Spacing Effect

Memory & Learning

Kornell (2009)

Is cramming less effective than spaced study sessions?


“The spacing effect—that is, the fact that spacing learning events apart results in more long-term learning than massing them together—is a robust phenomenon that has been demonstrated in hundreds of experiments dating back to Ebbinghaus (1885)” (Kornell, 2009). Testing himself on nonsense syllables, the German psychologist found that spaced repetition was one of the most effective techniques to flatten the forgetting curve.

The Spacing Effect can be explained through Atkinson and Shiffrin’s (1968) Multi-Store Model of Memory, as items in the long-term store lose their strength and decay over time, unless they are retrieved, brought back into conscious awareness in the short-term store, and rehearsed again.

Participants & Procedures

25 students from UCLA (Los Angeles, CA, USA)

The researcher investigated participants’ ability to learn GRE-type (college admisson level) vocabulary using flashcards pairing 40 synonyms. The pairs were selected so that subjects would know the meaning of the second, but not the first word (e.g., “effulgent: brilliant”). A pilot study was conducted to confirm this assumption. The experiment was conducted online over the course of 5 days, with participants studying 2 sets of 20 flashcards under 2 different conditions. In the “spaced” condition, a stack of 20 flashcards was studied twice each day of the 4 study sessions. In the “massed” condition, another stack of 20 was divided into 4 smaller stacks of 5 flashcards, each of which was studied 8 times on a single day.

Thus, each flashcard was studied the same number of times (8) in each condition. The only difference was that those 8 times were either “massed” in a single day or “spaced” across 4 days. At the end of each session, participants were also asked to predict how well they would do on the final cued recall test. This test took place on the 5th day: participants were given the first words in the pairs, and asked to remember their synonyms. In addition, the protocol enabled the researcher to analyze the specific effect of “cramming” by focusing on the 5 flashcards studied on the 4th day in the “massed” condition.


Unsurprisingly, the last small stack was remembered better than the other “massed” ones studied the previous days. However, spacing (54%) was much more effective than cramming (34%), and this even though participants reviewed all 20 “spaced” flashcards twice over 4 days, while they reviewed the 5 “crammed” ones 8 times on the final day.

Interestingly, at the end of session 1, participants falsely predicted that they would remember the “massed” flashcards better than the “spaced” ones. This can easily be explained within the Multi-Store Model of Memory: the “massed” flashcards were maintained in the short-term store, making it feel temporarily full, but decayed rapidly because, contrary to the “spaced” ones, they were not regularly retrieved from the long-term store. As noted by the researcher, “the impression of learners from during their first session are probably of primary importance, because those impressions are likely to serve as the basis for subsequent study decisions.” 

IA Tip

In line with IB guidelines, we recommend that students only compare two conditions in their experiment and obtain a single measurable result for each participant in each condition. Doing otherwise would complicate inferential statistics without any benefit as far as the IA is concerned.

Here, students may want to focus on participants’ scores on the final trial. To avoid confusions, they could also measure participants’ performance as a number rather than as a percentage of correct answers. Finally, if conducting the study over the course of several days creates proves too difficult, students might want to refer to Experiment 1 in the same article, which studied the same phenomenon during a single session.


Kornell, N. (2009). Optimising learning using flashcards: Spacing is more effective than cramming. Applied Cognitive Psychology, 23, 1297–1317.