Research in psychology has repeatedly shown that interim tests lead to better performance on a final test than restudying the same information—a finding known as the testing effect (also retrieval-practice effect). However, widespread use of interim tests in educational settings has been hindered due to a lack of their theoretical understanding. The episodic-context account (Karpicke et al., 2014) states that tests promote learning of new associations between the learning material and the learning context, thereby strengthening memory for the contextual aspects of the tested episode. The rapid-consolidation hypothesis (Antony et al., 2017) states that tests accelerate integration of the tested episode into preexisting knowledge, resulting in a more permanent and context-free memory representation. Hence, both accounts make opposing predictions regarding how well contextual aspects should be remembered. In order to investigate whether tests improve or impair context memory, participants watched a video of a theft once (control group), twice (restudy group), or once followed by forced-choice questions about the scene (test group). Two days later, all participants answered the forced-choice questions, which concerned the perpetrator (central information) and several details of the event (contextual information). The final-test results showed that (a) central and contextual information were remembered better after testing than restudying and (b) the accuracy for central and contextual information were positively correlated. The findings are discussed in light of the most popular theories of the testing effect.