Doctor of Philosophy, Case Western Reserve University, 2017, Psychology
Claims are more likely to be judged true when accompanied by information that, though related to the subject of the claim, is actually nonprobative. This finding, coined the truthiness effect, is thought to occur because the embellished claims are processed more fluently, and therefore seem more accurate, than the unembellished claims. Five experiments are reported that explore the explanations for this effect. After an initial replication using nonprobative images, two experiments diminish truthiness via manipulations designed to mitigate the use of fluency as a cue for determining truth value of the claims: first by slowing presentation rate of the stimuli and then by warning participants of the nonprobative nature of the accompanying images. The concluding two experiments test alternative accounts of the truthiness effect. In Experiment 4, truthiness was not observed when claims were presented in a forced-choice format, supporting the hypothesis that a shift in response criteria underlies the effect in the standard true-false format. Experiment 5 was conducted to determine if truthiness is better conceptualized as increased positive responding to embellished claims or diminished positive responding to unembellished claims. A full design was used in which presence of nonprobative information was manipulated between subjects for some participants and within subjects for others. Though results were equivocal, available evidence best aligns with an inhibition account of truthiness, in which participants adopt a more conservative criterion for statements without nonprobative information when other statements have such information.
Committee: Robert Greene PhD (Committee Chair); Elizabeth Short PhD (Committee Member); Brooke Macnamara PhD (Committee Member); Fey Parrill PhD (Committee Member)
Subjects: Cognitive Psychology; Experimental Psychology; Psychology