More is easier? Testing the role of fluency in the more-credible effect

Abstract

Abstract People are more likely to endorse statements of the form "A is more than B" than those of the form "B is less than A", even though the ordinal relationship being described is identical in both cases -– a result I dub the "more-credible" effect. This paper reports 9 experiments (totalN= 5643) that probe the generality and basis for this effect. Studies 1–4 replicate the effect for comparative statements relating to environmental change and sustainable behaviours, finding that it is robust to changes in participant population, experimental design, response formats and data analysis strategy. However, it does not generalize to all stimulus sets. Studies 5–9 test the proposition that the effect is based on the greater ease of processing "more than" statements. I find no meaningful effect of warning people not to base their judgments on the fluency of the sentences (Studies 5 and 6), but do find associations between comparative language, credibility, and processing time: when the more-credible effect manifests, the more-than statements are read more quickly than the less-than statements, and this difference partly mediates the effect of comparative on agreement with the statements; in contrast, for a set of comparisons for which changes in the more/less framing did not affect truth judgments, there was no meaningful difference in the time taken to read the more- and less-than versions of the statements. Taken together, these results highlight the importance of comparative language in shaping the credibility of important socio-political messages, and provide some limited support for the idea that the effect of language choice is partly due to differences in how easily the statements can be processed -– although other mechanisms are also likely to be at work.

Cite this work

No public copy is stored for this record, so only the metadata is cited.

Stored only when both papers are in the corpus — 8 citations exist upstream.

Papers that cite the same references — the relational fallback while embeddings are still backfilling.