Bridging Levels of Analysis: Learning, Information Theory, and the Lexicon

Autor/in	Dye, Melody
Titel	Bridging Levels of Analysis: Learning, Information Theory, and the Lexicon
Quelle	(2017), (190 Seiten) PDF als Volltext Verfügbarkeit Ph.D. Dissertation, Indiana University
Sprache	englisch
Dokumenttyp	gedruckt; online; Monographie
ISBN	978-0-3552-2364-4
Schlagwörter	Hochschulschrift; Dissertation; Computational Linguistics; Natural Language Processing; Artificial Languages; Computer Software; Language Usage; Psycholinguistics; Problem Solving; Learning Processes; Memory; Theories + Suchen Sie Ihr Suchwort? Thesis; Dissertations; Academic thesis; Linguistics; Computerlinguistik; Natürliche Sprache; Sprachgebrauch; Psycholinguistik; Problemlösen; Learning process; Lernprozess; Gedächtnis; Theory; Theorie
Abstract	While information theory is typically considered in the context of modern computing and engineering, its core mathematical principles provide a potentially useful lens through which to consider human language. Like the artificial communication systems such principles were invented to describe, natural languages involve a sender and receiver, a finite code, and a basic transmission problem. Human languages can thus be seen as socially evolved systems that have been structured to optimize information flow in communication. Over the past several decades, information theoretic approaches have attracted widespread interest among linguists and psychologists, and generated a productive research program focused on cataloging how speakers (in their utterances) and languages (in their design) conform to its principles. However, comparatively little work has been done to explicate how the cognitive mechanisms that subserve language give rise to such apparently rational behavior. Showing why speakers conform to such principles is an important correlate to showing that they do. A communication system, no matter how efficiently coded, must also be possible for humans to learn and to use. In this dissertation, I adopt an information theoretic approach to elucidating the challenges posed by the construction and maintenance of a workable lexicon, with the express aim of bridging the gap between rational and mechanistic accounts. To this end, I detail a series of large-scale corpus analyses and behavioral experiments that investigate (1) how languages (and linguistic sub-systems) solve the problem of assigning names to things, (2) how principles of learning and memory act to constrain the possible solution space, and (3) how different solutions present trade-offs for learnability and efficient processing. These investigations shed light on a few of the methods natural languages use to solve a complex constraint-satisfaction problem. [The dissertation citations contained here are published with the permission of ProQuest LLC. Further reproduction is prohibited without permission. Copies of dissertations may be obtained by Telephone (800) 1-800-521-0600. Web page: http://www.proquest.com/en-US/products/dissertations/individuals.shtml.] (As Provided).
Anmerkungen	ProQuest LLC. 789 East Eisenhower Parkway, P.O. Box 1346, Ann Arbor, MI 48106. Tel: 800-521-0600; Web site: http://www.proquest.com/en-US/products/dissertations/individuals.shtml
Erfasst von	ERIC (Education Resources Information Center), Washington, DC
Update	2020/1/01