Debunking Misinformation About a Causal Link Between Vaccines and Autism: Two Preregistered Tests of Dual-Process Versus Single-Process Predictions (With Conflicting Results)

Abstract

Dual-process and single-process theories lead to conflicting predictions about whether debunking messages negating a state of affairs should change responses on implicit measures in a manner intended by the message. Two preregistered studies (N1 = 550; N2 = 880) tested these predictions using official health information from the U.S. Centers for Disease Control and Prevention debunking the idea that vaccines would cause autism. Consistent with predictions derived from dual-process learning theories, Experiment 1 found that debunking-via-negation increased responses linking vaccines to autism on implicit measures, although it effectively reduced self-reported judgments linking vaccines to autism on explicit measures. Using the same measures and materials, Experiment 2 found that debunking-via-negation effectively reduced responses linking vaccines to autism on both implicit and explicit measures, consistent with predictions derived from single-process propositional theories. Potential reasons for the conflicting outcomes are discussed, including their implications for the debate between dual-process and single-process theories.