The goal of the research reported in this thesis is to explore analogical reasoning in complex situations. In three manuscripts, novel aspects of analogy use are investigated. In Manuscript 1, analogies used in a political campaign were analyzed with the framework developed in the analogical reasoning literature. Results show a number of novel features of analogy use. The majority of source analogs used were not superficially similar to the target problem, most of the mappings between source and target were implicit, and emotion appeared to be an important feature in the selection of source analogs. In Manuscript 2, three experiments were conducted that explored some of the effects uncovered in the first study. In these experiments, participants were asked either to generate their own source analogs in relation to a target problem or to retrieve a source from a predetermined set. Results show that when generating their own sources, people are not constrained by superficial similarity. However, when asked to retrieve from a predetermined set of sources, participants retrieve based on surface similarity. These results suggest that previous studies may have underestimated people's ability to use structural features in analogical retrieval. The research reported in Manuscript 3 explores the impact of analogical inferences on the representations of target problems. Descriptions of target problems were presented followed by potential source analogs. Results show that people engage in analogical mapping and draw inferences. These inferences, that were not presented, are incorporated in people's representation of the target and cannot be differentiated from information that was actually presented. People falsely recognize analogical inferences as having been presented when in fact they had not. Results from all these studies are discussed in terms of the novel insights they contribute to the analogy literature and in terms of their implications for theoretical models of

Human agents have to deal with a considerable amount of information from their environment and are also continuously faced with the need to take actions. As that information is largely of an uncertain nature, human agents have to decide whether, or how much, to believe individual pieces of information. To enable a reasoning system to deal in general with the demands of a real environment, and with information from human sources in particular, requires tools for uncertainty management and belief formation. This thesis presents a model for the management of uncertain information from human sources. Dealing, more specifically, with information which has been pre-processed by a natural language processor and transformed into an event-based representation, the model assesses information, forms beliefs and resolves conflicts between them in order to maintain a consistent world model. The approach is built on the fundamental principle that the uncertainty of information from people can, in the majority of situations, successfully be assessed through source models which record factors concerning the source's abilities and trustworthiness. These models are adjusted to reflect changes in the behaviour of the source. A mechanism is presented together with the underlying principles to reproduce such a behaviour. A high-level design is also given to make the proposed model reconstructible, and the successful operation of the model is demonstrated on two detailed examples.

No description available.

Deductive reasoning is a fundamental cognitive skill, and consequently has been the focus of much research over the past several decades. In the realm of syllogistic reasoning—judging the validity of a conclusion given two premises—a robust finding is the belief bias effect: broadly, the tendency for reasoners to judge as valid more believable than unbelievable conclusions. How the content believability of conclusions influences syllogistic reasoning has been the subject of hundreds of experiments and has informed several theories of deductive reasoning; however, how the content of premises influences the reasoning processes has been largely overlooked. In this thesis, I present 5 experiments that examine how premise content influences reasoning about categorical (i.e., statements with the words ‘some’ and ‘not’) and conditional (i.e., ‘if/then’ statements) syllogisms, which tend to be treated as interchangeable in deductive reasoning literature. It is demonstrated that premise content influences reasoning in these two types of syllogisms in fundamentally different ways. Specifically, Experiment 1 replicates and extends previous findings and demonstrates that for conditional syllogisms, belief bias results when premises are both believable and unbelievable; however, reasoners are more likely to judge that a conclusion is valid when it follows from believable than from unbelievable premises. Conversely, belief bias for categorical syllogisms results only when premises are believable; conclusion believability does not influence conclusion endorsement when premises are unbelievable. Based on these preliminary findings, I propose a theory that categorical and conditional syllogisms differ in the extent to which reasoners initially assume the premises to be true, and that this difference influences when in the reasoning process reasoners evaluate the believability of premises. Specifically, I propose that reasoners automatically assume that conditional, but not categorical, premises are true. It is proposed that, because the word “if” in conditional statements elicits hypothetical thinking, conditional premises are assumed to be true for the duration of the reasoning process. Subsequent to reasoning, premises can be “disbelieved” in a time-consuming process, and initial judgments about the conclusion may be altered, with a bias to respond that conclusions following from believable premises are valid. On the other hand, because categorical premises are phrased as factual propositions, reasoners initially judge the believability of categorical premises prior to reasoning about the conclusion. Unbelievable premises trigger the reasoner to disregard content from the rest of the syllogism, perhaps because the reasoner believes that the information in the problem will not be helpful in solving the problem. This theory is tested and supported by four additional experiments. Experiment 2 demonstrates that reasoners take longer to reason about conditional syllogisms with unbelievable than believable premises, consistent with the theory that unbelievable premises are “disbelieved” in a time-consuming process. Further, participants demonstrate belief bias for categorical syllogisms with unbelievable premises when they are instructed to assume that premises are true (Experiment 3) or when the word ‘if’ precedes the categorical premises (Experiment 4). Finally, Experiment 5 uses eye-tracking to demonstrate that premise believability influences post-conclusion premise looking durations for conditional syllogisms and pre-conclusion premise looking durations for categorical syllogisms. This finding supports the hypothesis that reasoners evaluate the believability of conditional premises after reasoning about the conclusion but that they evaluate the believability of categorical premises before reasoning about the conclusion. Further, Experiment 5 reveals that participants have poorer memory for the content of categorical syllogisms with unbelievable than believable premises, but memory did not differ for conditional syllogisms with believable and unbelievable premises. This suggests that unbelievable premise content in categorical syllogism is suppressed or ignored. These results and the theory of premise evaluation that I propose are discussed in the context of contemporary theories of deductive reasoning.

No description available.

Thesis (Ph. D.)--University of Washington, 1993. / Vita. Includes bibliographical references (leaves [245]-254).

Thesis (M.S.)--University of Wisconsin--Madison, 1963. / Typescript. eContent provider-neutral record in process. Description based on print version record. Includes bibliographical references (leaf 28)

Thesis (Ph. D.)--Brigham Young University. Dept. of Instructional Psychology and Technology, 2008. / Includes bibliographical references (p. 84-90).

Thesis (Ph. D.)--Columbia university, 1929. / Vita. "Reprinted from Archives of psychology ... no. 102." Bibliography: p. 79.

Thesis (M.S.) in Spatial Information Science and Engineering--University of Maine, 2002. / Includes vita. Includes bibliographical references (leaves 118-124 ).