Measuring Belief Bias with Ternary Response Sets

Belief bias in syllogistic reasoning refers to the finding that individuals are more likely to accept believable than unbelievable conclusions independent of their logical validity. Most theories argue that belief bias is driven by differences in reasoning processes between believable and unbelievable syllogisms. In contrast, Dube, Rotello, and Heit (2010) proposed that belief bias is solely an effect of response processes. We investigated belief bias without having to rely on response bias manipulations (Klauer, Musch, and Naumer, 2000) or confidence ratings (Dube et al., 2010). Instead, we added a third response (“I don't know”) to the usual binary response set (“Yes”/“No”). This allowed us to test belief bias with a fully identified multinomial processing tree model, in a hierarchical Bayesian framework. We found evidence that the belief bias is driven by differences in response processes. Evidence for a difference in reasoning processes was inconclusive.