The effect of concrete wording on truth judgements: A preregistered replication and extension of Hansen & Wänke (2010)

Abstract

When you lack the facts, how do you decide what is true and what is not? In the absence of knowledge, we sometimes rely on non-probative information. For example, participants judge concretely worded trivia items as more likely to be true than abstractly worded ones (the linguistic truth effect; Hansen & Wänke, 2010). If minor language differences affect truth judgements, ultimately they could influence more consequential political, legal, health, and interpersonal choices. This Registered Report includes two high-powered replication attempts of Experiment 1 from Hansen and Wänke (2010). Experiment 1a was a dual-site, in-person replication of the linguistic concreteness effect in the original paper-and-pencil format (n = 253, n = 246 in analyses). Experiment 1b replicated the study with an online sample (n = 237, n = 220 in analyses). In Experiment 1a, the effect of concreteness on judgements of truth (Cohen’s dz = 0.08; 95% CI: [–0.03, 0.18]) was smaller than that of the original study. Similarly, in Experiment 1b the effect (Cohen’s dz = 0.11; 95% CI [–0.01, 0.22]) was smaller than that of the original study. Collectively, the pattern of results is inconsistent with that of the original study.