Beyond belief bias: Reasoning from conceptual structures by mental models manipulation

Mental models constitute an alternative to the rule-based systems in the explanation of human reasoning (Johnson-Laird, 1983). In this paper, we claim that the concept of believability generally used to categorize content and context effects is of little use within a semantic theory. Thus, we propose the use of categories that are directly extracted from subjective relations among concepts within the reasoning problem. We demonstrate that manipulations based on this kind of categorization produce predictable patterns of responses in reasoning problems. We present two experiments to test our predictions, using conditional and syllogistic reasoning problems, and in both cases, we demonstrate the influence of conceptual knowledge not only in natural contexts, but also in experimentally created artificial contexts.     

An associative framework for probability judgment: an application to biases

Three experiments show that understanding of biases in probability judgment can be improved by extending the application of the associative-learning framework. In Experiment 1, the authors used M. A. Gluck and G. H. Bower's (1988a) diagnostic-learning task to replicate apparent base-rate neglect and to induce the conjunction fallacy in a later judgment phase as a by-product of the conversion bias. In Experiment 2, the authors found stronger evidence of the conversion bias with the same learning task. In Experiment 3, the authors changed the diagnostic-learning task to induce some conjunction fallacies that were not based on the conversion bias. The authors show that the conjunction fallacies obtained in Experiment 3 can be explained by adding an averaging component to M. A. Gluck and G. H. Bower's model.     

Mental models in propositional reasoning and working memory's central executive

We examine the role of working memory's central executive in the mental model explanation of propositional reasoning by using two working memory measures: the classical “reading span” test by Daneman and Carpenter (1980 Daneman, M. and Carpenter, P. A. 1980. Individual differences in working memory and reading. Journal of Verbal Learning and Verbal Behavior, 19: 450–466. [Crossref] , [Google Scholar]) and a new measure. This new “reasoning span” measure requires individuals to solve very simple anaphora problems, and store and remember the word solution in a growing series of inferential problems. We present one experiment in which we check the involvement of the central executive in conditional and disjunctive inference tasks and compare predictions of the new reasoning span test with those of the classical reading span test. The results of the experiment confirm that reasoning responses, which according to mental model theory require high cognitive work, are predicted by working memory measures. Results also show that some reasoning responses are probably obtained by means of superficial biases or strategies that do not load working memory. The reasoning span test, which involves the central executive to a greater degree, predicts reasoning performance better than the reading span test. The significance and possibilities of the new measure in studying reasoning are discussed.     

Unless reasoning

We report the results of two experiments investigating conditional inferences from conditional unless assertions, such as Juan is not in León unless Nuria is in Madrid. Experiments 1 and 2 check Fillenbaum's hypothesis about the semantic similarity of unless with if not and only if assertions; both also examine inferential endorsements (Experiment 1) and endorsements and latencies (Experiment 2) of the four logically equivalent conditional formulations: if A then B, if not-B then not-A, A only if B and notA unless B. The results of these experiments show the similarity of unless and only if confirming that the representation of both conditionals from the outset probably include two possibilities directionally oriented from B to A; results also confirm the especial difficulty of unless assertions. The implications of the results are discussed in the context of recent psychological and linguistic theories of the meaning of unless.     

Inferences from semifactual 'even if' conditionals

We report three experiments on semifactual conditionals such as 'even if he had worn his seatbelt he would have been injured'. Semifactuals contain a counterfactual antecedent (the presupposed fact is, he did not wear a seatbelt) and a true consequent (the fact is, he was injured). The experiments show that from the denial of the antecedent, 'he did not wear his seatbelt', reasoners do not infer the standard conclusion 'he was not injured' but instead they infer the asymmetric conclusion, 'he was injured'. From the affirmation of the consequent, 'he was injured', they do not infer the standard conclusion 'he wore his seatbelt' but instead they infer that there is no valid conclusion. The first experiment shows this pattern for 'even if' subjunctive conditionals compared to 'if' indicative conditionals, the second extends it to 'even if' subjunctive conditionals compared to 'even though' indicative concessives, and the third extends it to 'if...also/still' subjunctive conditionals. The results suggest that people think about two possibilities to understand a semifactual: the conjecture, he wore his seatbelt and he was injured, and the presupposed facts, he did not wear his seatbelt and he was injured.     

Priming in deduction: a spatial arrangement task

The mental model theory assumes that people reason by manipulating mental representations of states of the word, called "mental models." In the present study we used a new deduction task based on diagrammatic premises. We show that a premise can prime other premises that induce similar mental models in a way analogous to the case of words with related meanings, which prime one another. We present three experiments. In Experiment 1 we used an evaluation task. In Experiment 2, a construction task was used. The priming effect was obtained in both cases. In a third experiment we show that the priming effect was still present when partcipants were instructed to ignore a prime displayed before the premises. In all three experiments we compared determinate and indeterminate problems and found faster responses in the former.     

Knowledge and question asking

The ability and the motivation for question asking are, or should be, some of the most important aims of education. Unfortunately, students neither ask many questions, nor good ones. The present paper is about the capacity of secondary school pupils for asking questions and how this activity depends on prior knowledge. To examine this, we use texts containing different levels of information about a specific topic: biodiversity. We found a positive relationship between the amount of information provided and the number of questions asked about the texts, supporting the idea that more knowledgeable people ask more questions. Some students were warned that there would be an exam after the reading, and this led to a diminishing number of questions asked, and yet this still did not significantly improve their exam scores. In such a case, it seems that reading was more concerned with immediacy, hindering critical thinking and the dialog between their previous ideas and the new information. Thus, question asking seems to be influenced not only by the amount of knowledge, but also by the reader's attitude towards the information.     

The semantics of 'unless' conditionals: evidence from 'unless' and disjunctive inferences

We report the results of an experiment investigating conditional inferences from conditional assertions such as 'Juan won't go to León unless Nuria goes to Madrid' and 'Either Nuria goes to Madrid or Juan won't go to León'. This experiment addresses Dancygier's claims about the semantics of 'unless' by examining inferential endorsements of 'not-A unless B' and 'Either B or not-A' in the canonical order, presenting the categorical premise after the conditional assertions, and in the inverse order, presenting the categorical premise before the conditional assertions. The results of the experiment confirm that the representation of 'unless' includes two possibilities, although as Dancygier holds one of the possibilities may not be complete. The implications of the results are discussed in the context of the strategic nature of conditional reasoning and recent convergent theories of linguistic processing.