Abstract
This thesis represents a body of work to validate the Moral Foundations Conflict Task (MFCT), a novel measure that aims to probe directly at intuitive conflict between moral foundations. Moral Foundations Theory (MFT) explains variation in moral judgments on the basis of innate and intuitive foundations, representing distinct moral concerns. However, little work has explored how foundations might compete within individuals. Prior research has tended to rely on an explicit self-report measure – the Moral Foundations Questionnaire (MFQ). In contrast, the MFCT requires quick, intuitive choices between foundations, tracking endorsement based on how often they are chosen, and how long it takes to choose them (response times – RTs).
Across five studies, we test whether responses on the MFCT: reflect explicitly endorsed moral values measured by the MFQ (Study 1); are altered under cognitive load (Study 2 and 3); and under deliberation (Study 4); and are predicted by established correlates of the MFQ (Study 5). Endorsements on the MFCT reliably correlate with those on the MFQ. Increased load does not affect this correlation, enhancing confidence that the MFCT is an effective measure of moral intuitions. Increasing deliberation over choices in the MFCT also does not seem to alter this correlation, though this finding is subject to limitations to Study 4. Furthermore, the MFCT performs better than the MFQ in models with political orientation, authoritarianism and social dominance, indicating that the MFCT captures unique variance that the MFQ does not. Across all studies, we employ exploratory analyses of RTs, applying Ex-Gaussian decomposition alongside analyses of mean RT interpretation of the MFCT as a direct measure of inter-foundation conflict. An Ex-Gaussian approach models RTs as a combination of pure decision processes and conflict resolution, where the τ parameter corresponds to the latter. Generally, we find that mean RT and τ decrease as the value, and the difference in value, of foundations in a choice increases.
In conclusion, this work provides a significant theoretical and methodological contribution, allowing future research to explore inter-foundation conflict. It demonstrates that the MFCT captures systematic variance in foundation endorsement, and can therefore contribute to understanding of moral value conflicts and their consequences.
Across five studies, we test whether responses on the MFCT: reflect explicitly endorsed moral values measured by the MFQ (Study 1); are altered under cognitive load (Study 2 and 3); and under deliberation (Study 4); and are predicted by established correlates of the MFQ (Study 5). Endorsements on the MFCT reliably correlate with those on the MFQ. Increased load does not affect this correlation, enhancing confidence that the MFCT is an effective measure of moral intuitions. Increasing deliberation over choices in the MFCT also does not seem to alter this correlation, though this finding is subject to limitations to Study 4. Furthermore, the MFCT performs better than the MFQ in models with political orientation, authoritarianism and social dominance, indicating that the MFCT captures unique variance that the MFQ does not. Across all studies, we employ exploratory analyses of RTs, applying Ex-Gaussian decomposition alongside analyses of mean RT interpretation of the MFCT as a direct measure of inter-foundation conflict. An Ex-Gaussian approach models RTs as a combination of pure decision processes and conflict resolution, where the τ parameter corresponds to the latter. Generally, we find that mean RT and τ decrease as the value, and the difference in value, of foundations in a choice increases.
In conclusion, this work provides a significant theoretical and methodological contribution, allowing future research to explore inter-foundation conflict. It demonstrates that the MFCT captures systematic variance in foundation endorsement, and can therefore contribute to understanding of moral value conflicts and their consequences.
Original language | English |
---|---|
Type | PhD Thesis |
Publisher | University of Edinburgh |
Number of pages | 277 |
DOIs | |
Publication status | Published - 30 Nov 2020 |