Use this URL to cite or link to this record in EThOS:
Title: Rationality in an uncertain world : essays on the cognitive science of human reasoning
Author: Oaksford, M. R.
Awarding Body: University of Edinburgh
Current Institution: University of Edinburgh
Date of Award: 2001
Availability of Full Text:
Full text unavailable from EThOS.
Please contact the current institution’s library for further details.
In the psychology of reasoning there has been a longstanding assumption that people ought to be able to reason locally. However, prima facie this "logicist" position is called into doubt by the many results in the psychology of deductive reasoning showing that people do not give logical responses on tasks that have straightforward logical solutions. These results notwithstanding, some theorists have continued to insist that people ought to behave logically on these tasks. It is claimed that people's psychological procedures, although capable of logical inference in principle, fall into error in practice because of cognitive limitations of such working memory capacity. In this book it is argued that a more radical response is required especially when one considers that in some tasks this strategy writing off up to 96% of participants responding as "performance error". It is argued that logic has always provided a poor model of everyday human reasoning that people must do to survive in their uncertain world. Part 1 of the book concentrates on arguments against Logicism that indicate that a key issue is dealing with the "defeasibility" of every day inference. That is, the fact that inferences, such as if I turn the key the car starts and I turn the key, therefore the car starts can be defeated when the car does not start. The problems that have arisen in Artificial Intelligence in attempting to capture such inferences over everyday world knowledge are introduced and their implications for the psychology of reasoning are made clear. Part 2 of the book introduces a tentative probabilistic solution to these problems. It is argued that people's performance on deductive reasoning tasks results from generalising their everyday probabilistic reasoning strategies, that are generally adaptive in their normal environment, to the laboratory. These strategies, although normally adaptive and rational compared to a probabilistic standard, create the semblance of biased and irrational reasoning when compared to the standard provided by formal logic.
Supervisor: Not available Sponsor: Not available
Qualification Name: Thesis (D.Sc.) Qualification Level: Doctoral
EThOS ID:  DOI: Not available