Research has revealed that retrieval duration is a key factor of Retrieval Practice Effect (RPE). Appropriately extending the retrieval duration can generate more retrieval efforts and thus effectively improve memory performance and retention, especially for items being successfully retrieved in initial tests. However, when considering learning practice, conditions become more complicated. Furthermore, according to studies on study-time allocation, material difficulty will affect the monitoring and control of the learning process. Learners tend to adjust the choice of learning content according to the difficulty of the learning materials, which in turn leads to different learning outcomes. Besides, based on the Desirable Difficulty Hypothesis of RPE, compared with easier retrieval activities, learners will pay more effort to retrieve difficult items. Once retrieved successfully, those items’ storage strength can be improved, and so does the long-term retention. The Elaborative Retrieval account and the Episodic Context Account of RPE also predict that when learning items are relatively difficult, learners may reconstruct and update learning backgrounds, generating more retrieval clues during retrieval processes, and increasing the possibility of successful retrieval in the future.
The current research attempts to examine how material difficulty moderate the relationship between retrieval duration and RPE, and to find a theoretical framework to account for it. Experiment 1 is a 2 (retrieval duration: 8 seconds, 13 seconds) × 2 (material difficult: easy, difficult) mixed design. Sixty-one college students studied 20 pairs of the difficult material (weak cue related, e.g. farmer-tea) and easy material (strong cue related, e.g. lawyer-judge) respectively, and then retrieved the relevant target words based on the cue words for 8 or 13 seconds. Participants took final test to assess their learning performance (test scores and retention rates) after 5 minutes. Experiment 2 took a similar design, except that the final test was extended to 24 hours to examine material difficulty and retrieval duration on RPE in a delayed test.
The results showed that in both immediate and delayed test conditions, extending retrieval time led to better memory performance and retention rate. In the immediate test condition (Exp 1), there was no significant memory performance difference between different material difficulty levels when retrieval time was short. However, we observed that the memory performance of difficult materials was higher than that of easy materials when retrieval time was extended to 13s from 8s. And participants’ memory retention rate of difficult materials was always higher than that of easy materials no matter retrieval time was relatively long or short. When the final test was given 24 hours later (Exp 2), we observed that the memory performance and retention rate of the difficult materials were higher than that of the easy materials regardless of the retrieval durations.
These data suggest that extending the retrieval time can promote individuals’ memory retention, but this effect is moderated by the material difficulty levels. We discussed the results with respect to the Episodic Context Account and the Hypothesis of Desirable Difficulty. Retrieval practice involves attempting to reconstruct a prior learning context, and when the learning tasks are difficult, it may improve learners’ storage strength, meanwhile the representation of context is updated to include features of retrieved contexts and the current context, which thus enhances future retrieval.
