Doctoral Dissertations
Date of Award
8-1989
Degree Type
Dissertation
Degree Name
Doctor of Philosophy
Major
Psychology
Major Professor
Michael G. Johnson
Committee Members
Kathleen Emmett, John W. Lounsbury, Howard Pollio, John Maolne, Bruce MacLennan
Abstract
This dissertation concerns the importance of conceptual relations. For this study conceptual relations will be understood as concepts, or terms represenative of concepts (such as verbs like "makes" or prepositions like "on").
The dissertation begins with a consideration of a longstanding problem in psychology, the antinomy between psychological structures and context effects. This problem seems an appropriate starting place for the dissertation for two reasons. First, consideration of the antinomy has frequently implicated conceptual relations as a direction in which to explore a solution to the longstanding problem. As such, a review of thought concerning this antinomy is largely overlapping with a review of what we know about the importance of conceptual relations.
Second, and relatedly, while the research in the dissertation will attempt to illuminate our understanding of conceptual relations per se, it does so with a constant eye toward producing new insights to the antinomy. In this way, the study of conceptual relations, a rather etheral set of entities (as James, 1890, notes), can be validated and iii grounded in a very real problem-- the antinomy between structure and flexibility.
After reviewing the literature reguarding both the antinomy and conceptual relations, a model, incorporating a number of disparate insights about conceptual relations is outlined. The basic infrastructure of this model of "what relations do" is common to both Gestalt psychology and connectionism. This model joins these, and other previously isolated ideas, to spell out, at least in a preliminary fashion, more specifically how and why conceptual relations are important to perception, language, thought, as well as the antinomy which has haunted explanations of these cognitive processes.
The model provides two directions for further research. The first is to attempt to define which relations are indeed primary. The second is to provide a more complete account of the sense in which they are primary. The empirical studies which are reported in this dissertation have been designed 9 to both develop and evaluate this existing model of conceptual relations on these two points.
The first four studies attempt to recover which relations are indeed primary in structuring thought across a series of different tasks (word associations, analogies and simple descriptions). The findings of these studies are then contrasted with existing literature and with mattimatical set theory.
The final three studies attempt both to validate these findings and to explore in what sense these relations are primary. Results suggest that these terms make randomly generated phrases more meaningful, form the basis for analogical reasoning, and are primary in category formation.
The concluding discussion outlines then an enhanced model of how relations structure thought and are at play in the antinomy. The model draws heavily on the works of Husserl and Wittgenstein, and as such could be deemed phenomenological. Importantly, it is phenomenological in the sense that it is a model of experience as opposed to "unseen processes" or brain structures. As such, it is contrasted in detail to the current connectionist view. The conclusion drawn from this comparison is that the two accounts are very similar, but that important and useful benifits are derived from modeling experience as opposed to brain structure. Most important is that connectionism seems unable to resolve the antinomy, but that this new model can at least predict and explain its occurrence.
Recommended Citation
Henley, Tracy Branson, "The primacy of relations. " PhD diss., University of Tennessee, 1989.
https://trace.tennessee.edu/utk_graddiss/11663
