Many states have passed laws mandating pre- and in-service teacher training in dyslexia; however, research suggests the majority of stakeholders continue to hold misconceptions about the definition, causes, identification, treatment, and how dyslexia relates to school-based factors. In this study, three groups of preservice teachers were assessed on their knowledge of dyslexia to determine baseline levels of scientific conceptions (SCs), misconceptions (MCs) and uncertainties (UCs). Next, participants were randomly assigned one of three texts explaining dyslexia (informational [IDA text], refutation text [RT], refutation text with graphics embedded [RT-EG]) to determine how conditions impacted their SCs, MCs, and UCs. After reading, participants were assessed on their knowledge of dyslexia, cognitive engagement during reading, and demographic data was collected. Results indicate significant levels of MCs among preservice teachers regarding (a) the type of treatment needed for students with dyslexia and (b) the MC of dyslexia as a visual disability needing visually-based treatments. After readings, all groups improved in their SCs of dyslexia; however, the RT and RT-EG groups increased SCs more than the IDA text. Only the RT and RT-EG texts decreased MCs statistically significantly with large effect sizes (RT p < .001, ηp2 = .50; RT-EG p < .001, ηp2 = .32). All texts decreased UCs; however, the RT and RT-EG conditions produced large effects. Implications for training and future research are discussed. Using refutation-based materials in dyslexia trainings has the potential to increase participant understanding of dyslexia more than informational trainings that do not refute common MCs.