The Library
Adaptive rationality in communication
Tools
Pilgrim, Charlie (2023) Adaptive rationality in communication. PhD thesis, University of Warwick.
|
PDF
WRAP_Theses_Pilgrim_2023.pdf - Submitted Version - Requires a PDF viewer. Download (6Mb) | Preview |
Official URL: http://webcat.warwick.ac.uk/record=b3985043
Abstract
Modern telecommunications have transformed the way that people communicate. The situation is dynamic, with a rapidly changing technological and cultural landscape. Furthermore, interactions between this landscape and human behaviour are complex and difficult to predict. The work in this thesis is inspired by the general problem of describing these systems.
We begin with an investigation into historical trends in language, finding that the word entropy of American English has increased steadily since around 1900. We also find differences in word entropy across media categories. These changes are explored in the context of the attention economy, which is the dynamic of increasing competition for human attention in response to a rising abundance of information. A model of information foraging in the attention economy is developed to describe the trends in word entropy.
Word entropy is a property of word distributions, which follow Zipf’s law: a power law relationship between the frequency of words and their rank in that frequency distribution. As well as word entropy, we also see trends in changes in Zipf’s law over the 20th century. There is difficulty in investigating these trends due to bias in estimators of the Zipf exponent. The source of this bias is explored and shown to be due to inappropriate assumptions in the estimators. The correct estimator is derived but found to be computationally intractable.
Modern advances in information search and social media have been implicated in the creation of separated silos of thought in society and the driving of dangerous political polarisation. This is connected to confirmation bias, which is the tendency for people to search for and consume information in a way that maintains their existing beliefs. A model of confirmation bias is developed that is based on a boundedly rational model of belief updating, which takes into account cognitive limitations.
Item Type: | Thesis (PhD) | ||||
---|---|---|---|---|---|
Subjects: | B Philosophy. Psychology. Religion > BF Psychology P Language and Literature > P Philology. Linguistics Q Science > Q Science (General) Q Science > QA Mathematics Z Bibliography. Library Science. Information Resources > ZA Information resources |
||||
Library of Congress Subject Headings (LCSH): | Entropy (Information theory), Communication, Human information processing, Zipf's law, Language and languages -- Word frequency, Filter bubbles (Information filtering), Belief and doubt, Bayesian statistical decision theory | ||||
Official Date: | June 2023 | ||||
Dates: |
|
||||
Institution: | University of Warwick | ||||
Theses Department: | Mathematics Institute | ||||
Thesis Type: | PhD | ||||
Publication Status: | Unpublished | ||||
Supervisor(s)/Advisor: | Hills, Thomas | ||||
Format of File: | |||||
Extent: | xvii, 127 pages : illustrations | ||||
Language: | eng |
Request changes or add full text files to a record
Repository staff actions (login required)
View Item |