R3 2.9 May 28, 2024 Pretesting’s Surprising Effects on Memory for Material You Haven’t Studied Yet
With these effects, time is on your side - even more than anyone thought based on previous research on test-potentiated learning.
Those of us working on the traditional academic calendar typical of North America have reached the year’s biggest milestone: the end of the spring semester. For some, summer will bring a change of pace, and maybe, a slowing of pace as well as minds turn away from daily teaching and administrative responsibilities and toward other goals: research, writing, and getting to all kinds of stray tasks that didn’t make it to the top of packed academic-year to-do lists.
For many of us, professional development - reading, webinars, maybe a conference or two - is something that comes back into the foreground during this season. It’s always my hope that R3 can be one part of your catch-up, development, and reading goals, whether tied to academic-summer or not. In today’s issue, we’re focusing on one of the impacts of retrieval practice that I find especially intriguing: the forward testing, or pretesting effect. Essentially, this effect involves potentiating the formation of new memories by answering questions about material before you review it.
The idea that simple quiz questions can accelerate retention of concepts you have yet to encounter is surprising, powerful, and clearly something that could be put to good use in teaching. I’ve been mentioning test-potentiated learning of this sort in nearly every keynote and workshop I’ve given over the past year, and I think its existence can be one of the most persuasive reasons to take retrieval practice into account when setting up or refreshing a course. So I was especially intrigued to see this new article, which reports on a pair of relatively simple, straightforward experiments that sought to replicate the finding that word associations are remembered better when participants first guess at the answer, before studying. The article also asks if the advantage of pre-testing holds up with materials that are a better stand-in for typical educational materials that students would encounter, and finally, whether the advantage decreases, increases, or stays the same with longer time intervals between study and test.
Citation:
Kliegl, O., Bartl, J., & Bäuml, K. H. T. (2024). The pretesting effect comes to full fruition after prolonged retention interval. Journal of Applied Research in Memory and Cognition, 13(1), 63–70.
DOI:
https://doi.org/10.1037/mac0000085
Paywall or Open:
Paywall
Summary:
University students were assigned materials to study, then tested on recall. Materials either consisted of word pairs (Experiment 1) or realistic educational text passages (Experiment 2). Some of the to-be-retained information was pretested by having participants guess the answer before encountering the studied materials, and some of the information was not pretested. Tests took place after intervals ranging from one minute to one week. Across both experiments, pretesting produced better retention, and the advantage increased for longer retention intervals. Results support the practice of pre-testing as a way to boost retention of studied information, and also support the idea that semantic elaboration is one mechanism underlying the forward testing effect. That is, pre-testing begins to activate related ideas and concepts during initial study, then when participants have more time to think back on the material, these elaborations are further enhanced, and in turn, enhance recall of the material through an enriched set of connections or pathways in memory.
Research Questions:
- Does the pretesting effect (i.e., the advantage associated with answering a question about a fact versus simply studying it) change as the interval between study and test increases?
- Does the pretesting effect and its interaction with time interval also occur for realistic educational materials, similar to what has been found for contrived laboratory tasks (e.g., word pairs)?
Sample:
All participants were German university students. Experiment 1: N = 90 (66 female, 24 male); Experiment 2: N = 90 (67 female, 23 male)
Method/Design:
For both Experiment 1 and Experiment 2, all participants answered pretest questions on some items and not on others, and were randomly assigned to a short, medium, or long retention interval. Experiment 1 used weakly associated word pairs as the to-be-studied materials, meaning pairs of words that had some kind of relationship but that were unlikely to be guessed on the first try (e.g., frog-pond). The retention intervals, meaning the time delay between studying and the test, were fairly short: 1, 10, and 30 minutes. Experiment 2 used facts drawn from a text passage that were unlikely to be well-known by participants, and also featured longer retention intervals: 1 minute, 30 minutes, and one week. Accuracy of recall of the studied information was measured through either providing the first word and asking participants to fill in the second (frog - ?) or by presenting participants short-answer factual questions about the text passage that was read. In all but the 1-week interval, participants engaged in distractor tasks, such as counting backward or playing Tetris, between the study and test phases.
Key Findings:
In both studies, there was a significant interaction between interval and type of study (pretest versus simply reading the information), such that the advantage of pretesting increased along with the retention interval. In the case of the longest interval (one week), correct recall was 32.8% more likely for items that had been pretested, the largest advantage found across all conditions.
Choice Quote from the Article:
Testing material to be learned, like paired associates (e.g., plate–fork) or a prose passage (e.g., about the big bang theory), shortly after study can be highly beneficial for retention of the information—as demonstrated by the wealth of research on the so-called testing effect. Somewhat counterintuitively, this benefit of testing does not only arise when testing takes places after study but can also emerge when testing precedes study, and individuals are thus forced to guess the (unknown) correct answer—typically with little success. To be of potential use for educational settings, this so-called pretesting effect—like the classic testing effect—should survive prolonged retention intervals between acquisition and final testing and not be restricted to situations in which the final test follows shortly upon acquisition. We addressed the issue in two experiments. The one experiment employed paired associates as study material and retention intervals of up to 30 min; the other experiment employed an educationally more relevant prose passage as study material and retention intervals of up to one whole week. In both experiments, we examined whether pretesting some of the study material improved recall of the pretested information relative to other material that was not pretested. Both experiments replicated the benefit of pretesting for retention of studied information. Strikingly, in both experiments, the pretesting effect roughly doubled from the short to the longest retention interval.
Why it Matters:
Practically speaking, these findings should help put to rest the argument that retrieval practice, and other techniques designed to boost retention, aren’t relevant to learning complex information, or that they’ll simply contribute to a cram-and-purge dynamic where memories last only up to exam day, never to be heard from again. They’re also a nice support for the practices - often advised by teaching experts – of querying students about what they think, know, or believe before diving into a topic. This practice probably does double or triple duty – stoking curiosity, helping the instructor tailor instruction to what students already know, and communicating care, all at the same time as it primes students to take in and retain the material they’re about to encounter.
I also think that the deeper message here about mechanisms is worth amplifying, as it points to some big truths about memory that are useful for all teachers to know. The authors conclude from their results that, although there may be several reasons why delays amplify pretesting effects, one of them is likely elaboration. Elaboration is a general concept within memory theory that involves adding to or expanding on information in memory – things like noticing connections between something new and what you already know, connecting something you’ve learned to a personal experience you remember, or even forming an opinion about what you’re learning. The more complex and richly interwoven with other memories the new memories become, the better chance that you’ll be able to re-activate them later when you need to. This is a great way to think about memory, compared to thinking of it as just a big dumping ground for isolated, decontextualized facts. And it reinforces the intuition of many teachers that memorizing facts in isolation is not a productive way to build sophisticated and practically usable knowledge.
There are also some interesting implications here for guessing. One subtle but important distinction addressed within the article is the difference between weakly associated items (frog-pond) and non-associated or randomly paired items (frog-bread). The point here isn’t how word pairs are memorized (something you’d rarely see as an actual academic learning task). Rather, it’s that guessing – which is essentially what students are doing in a typical pre-test – is only productive when there’s some chance of getting it right. Mind you, guessing and getting it wrong still benefits memory – but not if students are so adrift that they can’t begin activating some kinds of associations, interpretations, or other meaningful trains of thought that produce elaboration. This is why studies of pre-testing don’t use random pairings, and why truly unguessable questions likely would not be helpful in actual class settings.
In the end, this article provides great food for thought from both theoretical and practical angles. Reading it had me envisioning how faculty would run with the suggestion that even moderately guessable facts are better remembered when they are introduced via a pre-test. I’m always interested to see how faculty adapt concepts from learning sciences for their own disciplines and styles, and especially so with these findings. I’d love to see what kinds of “moderately guessable” but important-to-remember information faculty would identify, and how they might incorporate pre-testing in different types of courses.
Most Relevant For:
Instructional designers; faculty; educational technology designers; researchers working on retrieval practice and related phenomena
Limitations, Caveats, and Nagging Questions:
I’ll reiterate a caveat that’s now become standard for me, as a cognitive psychologist: I realize that memory is not in and of itself “learning,” nor should it be the end-all and be-all of anyone’s teaching practice. That said, techniques like pre-testing can be a big help to faculty as they seek to balance foundational knowledge-building with higher-order application and critical thinking. Whenever you can accelerate the former, it should create more time and energy for the latter – a good thing, in any course or discipline that I can think of.
Apart from that, I didn’t see anything in the article’s methodology or interpretation that I think should raise any major cautions – but let me know in the comments if you noticed anything that I missed!
If you liked this article, you might also appreciate:
Ahn, D., & Chan, J. C. K. (2023). Does testing potentiate new learning because it enables learners to use better strategies? Journal of Experimental Psychology: Learning Memory and Cognition, 50(3), 435–457. https://doi.org/10.1037/xlm0001233
Arnold, K. M., & McDermott, K. B. (2013). Test-potentiated learning: Distinguishing between direct and indirect effects of tests. Journal of Experimental Psychology: Learning, Memory, and Cognition, 39(3), 940–945. https://doi.org/10.1037/a0029199
Chan, J., Meissner, C. A., & Davis, S. D. (2018). Retrieval potentiates new learning: A theoretical and meta-analytic review. Psychological Bulletin, (July). https://doi.org/10.1037/bul0000166
Miller, M.D. (2011). What college teachers should know about memory: A perspective from cognitive psychology. College Teaching, 59, 117-122. https://doi.org/10.1080/87567555.2011.580636
Miller, M.D. (2022). Remembering and forgetting in the age of technology: Teaching, learning, and the science of memory in a wired world. West Virginia University Press.
Pan, S. C., & Rickard, T. C. (2018). Transfer of test-enhanced learning: Meta-analytic review and synthesis. Psychological Bulletin. https://doi.org/10.1037/bul0000151
Pastötter, B., & Bäuml, K. H. T. (2014). Retrieval practice enhances new learning: The forward effect of testing. Frontiers in Psychology, 5(APR), 1–5. https://doi.org/10.3389/fpsyg.2014.00286
Uner, O., & Roediger, H. L. (2018). The effect of question placement on learning from textbook chapters. Journal of Applied Research in Memory and Cognition, 7(1), 116–122. https://doi.org/10.1016/j.jarmac.2017.09.002
File under:
Retrieval practice; forward testing effect; pre-testing; test-potentiated learning; memory theory