Masterton, Sarah
(2022)
Cognitive Bias Modification for Unhealthy Food Behaviours.
PhD thesis, University of Liverpool.
|
Text
200763862_Aug2022.pdf - Author Accepted Manuscript Download (5MB) | Preview |
Abstract
Cognitive Bias Modification (CBM) paradigms are hypothesised to reduce unhealthy food behaviours (and potentially weight) through the completion of non-invasive computerised tasks. Despite their potential, inconsistent task designs and study outcomes across the literature raise questions in relation to true training efficacy. The overall aim of this thesis was to investigate two specific CBM paradigms (cue-inhibitory control training (cue-ICT) and evaluative conditioning (EC)) to evaluate their impact in terms of behavioural change and identify mechanisms of effect while also addressing limitations of the current literature base. Factors associated with training outcomes (including cue-inhibition contingencies, belief in training and proxy measures of change) were also investigated to attempt to explain inconsistent training outcomes and accurately evaluate the efficacy of CBM training paradigms. The results from chapter 3 demonstrated that neither cue-ICT or EC administered in the lab had a significant influence on ad-libitum food consumption or implicit food preference. While there was a significant difference in explicit preferences between conditions, this was between the active cue-ICT and EC training groups rather than the active or passive control groups. The studies contained within chapter 4 demonstrated that systematically varying the cue-inhibition contingencies (cue-ICT) and critical pairing percentages (EC) experienced during online training had no influence on training outcomes, with no significant differences found at any task percentage for unhealthy food value for either cue-ICT and EC. There was some evidence to suggest EC may influence explicit choice, with healthier choices made in the 100% unhealthy food-negative image group compared to the control (50%) group. As these studies provided limited evidence to support the use of CBM in food contexts, chapter 5 examined the role of individual level variables (belief) in training outcomes across two online studies. The results revealed that active CBM only appeared to be effective at reducing unhealthy food value when a manipulation message describing the CBM technique (either cue-ICT or EC) in a positive way was presented prior to training completion. For EC these effects were still evident one week after training. The final experimental chapter (chapter 6) used an EMA design to examine the extent to which commonly used measures of food preference and value were related to real world food consumption. While these measures are commonly used as indicators of CBM training effectiveness, the results provided limited evidence to suggest that these measures are related to real world food behaviours, with only unhealthy food value predicting consumption over the study period. Overall, the results of this thesis provide limited evidence to support the use of CBM as a standalone intervention strategy for unhealthy food behaviours, with factors external to training (i.e., belief in training effectiveness) appearing to have a substantial influence on training efficacy. Future research is needed to further identify the role of individual differences within CBM contexts and validate alternative measures of food preference and value that can accurately predict real world food behaviours.
| Item Type: | Thesis (PhD) |
|---|---|
| Divisions: | Faculty of Health and Life Sciences > Institute of Population Health |
| Depositing User: | Symplectic Admin |
| Date Deposited: | 10 Nov 2022 15:20 |
| Last Modified: | 18 Jan 2023 20:36 |
| DOI: | 10.17638/03165430 |
| Supervisors: |
|
| URI: | https://livrepository.liverpool.ac.uk/id/eprint/3165430 |
Dimensions
Dimensions
