Although typically used for assessment, tests are considered as one of the most effective learning techniques. Practice tests can be provided after the to-be-learned information (i.e., posttests) or beforehand (i.e., pretests). Both types of practice tests have been shown to enhance prior learning. In addition to this backward effect, posttests also enhance subsequent learning of newly presented information (i.e., the forward effect of testing). This symposium aims to present recent findings from various labs on the benefits of practice tests and to examine its underlying mechanisms. Kliegl et al. examined the benefit of pretests and how its magnitude is moderated by retention interval and the presence of interfering information. Shanks et al. examined the grain size hypothesis of posttests proposing that several tests of smaller amounts of information enhance long-term retention more than a single test on all information. Bencze et al. investigated event-related potential (ERP) correlates of repeated retrieval (vs. restudy) practice to specify the contribution of episodic recollection and post-retrieval evaluation processes to long-term recall success. Rummer et al. examined students’ metacognitive accuracy for long-retention benefits of posttests compared to rereading and notetaking; they specifically used offline judgements of learning that are made independent of the current learning situation. Kubik et al. examined the forward effect of testing in visual-spatial learning and how the amount of proactive interference moderates its size. Finally, Pastötter et al. examined whether the forward effect of testing is immune to stress induced after encoding. Together, this symposium will provide insights on the underlying mechanisms of practice tests and its practical implications in educational settings.