Psychology ETDs

Publication Date

Spring 5-11-2019

Abstract

In order to successfully traverse an actively complex environment, an agent is required to learn from the consequences of their actions. For over a century, models of behavior have been developed demonstrating these consequence-based learning systems. More recently, underlying biological systems have been found to adhere to these constructs of learning. The electroencephalographic signal known as the Reward Positivity (RewP) is thought to reflect a dopamine-dependent cortical signal specific to reward receipt. Importantly, this signal has been shown to adhere to an axiomatic (rule-like) positive reward prediction error, whereby it is evoked following outcomes that are better than expected. These features of the RewP make it a candidate marker for clinical populations, such as major depressive disorder, substance use disorder, and Parkinson’s disease. Although recent experimental endeavors have highlighted key characteristics of the generation and modulation of the RewP, a major understudied feature of the RewP in humans is the link between hedonic experiences and reward processes, and how these interact to modulate learning. This dissertation aims to probe this overlooked hedonic aspect of RewP generation through the use of emotionally evocative image rewards. The first aim addresses methodological issues relating to the use of complex, ecologically valid stimuli in EEG experimentation. The second aim investigated techniques for rectifying these methodological issues. Lastly, the third aim investigated the use of emotionally salient images as rewards in a reinforcement learning paradigm.

Degree Name

Psychology

Level of Degree

Doctoral

Department Name

Psychology

First Committee Member (Chair)

James F. Cavanagh

Second Committee Member

Kent A. Kiehl

Third Committee Member

Jeremy Hogeveen

Fourth Committee Member

Philip A. Gable

Language

English

Keywords

EEG, Emotion, Gender Difference, Reinforcement Learning, Reward Processing, Time Frequency Analysis

Document Type

Dissertation

Share

COinS