Challenging the notion of a general disfluency effect: the moderating effect of element-interactivity on perceptual disfluency

Introduction The disfluency effect proposes that deliberately introducing challenges or difficulties into the learning process can be advantageous. Particularly, perceptual disfluency (e.g., harder to read fonts) might affect learning positively. However, an ongoing debate persists regardxing the robustness of this effect, as some studies have failed to replicate it or have uncovered opposing outcomes. Methods To investigate potential moderators of the disfluency effect, two experiments were conducted using different types of instructional materials (instructional texts: N 1 = 76; concept maps: N 2 = 74). In both experiments, the fluency was manipulated by using either a legible font or an illegible font, while element interactivity (high vs. low) was manipulated as a moderator. Learning outcomes, cognitive load, accuracy of metacognitive judgments, learning time, and efficiency were assessed in both experiments. Results Results indicated that disfluency did not have a general impact on the dependent variables, except for a detrimental effect on extraneous load. Notably, disfluency increased learning outcomes and germane load for low element interactivity. Contrary to common explanations of the disfluency effect, the use of a disfluent font did not yield metacognitive benefits.