization was not based on similarity. In our first experiment, we used sparse descriptions like Rips’s, as well asrich descriptions that contained features characteristic of the nonvariable category. We found that categorization trackedsimilarity with both kinds of descriptions. In a second experiment, we modified our procedure to be more like that of Rips’s by requiring subjects to think aloud while making their decisions. Now, like Rips, we found a dissociation between similarity and categorization with sparse items; with rich descriptions, categorization again tracked similarity. These findings are consistent with the hypothesis that categorization can be done in two ways, by similarity and byrule. An exclusive reliance on rule-based categorization seems to occur only when the description of the to-be-categorized object does not contain any features characteristic of any relevant category.

The advantage of timely intervention

Can people learn causal structure more effectively through intervention rather than observation? Four studies used a trial-based learning paradigm in which participants obtained probabilistic data about a causal chain through either observation or intervention and then selected the causal model most likely to have generated the data. Experiment 1 demonstrated that interveners made more correct model choices than did observers, and Experiments 2 and 3 ruled out explanations for this advantage in terms of informational differences between the 2 conditions. Experiment 4 tested the hypothesis that the advantage was driven by a temporal signal; interveners may exploit the cue that their interventions are the most likely causes of any subsequent changes. Results supported this temporal cue hypothesis.

Do we "do"?

A normative framework for modeling causal and counterfactual reasoning has been proposed by Spirtes, Glymour, and Scheines (1993; cf. Pearl, 2000). The framework takes as fundamental that reasoning from observation and intervention differ. Intervention includes actual manipulation as well as counterfactual manipulation of a model via thought. To represent intervention, Pearl employed the do operator that simplifies the structure of a causal model by disconnecting an intervened-on variable from its normal causes. Construing the do operator as a psychological function affords predictions about how people reason when asked counterfactual questions about causal relations that we refer to as undoing, a family of effects that derive from the claim that intervened-on variables become independent of their normal causes. Six studies support the prediction for causal (A causes B) arguments but not consistently for parallel conditional (if A then B) ones. Two of the studies show that effects are treated as diagnostic when their values are observed but nondiagnostic when they are intervened on. These results cannot be explained by theories that do not distinguish interventions from other sorts of events.

Frequency illusions and other fallacies

Cosmides and Tooby (1996) increased performance using a frequency rather than probability frame on a problem known to elicit base-rate neglect. Analogously, Gigerenzer (1994) claimed that the conjunction fallacy disappears when formulated in terms of frequency rather than the more usual single-event probability. These authors conclude that a module or algorithm of mind exists that is able to compute with frequencies but not probabilities. The studies reported here found that base-rate neglect could also be reduced using a clearly stated single-event probability frame and by using a diagram that clarified the critical nested-set relations of the problem; that the frequency advantage could be eliminated in the conjunction fallacy by separating the critical statements so that their nested relation was opaque; and that the large effect of frequency framing on the two problems studied is not stable. Facilitation via frequency is a result of clarifying the probabilistic interpretation of the problem and inducing a representation in terms of instances, a form that makes the nested-set relations amongst the problem components transparent.

The probability of causal conditionals.

Conditionals in natural language are central to reasoning and decision making. A theoretical proposal called the Ramsey test implies the conditional probability hypothesis: that the subjective probability of a natural language conditional, P(if p then q), is the conditional subjective probability, P(q/p). We report three experiments on causal indicative conditionals and related counterfactuals that support this hypothesis. We measured the probabilities people assigned to truth table cases, P(pq), P(p notq), P( notpq) and P( notp notq). From these ratings, we computed three independent predictors, P(p), P(q/p) and P(q/ notp), that we then entered into a regression equation with judged P(if p then q) as the dependent variable. In line with the conditional probability hypothesis, P(q/p) was by far the strongest predictor in our experiments. This result is inconsistent with the claim that causal conditionals are the material conditionals of elementary logic. Instead, it supports the Ramsey test hypothesis, implying that common processes underlie the use of conditionals in reasoning and judgments of conditional probability in decision making.

Categorical Inference Is Not a Tree: The Myth of Inheritance Hierarchies

One enduring principle of rational inference is category inclusion: Categories inherit the properties of their superordinates. In five experiments, I show that people do not consistently apply this principle when evaluating categorical arguments involving natural categories and a single nonexplainable predicate such as all electronic equipment has parts made of germanium, therefore all stereos have parts made of germanium. Participants frequently did not apply the category inclusion rule despite affirming the relevant categorical relation (e.g., stereos are electronic equipment). They failed to apply the rule even when categories were universally quantified unambiguously. Instead, judgments tended to be proportional to the similarity between premise and conclusion categories. Neglect of category inclusion relations was observed using arguments concerning natural kinds, artifacts, and social kinds.

Time as a guide to cause

How do people learn causal structure? In 2 studies, the authors investigated the interplay between temporal-order, intervention, and covariational cues. In Study 1, temporal order overrode covariation information, leading to spurious causal inferences when the temporal cues were misleading. In Study 2, both temporal order and intervention contributed to accurate causal inference well beyond that achievable through covariational data alone. Together, the studies show that people use both temporal-order and interventional cues to infer causal structure and that these cues dominate the available statistical information. A hypothesis-driven account of learning is endorsed, whereby people use cues such as temporal order to generate initial models and then test these models against the incoming covariational data.

When Explanations Compete: The Role of Explanatory Coherence on Judgements of Likelihood.

The likelihood of a statement is often derived by generating an explanation for it and evaluating the plausibility of the explanation. The explanation discounting principle states that people tend to focus on a single explanation; alternative explanations compete with the effect of reducing one anothers credibility. Two experiments tested the hypothesis that this principle applies to inductive inferences concerning the properties of everyday categories. In both experiments, subjects estimated the probability of a series of statements (conclusions) and the conditional probabilities of those conclusions given other related facts. For example, given that most lawyers make good sales people, what is the probability that most psychologists make good sales people? The results showed that when the fact and the conclusion had the same explanation the fact increased peoples willingness to believe the conclusion, but when they had different explanations the fact decreased the conclusions credibility. This decrease is attributed to explanation discounting; the explanation for the fact had the effect of reducing the plausibility of the explanation for the conclusion.