this paper. In particular, the protocols we have collected often involve picturing different situations (e.g., a mental map of South America, images of savannas, or an advertisement showing Juan Valdez on his coffee plantation in Colombia). These im- ages can be taken as evidence for the manipulation of mental models in Johnson-Laird's terms. But overlaying this manipulation of mental models are the systematic patterns in which they are deployed to support one's con- clusions (cf. Rips, 1986). So while mental models may be part of the story of plausible reasoning, there is another critical part which the theory we pro- pose addresses. The theory does not address the issue of whether people make systematic errors in their reasoning, as the psychological literature on decision making (Kahneman, Slovic, & Tversky, 1982) attempts to document. This issue does not arise in the theory because we are developing a formalism for representing the kinds of inferences people make and the parameters that affect their certainty, rather than a theory about how people make particular inferences. People may systematically ignore some kinds of information or undervalue particular certainty parameters--we have not attempted to determine whether they do or not. Instead we have tried to represent all the kinds of reasoning patterns and the kinds of certainty parameters that appear in the protocols we have analyzed (Collins, 1978a, 1978b). In this regard it is worth pointing out that certain fallacles in logic, such as affirming the consequent (Havi- land, 1974), become plausible inference patterns in the theory.' The theory was developed to account for protocols where. a question drives the search fo relevant information; in Artificial Intelligence this is called backward inferencing. One qu...

Theory-based Bayesian models of inductive reasoning 2 Theory-based Bayesian models of inductive reasoning

The ability to categorize and use concepts e#ectively is a basic requirementofany intelligent actor. The utility-based approach to categorization is founded on the thesis that categorization is fundamentally in service of action, i.e., the choice of concepts made by an actor is critical to its choice of appropriate actions. This is in contrast to classical and similarity-based approaches which seek logical completeness in concept description with respect to sensory data rather than action-oriented e#ectiveness. Utility-based categorization is normative and not descriptive. It prescribes howanintelligent agent ought to conceptualize to act e#ectively. It provides ideals for categorization, speci#es criteria for the design of e#ective computational agents, and provides a model of ideal competence. A decision-theoretic framework for utilitybased categorization whichinvolves reasoning about alternative categorization models of varying levels of abstraction is proposed. Categorization mode...

Typical examples, that is, examples that are representative for a particular situation or concept, play an important role in human knowledge representation and reasoning. In real life situations more often than not, instead of a lengthy abstract characterization, a typical example is used to describe the situation. This well-known observation has been the motivation for various investigations in experimental psychology, which also motivate our formal characterization of typical examples, based on a partial order for their typicality. Reasoning by typical examples is then developed as a special case of analogical reasoning using the semantic information contained in the corresponding concept structures. We derive new inference rules by replacing the explicit information about connections and similarity, which are normally used to formalize analogical inference rules, by information about the relationship to typical examples. Using these inference rules analogical reasoning proceeds by c...

This thesis develops a cognitive linguistic approach to language comprehension. The cognitive approach differs from traditional linguistic approaches in that linguistic description is seen as an integral part of the description of cognition, and that the object of description is the nature of conceptual structures, the processes which relate these conceptual structures, and the effect of context upon these processes. As a cognitive description within cognitive science, a computational approach is adopted: language comprehension is described in terms of two modules, a linguistic processing module and a discourse processing module. Within these modules, conceptual structures and processes are given a uniform characterization: structures are characterized as partial objects which are extended by processes into (potentially) less partial objects. In the linguistic processing module, linguistic expressions are characterized as signs which combine as head and modifier. The conceptual structu...

This chapter presents a brief review of a computational theory of human plausible reasoning developed by Collins and Michalski, and discusses experiments conducted toward its validation. This is a descriptive theory that attempts to describe how people actually mason from imperfect premises, in contrast to well-studied normarive theories, such as probabilistic reasoning, non-monotonic reasoning, fuzzy logic and multiple-valued logic. The theory proposes a variety of inference patterns that do not occur in formal logic-based theories. It combines semantic and parametric aspects of reasoning, and demonstrates that a large part of human plausible reasoning can be described as small "perturbations" of believed knowledge structures. Ideas are illustrated by the analysis of two protocols recording the explanations of the reasoning process given by human subjects. Preliminary conclusions and directions for future research are presented.

When Konrad Lorenz was awarded the Nobel Prize for medicine in 1973 he delivered the lecture "Analogy as a Source of Knowledge" and acknowledged that "...this procedure (analogical

Typical instances, that is, instances that are representative for a particular situation or concept, play an important role in human knowledge representation and reasoning, in particular in analogical reasoning. This well-known observation has been a motivation for investigations in cognitive psychology which provide a basis for our characterization of typical instances within concept structures and for a new inference rule for justified analogical reasoning with typical instances. In a nutshell this paper suggests to augment the propositional knowledge representation system by a non-propositional part consisting of concept structures which may have directly represented instances as elements. The traditional reasoning system is extended by a rule for justified analogical inference with typical instances using information extracted from both knowledge representation subsystems. Keywords: analogical reasoning, typical instance, hybrid knowledge representation 1 Introduction The traditio...

. The hallmark of traditional Artificial Intelligence (AI) research is the symbolic representation and processing of knowledge. This is in sharp contrast to many forms of human reasoning, which to an extraordinary extent, rely on cases and (typical) examples. Although these examples could themselves be encoded into logic, this raises the problem of restricting the corresponding model classes to include only the intended models. There are, however, more compelling reasons to argue for a hybrid representation based on assertions as well as examples. The problems of adequacy, availability of information, compactness of representation, processing complexity, and last but not least, results from the psychology of human reasoning, all point to the same conclusion: Common sense reasoning requires different knowledge sources and hybrid reasoning principles that combine symbolic as well as semantic-based inference. In this paper we address the problem of integrating semantic representations of...

In this paper we investigate the role of category size in category-based induction. In a series of three experiments we asked participants about the strength of inductive inferences from arbitrary subordinate categories to their superordinates.