On the Robustness of the Modality ­Effect: Attempting to Replicate a Basic Finding

Abstract

Abstract.Two experiments were conducted to replicate the modality effect and to test two opposing theoretical explanations: From the visuo-spatial load hypothesis (as implied by the Cognitive Theory of Multimedia Learning), we derived the prediction that the modality effect depends on the need to simultaneously store and process text and pictorial information in the same working memory subsystem, which implies that the modality effect is not influenced by text length. However, based on the concurrent auditory recency hypothesis, we expected the modality effect to occur with very short texts only, irrespective of whether a picture is present. Adopting a German translation of the learning materials used by Moreno and Mayer (1999) , we failed to detect a modality effect for sequential text-picture-presentation in Experiment 1. In Experiment 2, a modality effect was expected for simultaneous presentation of the same texts and pictures, yet again remained absent. We conclude with the suggestion that the apparent robustness of the modality effect reported in the literature might be explained – at least in part – by publication bias.