Fleig-Goldstein, Brendan
(2024)
Philosophical Foundations of Resource Rational Analysis.
Doctoral Dissertation, University of Pittsburgh.
(Unpublished)
This is the latest version of this item.
Abstract
Tacit appeals to systems being rational or apparently irrational are common in cognitive science, and for good reason: irrationality provides valuable evidence for cognitive models. A methodological approach in cognitive science called resource rational analysis attempts to systematize the use of irrationality to develop and test models of cognition. It does so by initially assuming that a system is rational, and then iteratively de-idealizing this assumption by identifying psychological facts that prevent a system from being more rational. This dissertation seeks to analyze how this strategy has worked, how it should work, why it will work, and why it can work better with the conceptual foundations proposed here. In Chapter 1, I develop a specific account of resource rationality. I argue that all epistemic norms are relative to cognitive constraints, and that there is no principled way to distinguish between agents doing their best relative to their limitations and agents being irrational. I advocate for a maximally broad view of cognitive constraints, which renders all agents trivially resource rational, but still allows for meaningful evaluation and prescription. Chapter 2 reviews arguments that intentionality presupposes rationality, and argues that this position is strengthened if the appropriate notion of rationality is understood as my notion of resource rationality. This conclusion shows why rationality considerations are important and even necessary for any intentional psychological science. In Chapter 3, I extend my account of resource rationality to normative commitments, proposing that what I call a meta-reflective capacity—maintaining resource rationality under varying conditions—is necessary and sufficient for possessing normative commitments. This perspective offers a framework for endowing AI systems with normative commitments and empirically investigating these commitments in humans and non-human animals. Chapter 4 presents resource rational analysis as a methodological strategy in cognitive science and argues for its effectiveness. This strategy, I show, implements a dynamic theory-testing method known as ``Closing-the-Loop," as described by Smith (2014). I use the Material Theory of Induction and Topological Learning Theory to provide an epistemic justification for this dynamic testing strategy. These considerations support the iterative de-idealization process and demonstrate the utility of rationality considerations in cognitive science.
Share
| Citation/Export: |
|
| Social Networking: |
|
Details
| Item Type: |
University of Pittsburgh ETD
|
| Status: |
Unpublished |
| Creators/Authors: |
|
| ETD Committee: |
|
| Date: |
27 August 2024 |
| Date Type: |
Publication |
| Defense Date: |
24 June 2024 |
| Approval Date: |
27 August 2024 |
| Submission Date: |
25 June 2024 |
| Access Restriction: |
No restriction; Release the ETD for access worldwide immediately. |
| Number of Pages: |
193 |
| Institution: |
University of Pittsburgh |
| Schools and Programs: |
Dietrich School of Arts and Sciences > History and Philosophy of Science |
| Degree: |
PhD - Doctor of Philosophy |
| Thesis Type: |
Doctoral Dissertation |
| Refereed: |
Yes |
| Uncontrolled Keywords: |
rationality, cognitive science, scientific evidence, resource rationality, bounded rationality |
| Date Deposited: |
27 Aug 2024 13:44 |
| Last Modified: |
27 Aug 2024 13:44 |
| URI: |
http://d-scholarship.pitt.edu/id/eprint/46632 |
Available Versions of this Item
-
Philosophical Foundations of Resource Rational Analysis. (deposited 27 Aug 2024 13:44)
[Currently Displayed]
Metrics
Monthly Views for the past 3 years
Plum Analytics
Actions (login required)
 |
View Item |
![[feed]](/images/feed-icon-32x32.png){kind=icon}