Use this URL to cite or link to this record in EThOS:
Title: Knowledge selection in category-based inductive reasoning
Author: Crisp-Bright, Aimee Kay
ISNI:       0000 0004 2697 2932
Awarding Body: Durham University
Current Institution: Durham University
Date of Award: 2010
Availability of Full Text:
Access from EThOS:
Access from Institution:
Current theories of category-based inductive reasoning can be distinguished by the emphasis they place on structured and unstructured knowledge. Theories which draw on unstructured knowledge focus on associative strength, or temporal and spatial contiguity between categories. In contrast, accounts which draw on structured knowledge make reference to the underlying theoretical frameworks which relate categories to one another, such as causal or taxonomic relationships. In this thesis, it is argued that this apparent dichotomy can be resolved if one ascribes different processing characteristics to these two types of knowledge. That is, unstructured knowledge influences inductive reasoning effortlessly and relatively automatically, whereas the use of structured knowledge requires effort and the availability of cognitive resources. Understanding these diverging processes illuminates how background knowledge is selected during the inference process. The thesis demonstrates that structured and unstructured knowledge are dissociable and influence reasoning in line with their unique processing characteristics. Using secondary task and speeded response paradigms, it shows that unstructured knowledge is most influential when people are cognitively burdened or forced to respond fast, whereas they can draw on more elaborate structured knowledge if they are not cognitively compromised. This is especially evident for the causal asymmetry effect, in which people make stronger inferences from cause to effect categories, than vice versa. This Bayesian normative effect disappears when people have to contend with a secondary task or respond under time pressure. The next experiments demonstrate that this dissociation between structured and unstructured knowledge is also evident for a more naturalistic inductive reasoning paradigm in which people generate their own inferences. In the final experiments, it is shown how the selection of appropriate knowledge ties in with more domain-general processes, and especially inhibitory control. When responses based on structured and unstructured knowledge conflict, people’s ability to reason based on appropriate structured knowledge depends upon having relevant background knowledge and on their ability to inhibit the lure from inappropriate unstructured knowledge. The thesis concludes with a discussion of how the concepts of structured and unstructured knowledge illuminate the processes underlying knowledge selection for category-based inductive reasoning. It also looks at the implications the findings have for different theories of category-based induction, and for our understanding of human reasoning processes more generally.
Supervisor: Not available Sponsor: Not available
Qualification Name: Thesis (Ph.D.) Qualification Level: Doctoral
EThOS ID:  DOI: Not available