Recently, Diemand-Yauman et al. "Cognition," 118, 114-118 (2011) demonstrated that learning with disfluent (hard-to-read) materials is more effective than learning with easy-to-read materials--a study that has since stipulated a number of follow-up studies (with mixed results). However, there is a potential confound in the original experiments: The disfluent materials were not only disfluent but also rather unusual. Therefore, they might have been particularly distinctive and have attracted more attention, which then resulted in better learning. We conducted three experiments to address this confound, all of them slightly modified replications of Diemand-Yauman et al.'s Experiment 1. Participants received five lists, either at their own pace on a computer screen (Experiment 1) or experimenter-paced on paper (Experiments 2 & 3). In Experiments 1 and 2, participants either received one fluent and four disfluent lists or they received four fluent lists and one disfluent list. The position of the distinct list varied across participants. In Experiment 3, the distinct list was always the penultimate one. In none of the experiments, learning performance was affected by any of the experimental manipulations. Our results question the generality of the disfluency effect with respect to learning.