Does testing potentiate new learning because it enables learners to use better strategies?

Abstract

Testing can potentiate new learning, which is often called the forward testing effect. One potential explanation for this benefit is that testing might enable participants to use more effective learning strategies subsequently. We investigated this possibility by asking participants to report their encoding strategies in a multilist foreign language learning paradigm with four preregistered experiments and one non-preregistered pilot experiment. In Experiments 1-3, participants learned three lists of Chinese-English pairs; one group took a test after every list (i.e., test condition) and the other group took a test only for the criterial List 3 (i.e., restudy condition). In addition, participants completed a transfer test and a study strategy survey. Although we found a forward testing effect in all experiments, participants in the test and restudy conditions did not report differences in strategies. In Experiments 4 and 5, we used a within-subject design so that we could correlate changes in strategy use with the magnitude of the forward testing effect on an individual level. Interestingly, individual differences in strategy change were moderately associated with the magnitude of the forward testing effect, but even here, strategy change did not mediate the effect of testing on performance. Overall, our data showed that, at least for foreign language learning of Chinese characters, interim testing did not enhance new learning by altering participants' subsequent encoding strategies. Moreover, our data showed that interim testing did not promote the transfer of Chinese language learning to novel characters.

The magnitude of the testing effect is independent of retrieval practice performance.

“…Because most existing studies were not explicitly designed to examine the relationship between retrieval practice performance and the testing effect, variations in retrieval practice performance coincided with changes in task demands. These changes might alter a learner’s behavior or strategies (Ahn & Chan, 2022, 2023; Cho & Neely, 2017), which could have downstream consequences on final test performance and the testing effect that are independent of retrieval practice performance per se. For example, recalling materials collaboratively (Vojdanoska et al, 2009) can alter how students approach a retrieval task and promote processes that are otherwise absent during individual recall (e.g., collaborative inhibition; Blumen & Rajaram, 2008).…”