Multinomial Models of the Repetition-Based Truth Effect: Investigating the Role of Prior Knowledge

The repetition-based truth effect refers to the phenomenon that repeated statements are more likely to be judged as true than new statements. Fazio et al. (2015) developed two multinomial processing tree (MPT) models to account for truth judgments. The knowledge-conditional model assumes that repetition leads to a shift in response bias conditional on the absence of knowledge. In contrast, the fluency-conditional model assumes that knowledge is used only when not relying on processing fluency, which results in reduced discrimination performance. We study the formal properties of the competing models using receiver operating characteristic (ROC) curves and highlight important auxiliary assumptions and identifiability constraints. In three experiments, we extended the classic truth-effect paradigm to validate and test different model versions by manipulating the base rate of true statements in the judgment phase. The results support the notion that repetition results in reduced discrimination performance. However, the alternative model conceptualizing the truth effect as a response bias cannot be rejected when assuming different knowledge for true and false statements.