No abstract

The sequence in which problems of different concepts are studied during instruction impacts concept learning. For example, several problems of a given concept can be studied together (blocking) or several problems of different concepts can be studied together (interleaving). In the current study, we demonstrate that the 2 sequences impact concept induction differently as they differ in the temporal spacing and the temporal juxtaposition of to-be-learned concept problems, and in the cognitive processes they recruit. Participants studied 6 problems of 3 different statistical concepts, and then were tested on their ability to correctly classify new problems on a final test. Interleaving problems of different to-be-learned concepts, rather than blocking problems by concept, enhanced classification performance, replicating the interleaving effect (Experiment 1). Introducing temporal spacing between successive problems decreased classification performance in the interleaved schedule-consistent with the discriminative-contrast hypothesis that interleaving fosters between-concept comparisons-and increased classification performance in the blocked schedule-consistent with the study-phase retrieval hypothesis that temporal spacing causes forgetting and subsequent retrieval enhances memory (Experiment 2). Temporally juxtaposing problems of concepts 3-at-a-time rather than 1-at-a-time improved overall classification performance, particularly in a blocked schedule-consistent with the commonality-abstraction hypothesis that blocking fosters within-concept comparisons (Experiment 3). All participants also completed a working memory capacity (WMC) task, findings of which suggest that the efficacy of the above study sequences may be related to individual differences in WMC.

Although online lectures have become increasingly popular, their effectiveness at promoting learning can be attenuated by mind wandering (shifts in attention away from the task at hand towards unrelated thoughts). We investigated whether taking tests on to-be-studied information, also known as pretesting, could mitigate this problem and promote learning. In two experiments, participants viewed a 26-min video-recorded online lecture that was paired with a pretest activity (answering questions about the lecture) or a control activity (solving algebra problems), and with multiple probes to measure attention. Taking pretests reduced mind wandering and improved performance on a subsequent final test compared to the control condition. This result occurred regardless of whether pretests were interspersed throughout the lecture (Experiment 1) or were administered at the very beginning of the lecture (Experiment 2). These findings demonstrate that online lectures can be proactively structured to reduce mind wandering and improve learning via the incorporation of pretests.

In some educational contexts, such as during assessments, it is essential to avoid errors. In other contexts, however, generating an error can foster valuable learning opportunities. For instance, generating errors can improve memory for correct answers. In two surveys conducted at three large public universities in North America, we investigated undergraduate students' and instructors' awareness of the pedagogical benefits of generating errors, as well as related practices, attitudes, and beliefs. Surveyed topics included the incorporation of errors into learning activities, opinions about the consequences of studying errors, and approaches to feedback. Many students had an aversion towards making errors during learning and did not use opportunities to engage in errorful generation, yet studied or analysed errors when they occurred. Many instructors had a welcoming attitude towards errors that occur during learning, yet varied in providing students with resources that facilitate errorful generation. Overall, these findings reveal the prevalence of an ambivalent approach to errors: Students and instructors avoid generating errors but prioritise learning from them when they occur. These results have important implications for the implementation of pretesting, productive failure, and other error-focused learning techniques in educational contexts.