Monica Bucciarelli and P.N. Johnson-LairdActa Psychologica
Volume 198, July 2019
Abstract
A dual-process theory postulates that belief and emotions about moral assertions can affect one another. The present study corroborated this prediction. Experiments 1, 2 and 3 showed that the pleasantness of a moral assertion – from loathing it to loving it – correlated with how strongly individuals believed it, i.e., its subjective probability. But, despite repeated testing, this relation did not occur for factual assertions. To create the correlation, it sufficed to change factual assertions, such as, “Advanced countries are democracies,” into moral assertions, “Advanced countries should be democracies”. Two further experiments corroborated the two-way causal relations for moral assertions. Experiment 4 showed that recall of pleasant memories about moral assertions increased their believability, and that the recall of unpleasant memories had the opposite effect. Experiment 5 showed that the creation of reasons to believe moral assertions increased the pleasantness of the emotions they evoked, and that the creation of reasons to disbelieve moral assertions had the opposite effect. Hence, emotions can change beliefs about moral assertions; and reasons can change emotions about moral assertions. We discuss the implications of these results for alternative theories of morality.
The research is here.
Here is a portion of the Discussion:
In sum, emotions and beliefs correlate for moral assertions, and a change in one can cause a change in the other. The main theoretical problem is to explain these results. They should hardly surprise Utilitarians. As we mentioned in the Introduction, one interpretation of their views (Jon Baron, p.c.) is that it is tautological to predict that if you believe a moral assertion then you will like it. And this interpretation implies that our experiments are studies in semantics, which corroborate the existence of tautologies depending on the meanings of words (contra to Quine, 1953; cf. Quelhas, Rasga, & Johnson-Laird, 2017). But, the degrees to which participants believed the moral assertions varied from certain to impossible. An assertion that they rated as probable as not is hardly a tautology, and it tended to occur with an emotional reaction of indifference. The hypothesis of a tautological interpretation cannot explain this aspect of an overall correlation in ratings on scales.
Here is a portion of the Discussion:
In sum, emotions and beliefs correlate for moral assertions, and a change in one can cause a change in the other. The main theoretical problem is to explain these results. They should hardly surprise Utilitarians. As we mentioned in the Introduction, one interpretation of their views (Jon Baron, p.c.) is that it is tautological to predict that if you believe a moral assertion then you will like it. And this interpretation implies that our experiments are studies in semantics, which corroborate the existence of tautologies depending on the meanings of words (contra to Quine, 1953; cf. Quelhas, Rasga, & Johnson-Laird, 2017). But, the degrees to which participants believed the moral assertions varied from certain to impossible. An assertion that they rated as probable as not is hardly a tautology, and it tended to occur with an emotional reaction of indifference. The hypothesis of a tautological interpretation cannot explain this aspect of an overall correlation in ratings on scales.
