Studying the behavioural, physiological, and neural indices of associative learning in multi-trial paradigms: methodological and analytical considerations

Jambazova, Antonia Antony (2021) Studying the behavioural, physiological, and neural indices of associative learning in multi-trial paradigms: methodological and analytical considerations. PhD thesis, University of Glasgow.

Full text available as:
[thumbnail of 2021JambazovaPhD.pdf] PDF
Download (18MB)

Abstract

Since the pioneering work of Ivan Pavlov nearly a century ago, the empirical study of associative learning through classical conditioning has continued to grow. However, the high volume of classical conditioning investigations has resulted in an equal in magnitude methodological and analytical variation, which can often challenge cross-study comparisons, replicability and generalisability of findings (Haaker et al., 2019; Lonsdorf et al., 2017). Consequently, the field of conditioning has begun to focus on reducing excessive flexibility in data practices through increasing methodological rigour, consistency, and transparency. So far, research has concentrated on improving methods in areas such as the quantification of conditioned responding, analytical strategies, translational research and individual differences (Bach et al., 2018; Haaker et al., 2019; Korn et al., 2017; Lonsdorf et al., 2019; Lonsdorf & Merz, 2017; Ney et al., 2018; Sjouwerman & Lonsdorf, 2019). The aim of this thesis was to provide an additional contribution to recent methodological efforts in the field by focusing on an area that has not received as much empirical attention. Specifically, we discuss and examine the potential utility of multi-trial conditioning for studying psychophysiological indices of learning. In addition, throughout this thesis, we aimed to reinforce the value of transparent and robust data practises in aiding replicability and generalisability of conditioning research.

In Chapter 2, we report findings from an indirect behavioural replication of an established multi-trial task (i.e., Multi-CS Conditioning, Steinberg et al., 2013), accompanied by a discussion about the role of contingency awareness in conditioning. We also provide a re-analysis of a previous Multi-CS dataset (Rehbein et al., 2014) to highlight the value of robust and transparent data visualisation in guiding analytical decisions, and to illustrate how poor consideration of individual differences and underlying data distributions may explain the inconsistency in previous research using this task. Chapter 3 reports a novel visual blocked conditioning paradigm that delivers a high number of trials through attempting to elicit associative learning in multiple successive blocks. We investigated the potential utility of this task to overcome some of the technical and design challenges (e.g., detecting deep source activity, time-frequency analysis) present in magnetoencephalography (MEG) research, studying the cortical and subcortical oscillatory dynamics of learning and extinction. The findings from this study suggested that the task does not reliably elicit conditioning in any of the outcome measures that we considered (MEG, pupil size, valence, and arousal ratings). Nevertheless, the reported results identified several design modifications that can aid future paradigm development. These were related to aspects such as trial duration, the type of CSs employed, and maintaining attention and contingency awareness. Chapter 4 reports findings from an auditory blocked conditioning task, modified based on the results from Chapter 3. The task was examined in the context of pupillary and subjective behavioural indices of conditioning, with a discussion of its application in future MEG designs. In addition, the study considers the potential of this multi-trial paradigm to offer better generalisability of findings when used in combination with robust analytical strategies (i.e., data-driven time window selection and design-appropriate mixed modelling). Finally, Chapter 5 discusses the implications of the findings reported in this thesis for future multi-trial conditioning research.

Item Type: Thesis (PhD)
Qualification Level: Doctoral
Subjects: B Philosophy. Psychology. Religion > BF Psychology
Colleges/Schools: College of Science and Engineering > School of Psychology
Supervisor's Name: Scheepers, Dr. Christoph and van Rijsbergen, Dr. Nicola
Date of Award: 2021
Depositing User: Theses Team
Unique ID: glathesis:2021-82335
Copyright: Copyright of this thesis is held by the author.
Date Deposited: 22 Jul 2021 13:48
Last Modified: 22 Jul 2021 15:56
Thesis DOI: 10.5525/gla.thesis.82335
URI: https://theses.gla.ac.uk/id/eprint/82335
Related URLs:

Actions (login required)

View Item View Item

Downloads

Downloads per month over past year