Inside StudyGo Learn: spacing, retrieval, and real retention

A practice method called Leren (Learn) in StudyGo operationalizes this principle even inside a single study session. While learning words, instead of showing the same word pair again immediately, we interleave other items between repetitions and escalate the response mode toward open questions (typed recall, not just recognition). We take this approach because it isn’t our goal to inflate practice scores; it’s to make knowledge stick for the moments that matter (quizzes, tests, and real-world use).

A student selects a vocabulary list, either from their school book, or a handmade one. In Learn, each item progresses through a small ladder of question types:

Once an item is answered correctly in each question type, it’s marked as completed and leaves the cycle. To solidify weak spots, a short final retrieval round resurfaces the most error-prone items. Spacing refers to how much time passes in between item repetitions. For example, let’s consider a student that is studying French, who encounters that ‘library’ means ‘bibliothèque’. This is first introduced as a Flashcard, after which several other items are presented to the student. Then, the ’library’–’bibliothèque’ item is presented again, this time as a multiple choice question, and so on. Spacing happens within a block of items: newly introduced words push earlier ones apart; as a practice block is ending, spacing narrows again. You can find more information on how Learn works here.

Students, parents and teachers rightly care about whether something works in the classroom and at home, not just in a lab. That’s why we compared two versions of Learn in StudyGo itself, with the only difference being how much spacing they applied. Everything else stayed the same: the content, the steps inside Learn, and the way we grade answers. To make the comparison fair, the system quietly and randomly assigned learners to one version or the other. This is often called an A/B test, but you can think of it simply as a coin toss that balances the two groups so we can attribute any later difference to the spacing, not to who happened to use which version.

We then watched what happened next, without asking learners to change their habits. Some students moved straight from Learn into a later Test (a different exercise type) on the same list, others came back after a short break, and others waited longer. That natural variety is useful, because it lets us see if the spacing advantage depends on the time gap between studying and testing. The research was conducted as part of a Master’s thesis by our Data Scientist Tatiana Litvin, in collaboration with the Vrije Universiteit Amsterdam and prof. dr. Martijn Meeter.

We cared most about the grade in a later Test on the same list, because that’s what signals real knowledge retention. However, we also looked at the Learn score during practice, because that tells us how the session felt in the moment. To keep the comparison fair, we accounted for the time between Learn and Test, for how long the Learn session took, and for whether a student did any extra practice on the same list in between. Technically, we used a statistical model that treats each learner and each list as its own “baseline”, so we’re not mixing up a very hard list with a very easy one, or a very advanced student with a beginner.

As expected, the version with less (tighter) spacing felt easier. As you can see in Figure 1, grades during practice with Learn were modestly higher when items came back sooner (small block). This is the kind of result you can feel as a learner: fewer moments of “hmm, what was that again?” and more smooth, familiar answers. The cost is that a lot of those easy wins come from recognition and short-interval retrieval. They boost confidence – great! – but perhaps they don’t challenge memory enough to make it stick.

Figure 1. Effect of the manipulation in spacing and quiz type on grade.

Also, as shown in Figure 1, when students tested themselves on the same list, the students that practiced with broader spacing came out ahead (large block). On average, Test grades were several points higher in the broader-spacing version. We also checked whether this advantage only appears after long breaks. It doesn’t. Many students take a Test exercise within an hour of Learn, and the broader-spacing benefit is already visible there; it simply persists as the gap grows. You can see this in Figure 2; large block students (more spacing) on average consistently outperform small block (less spacing) students.

Figure 2. The grade gap in practice method Test remains consistent over time.

Typing an answer from memory is like lifting a slightly heavier weight at the gym. You don’t need to make it impossible, but you need enough challenge to trigger the adaptation you want. Spacing creates that challenge by letting a little forgetting happen between encounters. When an item returns after other items, you have to reconstruct it. That effort is precisely what strengthens memory. So the short-term dip in practice smoothness is not unexpected, it’s the path to better recall at the moment that matters.

If the practice method Learn sometimes feels a touch tougher, especially compared to drills where the same item repeats right away, that’s by design. Lower practice scores don’t signal failure; they’re a sign that you’re doing the kind of work that generalizes to a later test. For teachers, the takeaway is encouraging: you don’t need to assign extra homework or new materials to get this benefit. The algorithm does the rebalancing inside the session, moving a bit of “payoff” from now to later. For parents, it helps to reframe the practice phase as training, not a report card. A few more productive “hmm, let me think” moments today can mean fewer blank stares tomorrow.

Based on these results, we’re standardizing the broader-spacing configuration in practice method Learn. Behind the scenes, this simply means that more items are kept “in play” at once, so repetitions land further apart within a session. The steps you see as a learner (flashcard, multiple choice, hints, open question) stay exactly the same, and the lists you choose don’t change. What changes is the rhythm between repetitions.

Overall, we can conclude that broader spacing inside a single Learn session leads to better recall on a later Test. Even if practicing can feel a bit tougher in the moment, students’ grades will benefit, which is why we’ve made it our new default. For students, that means more of today’s effort shows up when it counts; for teachers and parents, it means StudyGo quietly does the heavy lifting in the background so learners don’t have to study longer or change their routine. We’ll keep holding ourselves to this standard: design with learning science, check it carefully in real use, and ship only what helps knowledge stick.

[1] Soderstrom, N. C., & Bjork, R. A. (2015). Learning versus performance: An integrative review. Perspectives on Psychological Science, 10(2), 176-199. (link)

[2] Bjork, E. L., & Bjork, R. (2014). Making things hard on yourself, but in a good way: Creating desirable difficulties to enhance learning. In Psychology and the real world: Essays illustrating fundamental contributions to society (2nd ed., pp. 59-68). New York: Worth Publishing.

[3] Karpicke, J. D., & Roediger, H. L. (2008). The critical importance of retrieval for learning. Science, 319, 966–968. (link)

[4] Roediger, H. L., & Karpicke, J. D. (2006). Test-enhanced learning: Taking memory tests improves long-term retention. Psychological Science, 17(3), 249-255. (link)

[5] Kornell, N. (2009). Optimising learning using flashcards: Spacing is more effective than cramming. Applied Cognitive Psychology, 23(9), 1297-1317. (link)

[6] Karpicke, J. D., & Bauernschmidt, A. (2011). Spaced retrieval: Absolute spacing enhances learning regardless of relative spacing. Journal of Experimental Psychology: Learning, Memory, and Cognition, 37(5), 1250-1257. (link)

[7] Nakata, T., & Suzuki, Y. (2019). Effects of massing and spacing on the learning of semantically related and unrelated words. Studies in Second Language Acquisition, 41(2), 287-311. (link)

[8] Pyc, M. A., & Rawson, K. A. (2009). Testing the retrieval effort hypothesis: Does greater difficulty correctly recalling information lead to higher levels of memory? Journal of Memory and Language, 60(4), 437-447. (link)