The Learning Scientists

View Original

The Failed Spacing Effect 30 Years Later (Part 1)

By Yana Weinstein

Thirty years ago, Frank N. Dempster wrote an article entitled “A Case Study in the Failure to Apply the Results of Psychological Research” (1). His case study was the spacing effect - the finding that studying information presented spaced out over time is more effective for learning than is studying the same information in massed study sessions. In this article, Dempster outlined 9 possible reasons why the spacing effect might not have become mainstream practice, particularly in American classrooms (he does mention that other countries, such as Russia, had embraced this practice more readily - a statement I can anecdotally confirm from my Russian childhood). I thought it might be interesting to revisit this article today (i.e., 30 years later), and see how many of these reasons for failure to adopt the spacing effect still apply. 

In Part 1 of this blog, I look at the first 5 potential reasons described by Dempster in his review.

Reason #1: "The Phenomenon Has Not Been Known Long Enough”

Dempster argued that this was not a likely reason why the spacing effect was still largely absent from classrooms 30 years ago. To back up his point, he cited studies all the way from the 19th century (2) and a review from 1928 (3). We have now added another 30 years of knowledge, so this obviously can’t be the problem.

Reason #2: “The Phenomenon Has Not Received Recent Documentation”

This is another reason that Dempster suggested as a hypothesis, but rejected: The idea that perhaps the spacing effect was popular at the beginning of the 20th century, and then forgotten. Dempster cited studies from the 1970s (4) and 1980s (5) to reject this hypothesis, and of course we now have a very rich literature on the spacing effect from the past few decades (6). Clearly, there has been a sustained effort to investigate this phenomenon (cue jokes about spacing out research on the spacing effect!), so a dearth of recent research is not the answer.

Reason #3: “The Phenomenon Cannot Be Linked to Issues of Current Concern to Educators”

In plain English, this means “why should teachers care?”. Dempster argued that the benefits of spaced practice are “immediate and obvious”, as opposed to some obscure cognitive process investigated in the lab that we need to convince teachers to care about. After all, he says, we’re all very concerned about how students are spending their time in the classroom. Such concerns have only become more vital since the amount of information students are expected to learn seems to increase from year to year (see, for example, this article from the UK about the recent changes to examinations taken by 16-17 years olds).

Reason #4: “The Phenomenon Has Not Been Demonstrated Satisfactorily in School-Like Activities”

This possible concern is directly related to the lab-to-classroom model that Megan and I like to talk about and describe in Chapter 2 of our new book (7) - see p.17 for an illustration. The idea here is that perhaps the spacing effect had only been demonstrated with basic materials, such as word lists or unrelated facts. Dempster cited various studies including his own (8) to counter this possible concern; for example, he looked at the effect of spacing on recall of idea units from text passages. A related concern could be that spacing hadn’t been demonstrated at long enough retention intervals, but once again there was already evidence of spacing effects emerging after as long as a month (9). Since the late 1980s, numerous spacing studies have looked at educational materials such as clinical knowledge in medical school (10) and across long retention intervals as long as almost a year (11), further diminishing this concern.

Reason #5: There Are Serious Discontinuities in the Literature on the Spacing Effect

Now we finally get to the first reason that Dempster actually believed might be true! Here Dempster points out a very interesting aspect of the literature on the spacing effect: That there were many similar research questions addressed throughout the years, but that these weren’t necessarily connected up by appropriate citations of older work. To use a concrete example, Dempster mentions that a review published in 1974 (4) includes many spacing effect studies with simple materials and discusses theories of spacing effect that are directly tested in previous applied studies (12) that are not cited in the review. As a very quick check of whether this issue persists to this day, I took a 2016 review of the spacing effect (6) and compared the references in that review to the ones in this Dempster paper. 

The results were almost unbelievable to me: the 2016 review contains only 7 of the 66 references cited in Dempster’s 1988 review! This clearly shows that the discontinuity in the spacing literature still exists, and suggests that researchers who are studying the spacing effect today may not be drawing upon all the existing knowledge we have already acquired through previous studies with similar research questions. This type of discontinuity slows scientific progress and leads to researchers sometimes reinventing the wheel in their work. 

If I may speculate, part of the problem could be that deep reading of the literature is not sufficiently incentivized in the current publication model. Scientific publications are typically judged based on whether they provide a “novel contribution”; and since no reviewer or editor can be expected to go off for a few years and read all of the previous research on a particular topic, the “novelty” factor isn’t so much judged objectively in terms of whether a finding is truly novel (i.e., has not previously been found in the literature), but instead based on how exciting it is, or at best, how novel it is compared to a handful of other studies that are cited in that paper.

In Part 2 of this blog, I will discuss the remaining 4 reasons that Dempster suggested in 1988 for why the spacing effect was, at that time, not prevalent in the classroom.


References:

(1) Dempster, F. N. (1988). The spacing effect: A case study in the failure to apply the results of psychological research. American Psychologist43, 627-634.

(2) Ebbinghaus, H. (1913). Memory (HA Ruger & CE Bussenius, Trans.). New York: Teachers College (Original work published 1885).

(3) Ruch, T. C. (1928). Factors influencing the relative economy of massed and distributed practice in learning. Psychological Review35, 19-45.

(4) Hintzman, D. L. (1974). Theoretical implications of the spacing effect. In R. L. Solso (Ed.) Theories in cognitive psychology: The Loyola Symposium (pp. 77-99). Potomac, MD: Erblaum.

(5) Toppino, T. C., & DiGeorge, W. (1984). The spacing effect in free recall emerges with development. Memory & Cognition12, 118-122.

(6) Kang, S. H. (2016). Spaced repetition promotes efficient and effective learning: Policy implications for instruction. Policy Insights from the Behavioral and Brain Sciences3, 12-19.

(7) Weinstein, Y., & Sumeracki, M. (2019). Understanding How We Learn: A Visual Guide. Routledge.

(8) Dempster, F. N. (1986). Spacing effects in text recall: An extrapolation from laboratory to the classroom. Journal of Educational Psychology79, 162-170.

(9) Dempster, F. N. (1987). Effects of variable encoding and spaced presentations on vocabulary learning. Journal of Educational Psychology79, 162-170.

(10) Kerfoot, B. P., DeWolf, W. C., Masser, B. A., Church, P. A., & Federman, D. D. (2007). Spaced education improves the retention of clinical knowledge by medical students: a randomised controlled trial. Medical Education, 41, 23–31.

(11) Morris, P. E., & Fritz, C. O. (2000). The name game: Using retrieval practice to improve the learning of names. Journal of Experimental Psychology: Applied, 6, 124-129.

(12) Ausubel, D. P. (1966). Early versus delayed review in meaningful learning. Psychology in the Schools3, 195-198.