Rothenhoefer, Kathryn M.
(2023)
The Effects of Uncertainty on Behavior: Reward Distributions Modulate Dopamine Learning Signals and Decision Making.
Doctoral Dissertation, University of Pittsburgh.
(Unpublished)
Abstract
To make optimal decisions and get the best outcomes, human and animal decision-makers must traverse an uncertain world. Uncertainty is divided into two separate types: risk and ambiguity. Risk describes uncertainty where the outcomes and their probabilities are known, such as a fair coin flip. Ambiguity is uncertainty where there is incomplete information about the possible outcomes and their probabilities of occurring. Decision-makers often avoid choices with ambiguity, even if they are objectively better than the alternatives. Our ability to make the best decisions given the uncertainty of our options is dependent on learning from our past decisions and updating our expectations accordingly. Learning which behaviors to repeat, and which to discontinue is dependent on learning signals from midbrain dopamine neurons. Dopamine neurons compute reward prediction errors and transmit this signal to various brain regions, but mostly to the striatum, a critical mediator of reward learning. Reward predictions errors code the difference between the reward that was actually received and the reward that was expected. Expected value estimates in dopamine neuron signals incorporate the probability and size of rewards. However, it is not known whether dopamine neurons code uncertainty independent from expected value. The results of this body of work are threefold. First, we determined that higher uncertainty, when independent from the expected value of the cue, decreased learning and autonomic responses, and minimized the absolute magnitude of dopamine neuron reward prediction error responses. Second, we determined that animals differentiate between ambiguous and risky choice options that have the same expected value. Choice behavior illustrated that ambiguity preferences vary based on expected value. This result is the first to demonstrate that decision-makers have value-dependent ambiguity preferences, like they do risk preferences. Finally, we characterized the functional and anatomical diversity of neurons in the striatum, specifically medium spiny neurons, which are a key relay station between dopamine neurons and the cortico-basal ganglia-thalamo-cortical circuits that mediate sophisticated behavior. These results together provide clear direction for future research into the effect of uncertainty on behavior and neural coding, and new opportunities for investigating circuit-specific functions that mediate these behaviors.
Share
| Citation/Export: |
|
| Social Networking: |
|
Details
| Item Type: |
University of Pittsburgh ETD
|
| Status: |
Unpublished |
| Creators/Authors: |
|
| ETD Committee: |
|
| Date: |
15 September 2023 |
| Date Type: |
Publication |
| Defense Date: |
28 November 2022 |
| Approval Date: |
15 September 2023 |
| Submission Date: |
23 December 2022 |
| Access Restriction: |
No restriction; Release the ETD for access worldwide immediately. |
| Number of Pages: |
187 |
| Institution: |
University of Pittsburgh |
| Schools and Programs: |
School of Medicine > Neurobiology |
| Degree: |
PhD - Doctor of Philosophy |
| Thesis Type: |
Doctoral Dissertation |
| Refereed: |
Yes |
| Uncontrolled Keywords: |
Dopamine; Reward Prediction Error; Uncertainty; Risk; Ambiguity; Learning; Decision Making; Medium Spiny Neurons; Striatum |
| Related URLs: |
|
| Date Deposited: |
15 Sep 2023 14:05 |
| Last Modified: |
15 Sep 2023 14:05 |
| URI: |
http://d-scholarship.pitt.edu/id/eprint/44064 |
Metrics
Monthly Views for the past 3 years
Plum Analytics
Actions (login required)
 |
View Item |
![[feed]](/images/feed-icon-32x32.png){kind=icon}