Beauty and Truth, Truth and Beauty: Chiastic Structure Increases the Subjective Accuracy of Statements

The Keats heuristic suggests that people find esthetically pleasing expressions more accurate than mundane expressions. We test this notion with chiastic statements. Chiasmus is a stylistic phenomenon in which at least two linguistic constituents are repeated in reverse order, conventionally represented by the formula A-B-B-A. Our study focuses on the specific form of chiasmus known as antimetabole, in which the reverse-repeated constituents are words (e.g., All for one and one for all; A = all, B = one). In three out of four experiments (N = 797), we find evidence that people judge antimetabolic statements (e.g., Success is getting what you want. Happiness is wanting what you get.) as more accurate than semantically equivalent nonantimetabolic statements (e.g., Success is getting what you wish. Happiness is wanting what you receive.). Furthermore, we evaluate fluency as a potential mechanism explaining the observed accuracy benefit afforded to antimetabolic statements, finding that the increased speed (i.e., fluency) with which antimetabolic statements were processed predicted judgments of accuracy. Overall, the present work is consistent with the growing literature on stylistic factors biasing assessments of truth, using the distinctive stylistic pattern of antimetabole. We find that information communicated using an antimetabolic structure is judged to be more accurate than nonantimetabolic paraphrases. Public Significance Statement The rhetorical figure called antimetabole repeats words in the reverse order (e.g., nice and important in It's nice to be important, but it's more important to be nice). We found that antimetabolic statements evoke stronger feelings of truthfulness compared to the same meanings expressed without reverse repetition (e.g., It's nice to be important, but it's more critical to be kind). We argue that the increased speed with which antimetabolic statements are judged partially explains the accuracy benefit that they confer. Our findings suggest that stylistic factors-in this case, antimetabole-bias evaluations of truth.