Reasoning Dialogues

When people argue with others in conversation, they make a variety of conversational moves: They make claims, ask for justification of others' claims, attack claims, and attack claims' justifications. The arrangement of these moves gives argumentation its characteristic shape. This article illustrates a proposed format for conversations of this type, and it reviews some findings about the way people understand and evaluate these conversations. The findings suggest that judgments of the arguers' burden depend not only on the content of their claims, but also on the conversation's structure. In addition, judgments of the strength of a justification—an arguer's evidence or explanation—are a function of the argument's setting.     

Reasoning about evidence

Bayesians understand the notion of evidential support in terms of probability raising. Little is known about the logic of the evidential support relation, thus understood. We investigate a number of prima facie plausible candidate logical principles for the evidential support relation and show which of these principles the Bayesian evidential support relation does and which it does not obey. We also consider the question which of these principles hold for a stronger notion of evidential support.     

Reasoning with uncertain categories

Five experiments investigated how people use categories to make inductions about objects whose categorisation is uncertain. Normatively, they should consider all the categories the object might be in and use a weighted combination of information from all the categories: bet-hedging. The experiments presented people with simple, artificial categories and asked them to make an induction about a new object that was most likely in one category but possibly in another. The results showed that the majority of people focused on the most likely category in making inductions, although there was a group of consistently normative responders who used information from both categories (about 25% of our college population). Across experiments the overall pattern of results suggests that performance in the task is improved not by understanding the underlying principles of bet-hedging but by increasing the likelihood that multiple categories are in working memory at the time of the induction. We discuss implications for improving everyday inductions.     

The Selective Laziness of Reasoning

Reasoning research suggests that people use more stringent criteria when they evaluate others' arguments than when they produce arguments themselves. To demonstrate this “selective laziness,” we used a choice blindness manipulation. In two experiments, participants had to produce a series of arguments in response to reasoning problems, and they were then asked to evaluate other people's arguments about the same problems. Unknown to the participants, in one of the trials, they were presented with their own argument as if it was someone else's. Among those participants who accepted the manipulation and thus thought they were evaluating someone else's argument, more than half (56% and 58%) rejected the arguments that were in fact their own. Moreover, participants were more likely to reject their own arguments for invalid than for valid answers. This demonstrates that people are more critical of other people's arguments than of their own, without being overly critical: They are better able to tell valid from invalid arguments when the arguments are someone else's rather than their own.     

Infinite Reasoning

Our relationship to the infinite is controversial. But it is widely agreed that our powers of reasoning are finite. I disagree with this consensus; I think that we can, and perhaps do, engage in infinite reasoning. Many think it is just obvious that we can't reason infinitely. This is mistaken. Infinite reasoning does not require constructing infinitely long proofs, nor would it gift us with non-recursive mental powers. To reason infinitely we only need an ability to perform infinite inferences. I argue that we have this ability. My argument looks to our best current theories of inference and considers examples of apparent infinite reasoning. My position is controversial, but if I'm right, our theories of truth, mathematics, and beyond could be transformed. And even if I'm wrong, a more careful consideration of infinite reasoning can only deepen our understanding of thinking and reasoning.     

Reasoning serves argumentation in children

The argumentative theory of reasoning (Mercier & Sperber, in press-c) claims that reasoning evolved for argumentation: to find and evaluate arguments in dialogic contexts. The theory has drawn most of its supportive evidence from work with adults, leaving open the possibility that argumentive features of reasoning are in fact entirely learned. Evidence is reviewed here suggesting that the special relation between reasoning and argumentation holds at all ages. More specifically, it is argued that (a) children possess at least rudimentary argument skills, (b) they are able to reap the benefits of social reasoning from very early on, (c) confirmation bias is present as soon as they start to argue, and (d) children can be victims of the same biases that affect adults when they use reasoning in the wrong contexts. These claims strengthen the argumentative theory of reasoning and support a call for more research on the interactive features of reasoning in both adults and children.     

Soft constraints in interactive behavior: the case of ignoring perfect knowledge in-the-world for imperfect knowledge in-the-head

Constraints and dependencies among the elements of embodied cognition form patterns or microstrategies of interactive behavior. Hard constraints determine which microstrategies are possible. Soft constraints determine which of the possible microstrategies are most likely to be selected. When selection is non-deliberate or automatic the least effort microstrategy is chosen. In calculating the effort required to execute a microstrategy each of the three types of operations, memory retrieval, perception, and action, are given equal weight; that is, perceptual-motor activity does not have a privileged status with respect to memory. Soft constraints can work contrary to the designer’s intentions by making the access of perfect knowledge in-the-world more effortful than the access of imperfect knowledge in-the-head. These implications of soft constraints are tested in two experiments. In experiment 1 we varied the perceptual-motor effort of accessing knowledge in-the-world as well as the effort of retrieving items from memory. In experiment 2 we replicated one of the experiment 1 conditions to collect eye movement data. The results suggest that milliseconds matter. Soft constraints lead to a reliance on knowledge in-the-head even when the absolute difference in perceptual-motor versus memory retrieval effort is small, and even when relying on memory leads to a higher error rate and lower performance. We discuss the implications of soft constraints for routine interactive behavior, accounts of embodied cognition, and tool and interface design.     

Reasoning with conditionals: a test of formal models of four theories

The four dominant theories of reasoning from conditionals are translated into formal models: The theory of mental models (Johnson-Laird, P. N., & Byrne, R. M. J. (2002). Conditionals: a theory of meaning, pragmatics, and inference. Psychological Review, 109, 646-678), the suppositional theory (Evans, J. S. B. T., & Over, D. E. (2004). If. Oxford: Oxford University Press), a dual-process variant of the model theory (Verschueren, N., Schaeken, W., & d'Ydewalle, G. (2005). A dual-process specification of causal conditional reasoning. Thinking &Reasoning, 11, 278-293), and the probabilistic theory (Oaksford, M., Chater, N., & Larkin, J. (2000). Probabilities and polarity biases in conditional inference. Journal of Experimental Psychology: Learning, Memory, and Cognition, 26, 883-899). The first three theories are formalized as multinomial models. The models are applied to the frequencies of patterns of acceptance or rejection across the four basic inferences modus ponens, acceptance of the consequent, denial of the antecedent, and modus tollens. Model fits are assessed for two large data sets, one representing reasoning with abstract, basic conditionals, the other reflecting reasoning with pseudo-realistic causal and non-causal conditionals. The best account of the data was provided by a modified version of the mental-model theory, augmented by directionality, and by the dual-process model.     

Children's working understanding of knowledge sources: Confidence in knowledge gained from testimony

In three experiments, children aged between 3 and 5 years (N = 38, 52, 94; mean ages 3–7 to 5–2) indicated their confidence in their knowledge of the identity of a hidden toy. With the exception of some 3-year-olds, children revealed working understanding of their knowledge source by showing high confidence when they had seen or felt the toy, and lower confidence when they had been told its identity by an apparently well-informed speaker. Correct explicit source reports were not necessary for children to show relative uncertainty when the speaker subsequently doubted the adequacy of his access to the toy. After a 2-min delay, 3–4-year-olds, unlike 4–5-year-olds, failed to see the implications of the speaker's doubt about his access.     

Question answering from structured knowledge sources

We present an implemented approach for domain-restricted question answering from structured knowledge sources, based on robust semantic analysis in a hybrid NLP system architecture. We perform question interpretation and answer extraction in an architecture that builds on a lexical-conceptual structure for question interpretation, which is interfaced with domain-specific concepts and properties in a structured knowledge base. Question interpretation involves a limited amount of domain-specific inferences, and accounts for higher-level quantificational questions. Question interpretation and answer extraction are modular components that interact in clearly defined ways. We derive so-called proto queries from the linguistic representations, which provide partial constraints for answer extraction from the underlying knowledge sources. The search queries we construct from proto queries effectively compute minimal spanning trees from the underlying knowledge sources. Our approach naturally extends to multilingual question answering, and has been developed as a prototype system for two application domains: the domain of Nobel prize winners, and the domain of Language Technology, on the basis of the large ontology underlying the information portal LT World.     

问题类型和情绪状态对贝叶斯推理的影响

本研究以240名大学生为被试,探讨了问题类型和情绪状态对贝叶斯推理的影响。结果表明：(1)问题类型对贝叶斯推理有影响;(2)情绪状态对贝叶斯推理有影响;(3)问题类型与情绪状态对贝叶斯推理的影响存在交互作用。具体地,在经典乳癌问题中,被试在消极情绪状态下比在积极情绪状态下做出更高的概率估计,而在中彩问题中,被试在积极情绪状态下比在消极情绪状态下概率估计更高。     

因果信息在贝叶斯推理中的作用:问题情境的调节

以大学生为被试,考察因果信息在不同贝叶斯推理问题中的作用。结果表明:(1)问题情境的主效应显著(F(1,122)=16.33,p0.05,η~2=0.120),因果信息的主效应不显著((F(1,122)=1.161,p=0.283,η~2=0.01),但问题类型和因果信息的交互作用显著(F(1,122)=4.757,p0.05,η~2=0.038)。(2)不同类型的问题上因果信息的作用是不同的。具体表现为,在乳癌问题中,增加了因果信息,被试的正确率明显提高(χ~2(1)=6.759,p0.05,V=0.607),而在中彩问题中没有因果信息的推理成绩反而更好(χ~2(1)=1.864,p=0.221,V=0.167)。     

分类活动对归纳推理多样性效应的影响

归纳推理多样性效应被证实为普遍存在的,然而研究者Heit和Medin却对多样性效应的普遍性存在分歧。本文通过三个实验针对两位学者的分歧展开研究。结果发现:1.两位学者之间的分歧是由不同的分类活动引起的。当分类活动为外显时,被试的推理结果既不支持Medin,也不支持Heit;当分类活动内隐时,被试的推理结果支持Heit的观点。2.人们在日常生活及现实情境中,往往是根据直觉对事物进行分类及多样性推理。     

Rationality and reasoning with metaphors

Traditional accounts of rationality typically preclude metaphorical reasoning. We review research that has highlighted the pervasiveness of metaphors in creative problem solving, jurisprudence and history of science, and argue that any account of rationality must explicitly acknowledge the ontology of representation and include an ontology-changing mechanism. From these considerations, we present an interaction-based view of cognition and examine the problem of rationality in its terms. We argue that rationality is closely related to the attitude of a cognitive agent towards incoherency—by which we do not mean internal inconsistency but operational incongruity in the external world. We conclude that though rationality permits a change of ontology, it requires a healthy respect towards the autonomous structure of the environment.     

Distributed skill functions and the meshing of knowledge structures

The scope of knowledge space theory was extended by bringing into the picture the underlying skills and capabilities that are relevant to solving the problems in a knowledge domain. A major challenge to this approach comes from the need to aggregate distributed information on (partially) overlapping domains and skill sets. The notion of a distributed skill function is introduced for formalizing the integration of several skill functions that represent the assignment of skills to problems. It is shown that their consistency is captured by the meshability of the delineated knowledge structures. This result draws upon a characterization of the meshing of finite or infinite collections of knowledge structures, which extends and generalizes previous results on the binary case. The discussion covers implications for knowledge assessment and for practical applications, such as integrating skill assignments coming from different experts or distributed resources in technology-enhanced learning.     

Facts and Possibilities: A Model‐Based Theory of Sentential Reasoning

This article presents a fundamental advance in the theory of mental models as an explanation of reasoning about facts, possibilities, and probabilities. It postulates that the meanings of compound assertions, such as conditionals (if) and disjunctions (or), unlike those in logic, refer to conjunctions of epistemic possibilities that hold in default of information to the contrary. Various factors such as general knowledge can modulate these interpretations. New information can always override sentential inferences; that is, reasoning in daily life is defeasible (or nonmonotonic). The theory is a dual process one: It distinguishes between intuitive inferences (based on system 1) and deliberative inferences (based on system 2). The article describes a computer implementation of the theory, including its two systems of reasoning, and it shows how the program simulates crucial predictions that evidence corroborates. It concludes with a discussion of how the theory contrasts with those based on logic or on probabilities.     

Dynamics properties of knowledge acquisition

Repeated counteraction of current knowledge and reality observation generates new knowledge and can be viewed as a dynamic process. When observed objects are not results of human activity, the dynamics results in acquisition of new knowledge, usually as a refinement of the existing one. Otherwise, adjustment between incomplete knowledge and reality can be a result of intentional modification of objects in question, as it happens in design of any artifact. In this case convergence of the process is questionable, for example, it may imply the end of the technical progress. In worst case scenario, the same computable criteria used simultaneously to modify objects and evaluate the result quality causing drift from an established goal and devaluation of acquired knowledge.