Learning material with different difficulty : When is testing more beneficial than study?

Abstract: Repeated testing is known to promote more lasting memories than repeated study of the same material. This thesis tested the hypothesis that the benefit of repeated testing, compared to repeated study, should be more evident for difficult than for easy material. This effect should be due to the potential demand for more effortful retrieval of difficult material, and thus, following the "retrieval effort hypothesis", difficult material should lead to better memory. A test-group and a study-group encountered easy related word-pairs and difficult unrelated word-pairs under different learning conditions. Afterwards, participants did one test directly after training and one after a one week delay. Repeated measurement ANOVA was used to investigate differences in number of correct recalls within categories and between groups with 18 participants in each condition. The results replicate results from previous studies, indicating that testing is better than study for long-term retention. Interestingly, the test-group benefitted on both easy and difficult material compared to the study-group, but on average forgot more word-pairs within the difficult category than in the easy category. This is in contrast to what might be predicted by the "retrieval difficulty hypothesis" and suggests that the effect of testing on difficult material is more fragile than the effect on easy material.