Falsifying mental models: Testing the predictions of theories of syllogistic reasoning

Four experiments are reported that tested the claim, drawn from mental models theory, that reasoners attempt to construct alternative representations of problems that might falsify preliminary conclusions they have drawn. In Experiment 1, participants were asked to indicate which alternative conclusion(s) they had considered in a syllogistic reasoning task. In Experiments 2-4, participants were asked to draw diagrams consistent with the premises, on the assumption that these diagrams would provide insights into the mental representation being used. In none of the experiments was there any evidence that people constructed more models for multiple-model than for single-model syllogisms, nor was there any correlation between number of models constructed and overall accuracy. The results are interpreted as showing that falsification of the kind proposed by mental models theory may not routinely occur in reasoning.     

Mental model construction in linear reasoning: Evidence for the construction of initial annotated models

According to the mental model theory, reasoners build an initial model representing the information given in the premises. In the context of relational reasoning, the question arises as to which kind of representation is used to cope with indeterminate or multimodel problems. The present article presents an array of possible answers arising from the initial construction of complete explicit models, partial explicit models, partial implicit models, a single "isomeric" model, or a single annotated model. Predictions generated from these views are tested in two experiments that vary the problem structure and the number of models consistent with the premises. Analyses of the premise processing times, answering times and accuracy show that the annotated model yields the best fit of the data. Implications of these findings for the mental model theory as developed for relational reasoning are discussed.     

Working memory and syllogistic reasoning

The purpose of this study was to examine the relation between working memory span and syllogistic reasoning performance. In addition, performance for the reasoning task was compared to predictions made by mental model theory and the probability heuristics model. According to mental model theory, syllogisms that require the use of more mental models are more difficult. According to the probability heuristics model difficulty is related to the number of probabilistic heuristics that must be applied, or (for invalid syllogisms) inconsistencies between the derived and correct conclusion. The predictions of these theories were examined across two experiments. In general, people with larger working memory capacities reasoned better. Also, the responses made by people with larger capacities were more likely to correspond to the predictions made by both mental model theory and the probability heuristics model. Relations between working memory span and performance were also consistent with both theories.     

A Cognitive Theory of Graphical and Linguistic Reasoning: Logic and Implementation

We discuss external and internal graphical and linguistic representational systems. We argue that a cognitive theory of peoples' reasoning performance must account for (a) the logical equivalence of inferences expressed in graphical and linguistic form, and (b) the implementational differences that affect facility of inference. Our theory proposes that graphical representation limit abstraction and thereby aid “processibility”. We discuss the ideas of specificity and abstraction, and their cognitive relevance. Empirical support both comes from tasks which involve the manipulation of external graphics and tasks that do not. For the former, we take Euler's (1772) circles, provide a novel computational reconstruction, show how it captures abstractions, and contrast it with earlier construals and with Johnson-Laird's (1983) mental models representations. We demonstrate equivalence of the graphical Euler system, and the nongraphical mental models system. For tasks not involving manipulation of external graphics, we discuss text comprehension, and the mental performance of syllogisms. By positing an internal system with the same specificity as Euler's circles, we cover the mental models data, and generate new empirical predictions. Finally, we consider how the architecture of working memory explains why such specific representations are relatively easy to store.     

Spatial Mental Models in Linear Reasoning

There is ample evidence that linear syllogisms are solved by constructing mental models, which semantically represent states in the world (Evans, Newstead, &; Byrne, 1993; Vandierendonck &; De Vooght, 1996 a). Such models are built by combining the information presented in the premises into an integrated representation, which is often thought of as spatial. The present article (a) presents evidence for an integrated spatial representation of premise information and (b) proposes a model construction process based on scanning processes that locate the position where new information is to be integrated into the spatial model.     

Strategies in relational inference

Three experiments are reported in which the relationships between task format, item type, and strategy usage were investigated for a two-dimensional relational inference task. Contrary to past findings with linear syllogisms, it was found that parallel presentation (presenting problem statements simultaneously) did not result in any increased use of deduction rule processes compared with serial presentation (presenting problem statements individually). Instead, the results suggested that mental models were used by the majority of subjects, and that multiple models were more likely to be constructed with parallel presentation. It is proposed that, in general, multiple model construction will be more frequent for deduction tasks where the cognitive load is relatively low. Hence, contrary to suggestions by Polk and Newell (1995), reasoning in this way appears to be prevalent and highly robust—where supported by task format—even where the use of this strategy is disadvantageous.     

如何利用广义量词的语义性质判断扩展三段论的有效性

在国内外最新研究成果的基础上,笔者通过对文中的定理和推论的证明,主要说明了以下几点：（1）利用广义量词的相关语义性质,比如单调性和对称性,既可以解释亚氏三段论的有效性,又可以解释带有广义量词的扩展三段论的有效性;（2）一些有效的扩展三段论仅仅表征了广义量词的左或右单调性,还有一些有效的扩展三段论同时表征了广义量词的多个语义性质;（3）利用广义量词的东南或西北或西南或东北方向的单调性可以判断一些带有限制条件的扩展三段论的有效性。此研究将有利于广义量词理论的发展,对于计算机科学中的知识表示和知识推理的研究都具有较为重要的理论价值和实践意义。     

Problem decomposability as a factor in complexity of the dimensional change card sort task

According to cognitive complexity and control (CCC) theory complexity depends on number of levels of a hierarchy of rules. According to relational complexity (RC) theory complexity is a function of the number of related variables in the task, and the most difficult tasks are those in which there is a constraint on decomposition into simpler subtasks. One hundred and twenty, 3–6-year old children were tested on the standard dimensional change card sort (DCCS) task, or a modified version, the DecompDCCS, in which the dimensions could be decomposed into subtasks. The standard version requires two levels of a hierarchy to be processed, and is ternary relational according to RC theory, whereas the subtasks of the DecompDCCS are binary relational. The DecompDCCS was easier than the DCCS for 3–4 year-olds, but all 5–6 year-olds succeeded on both. The results indicate that decomposability into simpler subtasks, as suggested by RC theory, is a factor in difficulty of DCCS. The role of decomposability in other tasks that are persistently difficult for young children is discussed.     

Essay Review

The incompleteness and artificiality of the ‘traditional logic’ of the textbooks is reflected in the way that syllogisms are commonly enumerated. The number said to be valid varies, but all the numbers given are of a kind that logicians should find irritating. Even the apparent harmony of what is almost invariably said to be the total number of syllogisms, 256, turns out to be illusory. In the following, it is shown that the concept of a distribution-value, which is related to the traditional theory of distribution, and the familiar concept of quantity together suffice to produce a far better way of enumerating syllogisms and a more complete understanding of the systematic features of syllogistic logic.     

Children's Reasoning in Solving Relational Problems of Deduction

This article reports on a study of children's deductive reasoning in solving novel relational problems. Detailed protocols were obtained from 264 children (aged 9- 12 years) who verbalised their thinking as they solved the problems. The study included the development of a three-phase theory based on Johnson-Laird and Byrne's mental models perspective, but with some distinct modifications. These include a focus on the relational complexity entailed in model construction and in premise integration, and the advancement of four reasoning principles that are applied throughout problem solution (in contrast to Johnson-Laird's falsification processes as the hallmark of deductive reasoning). The reported case studies and the results of statistical analyses supported predictions arising from the proposed theory, including the key role of the reasoning principles. The results also showed that problem difficulty is a function of relational complexity, not of the number of models to be constructed, as argued by Johnson-Laird and Byrne.     

Cognitive approaches to posttraumatic stress disorder: the evolution of multirepresentational theorizing

The evolution of multirepresentational cognitive theorizing in psychopathology is illustrated by detailed discussion and analysis of a number of prototypical models of posttraumatic stress disorder (PTSD). Network and schema theories, which focus on a single, explicit aspect/format of mental representation, are compared with theories that focus on 2 or more explicit representational elements. The author argues that the latter theories provide a more complete account of PTSD data, though are not without their problems. Specifically, it is proposed that at least 3 separate representational elements-associative networks, verbal/propositional representations, and schemas-are required to generate a comprehensive cognitive theory of PTSD. The argument that the development of multirepresentational cognitive theory in PTSD is a paradigm case for the development of similar theories in other forms of psychopathology is elaborated, and a brief agenda is proposed promoting 2 levels of theorizing-deep, formal theory alongside more localized, applied theory.     

Syllogistic inference

This paper reviews current psychological theories of syllogistic inference and establishes that despite their various merits they all contain deficiencies as theories of performance. It presents the results of two experiments, one using syllogisms and the other using three-term series problems, designed to elucidate how the arrangement of terms within the premises (the ‘figure’ of the premises) affects performance. These data are used in the construction of a theory based on the hypothesis that reasoners construct mental models of the premises, formulate informative conclusions about the relations in the model, and search for alternative models that are counterexamples to these conclusions. This theory, which has been implemented in several computer programs, predicts that two principal factors should affect performance: the figure of the premises, and the number of models that they call for. These predictions were confirmed by a third experiment.     

依恋内部工作模式的社会认知研究

社会认知取向的成人依恋研究,在验证依恋理论的核心假设和探索与情绪调节策略、依恋系统活动有关的心理活动过程方面取得了长足的进步。这种把当前社会认知理论特别是心理表征这一重要概念运用到理解依恋的内部工作模式（internalworkingmodel）中去的价值已经逐渐显露。文章概述了关于内部工作模式与心理表征的关系及其在启动效应、可得性与通达性、记忆、情绪、归因等五个重要方面的研究进展,并对将来的研究进行了思考与展望     

The source of belief bias effects in syllogistic reasoning.

In studies of the belief bias effect in syllogistic reasoning, an interaction between logical validity and the believability of the conclusion has been found; in essence, logic has a larger effect on unbelievable than on believable conclusions. Two main explanations have been proposed for this finding. The selective scrutiny account claims that people focus on the conclusion and only engage in logical processing if this is found to be unbelievable; while the misinterpreted necessity account claims that subjects misunderstand what is meant by logical necessity and respond on the basis of believability when indeterminate syllogisms are presented. Experiments 1 and 2 compared the predictions of these two theories by examining whether the interaction would disappear if only determinate syllogisms were used. It did, thus providing strong support for the misinterpreted necessity explanation. However, the results are also consistent with a version of the mental models theory, and so Experiment 3 was carried out to compare these two explanations. The mental models theory received strong support, as it did also in the follow-up Experiments 4 and 5. It is concluded that people try to construct a mental model of the premises but, if there is a believable conclusion consistent with the first model they produce, then they fail to construct alternative models.     

Model theory of deduction: a unified computational approach

One of the most debated questions in psychology and cognitive science is the nature and the functioning of the mental processes involved in deductive reasoning. However, all existing theories refer to a specific deductive domain, like syllogistic, propositional or relational reasoning.
Our goal is to unify the main types of deductive reasoning into a single set of basic procedures. In particular, we bring together the microtheories developed from a mental models perspective in a single theory, for which we provide a formal foundation. We validate the theory through a computational model (UNICORE) which allows fine-grained predictions of subjects' performance in different reasoning domains.
The performance of the model is tested against the performance of experimental subjects—as reported in the relevant literature—in the three areas of syllogistic, relational and propositional reasoning. The computational model proves to be a satisfactory artificial subject, reproducing both correct and erroneous performance of the human subjects. Moreover, we introduce a developmental trend in the program, in order to simulate the performance of subjects of different ages, ranging from children (3–6) to adolescents (8–12) to adults (>21). The simulation model performs similarly to the subjects of different ages.
Our conclusion is that the validity of the mental model approach is confirmed for the deductive reasoning domain, and that it is possible to devise a unique mechanism able to deal with the specific subareas. The proposed computational model (UNICORE) represents such a unifying structure.     

The effects of relational structure on analogical learning

Relational structure is important for various cognitive tasks, such as analogical transfer, but its role in learning of new relational concepts is poorly understood. This article reports two experiments testing people’s ability to learn new relational categories as a function of their relational structure. In Experiment 1, each stimulus consisted of 4 objects varying on 2 dimensions. Each category was defined by two binary relations between pairs of objects. The manner in which the relations were linked (i.e., by operating on shared objects) varied between subjects, producing 3 logically different conditions. In Experiment 2, each stimulus consisted of 4 objects varying on 3 dimensions. Categories were defined by three binary relations, leading to six logically different conditions. Various learning models were compared to the behavioral data, based on the theory of schema refinement. The results highlight several shortcomings of schema refinement as a model of relational learning: (1) it can make unreasonable demands on working memory, (2) it does not allow schemas to grow in complexity, and (3) it incorrectly predicts learning is insensitive to relational structure. We propose schema elaboration as an additional mechanism that provides a more complete account, and we relate this mechanism to previous proposals regarding interactions between analogy and representation construction. The current findings may advance understanding of the cognitive mechanisms involved in learning and representing relational concepts.     

Premise Misinterpretation and Syllogistic Reasoning

A study by Ceraso and Provitera (1971) found that elaboration of the premises used in syllogistic reasoning led to substantially improved performance. This finding is of considerable importance because of the implications it has for mental logic and mental models theories of reasoning. Three experiments are reported, which replicated and extended the original findings. It was found that elaboration led to a significant improvement in performance, but that this was confined to multiple model syllogisms, where the elaboration has the effect of reducing the number of models involved. A fourth experiment indicated that elaboration can vary within the same syllogism depending on the direction of the conclusion drawn. These findings are best explained under the assumption that reasoners build mental models when solving problems and that elaboration can reduce the number of possible models.     

Reasoning deficits among illicit drug users are associated with aspects of cannabis use

Deficits in deductive reasoning have been observed among ecstasy/polydrug users. The present study seeks to investigate dose-related effects of specific drugs and whether these vary with the cognitive demands of the task. One hundred and five participants (mean age 21.33, SD 3.14; 77 females, 28 males) attempted to generate solutions for eight one-model syllogisms and one syllogism for which there was no valid conclusion. All of the one-model syllogisms generated at least one valid conclusion and six generated two valid conclusions. In these six cases, one of the conclusions was classified as common and the other as non-common. The number of valid common inferences was negatively associated with the aspects of short-term cannabis use and with measures of IQ. The outcomes observed were more than simple post-intoxication effects since cannabis use in the 10 days immediately before testing was unrelated to reasoning performance. Following adjustment for multiple comparisons, the number of non-common valid inferences was not significantly associated with any of the drug-use measures. Recent cannabis use appears to impair the processes associated with generating valid common inferences while not affecting the production of non-common inferences. It is possible, therefore, that the two types of inference may recruit different executive resources, which may differ in their susceptibility to cannabis-related effects.     

Theories of categorical reasoning and extended syllogisms

The aim of this study was to examine the predictions of three theories of human logical reasoning, (a) mental model theory, (b) formal rules theory (e.g., PSYCOP), and (c) the probability heuristics model, regarding the inferences people make for extended categorical syllogisms. Most research with extended syllogisms has been restricted to the quantifier “All” and to an asymmetrical presentation. This study used three-premise syllogisms with the additional quantifiers that are used for traditional categorical syllogisms as well as additional syllogistic figures. The predictions of the theories were examined using overall accuracy as well as a multinomial tree modelling technique. The results demonstrated that all three theories were able to predict response selections at high levels. However, the modelling analyses showed that the probability heuristics model did the best in both Experiments 1 and 2.