Retrieval Practice

Paving pathways in the memory maze

What is retrieval practice?  What make is effective, and why it is not as common as we hope?

Evidence, explanations, examples , considerations towards applications and examples of practice in various classrooms.

Retrieval practice is a strategy to rehearse already learned information by trying to recall it from memory, or to actively locate the piece of information in one's mind. It is based on what is known as “The Testing Effect” – the findings that future long-term memory performance is enhanced for materials practiced by testing when compared to rehearsal. Unfortunately, tests have an unfavorable reputation: they are associated with assessment, performance measurement, and many students and former students tend not to like them because of the consequences that they often bear. In retrieval practice we use tests as an effective way to PRACTICE. 

Retrieval practice is a strategy to rehearse already learned information by trying to recall it from memory, or to actively locate the piece of information in one's mind. When applied effectively it allows learners to identify gaps in their knowledge and actively resolve them. Therefore, it’s a crucial stage of a meaningful learning process that allows learners to use the knowledge they have acquired in any relevant future situation.   It is based on what is known as “The Testing Effect” – an established series of findings from cognitive psychology that shows enhancement of long-term memory performance after practicing by using retrieval as compared to rote rehearsal.  One barrier to using retrieval practice is the unfavorable reputation of Tests: they are associated with assessment, performance measurement, and most students would avoid them if they could, because of the consequences that they often bear. In retrieval practice, we use positive aspects of tests as an effective way to practice, which in turn may improve performance in high-stakes tests and even reduce anxiety. More importantly, retrieval practice has every potential to improve performance not just in high-stakes tests, but when the information is needed in the much longer term.

Retrieval Practice – the evidence 

Retrieval practice has made it to the top of a list of the most effective evidence-based learning strategies1, and it was demonstrated to be effective for learners across ages and learning materials.  

Retrieval practice is often compared to strategies like re-reading or re-studying.  It would probably not surprise you that re-reading is a very common way of practice among students, and it is often recommended by teachers as well. Re-reading the book or the notes seems perfectly reasonable: we are re-exposing ourselves to previously learned material in order to stabilize its representation in our minds. We can actually feel the benefit as we are practicing this method of rehearsal. Interestingly, however, this method was found to be ineffective. More accurately, re-reading may be effective for the short-term (defined loosely between minutes and a few days), but on longer time scales it is almost entirely ineffective.  The major difference between re-reading and retrieval practice is that when re-reading, the information is presented to the learners in its whole from an external source, while with retrieval the learners are searching the information in their own mind, following just a cue.

An influential study2 published in the journal Science in 2008 clearly demonstrated this point: Karpicke and Roediger let college students learn 40 words in Swahili and their translation to English. The Experiment included Study phases in which each word appeared next to its translation on a computer screen, and Retrieval phases in which a Swahili word appeared and participants were requested to type the English translation if they could recall it.

These phases repeated alternately 4 times each: 

Study1 >>Retrieval1  >>Study2 >>Retrieval2 >> Study3 >> Retrieval3  >>Study4 >> Retrieval4  

All participants were tested on their memory one week later.  The goal of the study was to find out which of the study phases, Re-Study or Retrieval contributed more to the long-term performance. To test that the number of words in each phase (study or retrieval) was changed: one group had twice as many Retrieval repetitions than Re-study repetitions, and another group had it the other way around

Which ratio would you choose for your own learning? 

You can see the memory performance after one week in that image: 

The group that had two-thirds of their study trials in the form of RETRIEVAL, averaged 8o% of the words

The group that had two-thirds of their study trials in the form of STUDY, averaged 36% of the words.

Additional experimental groups (not shown here) demonstrated that any extra STUDY repetitions did not improve the final recall performance for either of these groups and were therefore a waste of learning time.

This study proved that testing that induced retrieval is a very effective practice strategy!


> The figure is reconstructed on the basis of Karpicke and Roediger, 2008. It is partial and simplified for purposes of clarity. See the original here where you can find the full experimental design (4 groups) .

Why does retrieval practice work?

When attempting to retrieve a memory, we invest deliberate effort in reconstructing the pathway to the stored information, by reactivating a sequence of neural networks that are required in order to bring the information to mind. This pathway is represented in a simplified model on the right panel using the model that was introduced in the previous pages

When restudying the focus is on the information that is already presented, and by that bypassing the pathway leading to it (left image).

When practicing by retrieval the entire pathway is rehearsed, not just the end-goal. It is obvious that our ultimate goal is to be able to retrieve the knowledge voluntarily, not just recognize it, and hence practicing this ability is the way to go. ( unfortunately, this is not always in line with our intuition, as restudying is easier and immediately rewarding...)

It is like navigating to a familiar destination

an analogy would be navigation to a familiar place: knowing the way is as important as knowing the destination (in a traditional, no GPS situation). If you don’t know how to arrive at a place you are dependent on someone else that would lead you to or “land” you at the correct location. By learning to navigate by yourself, and preferably from multiple starting points you gain ownership of the information and independence in using it. 

Retrieval practice is a more effortful task, but the effort is the reason it works. It is interesting to note that the concept of rehearsal is intuitive to most learners; we usually appreciate the importance of practice in acquiring mastery. Yet, learners often disregard the question:

“What to rehearse?” 

Rehearsing the learned information is straightforward, easier, and more rewarding so it is often the intuitive choice. However, if the end goal is to be able to retrieve the information and use it whenever required, it becomes clear that we need to rehearse just that: the act of retrieving information and using it. 

How to make Retrieval Practice effective?

To put all the above together for a simple recommendation – it is possibly most beneficial to use retrieval practice to probe previously learned relevant information such that it is practiced and also active in students' minds as the basis for new learning.

Examples from different classrooms

Retrieval practice is highly effective and it is spreading as a teaching approach, however, it is important to note two major challenges: one issue is the unfavorable reputation of tests, which usually influence older students’ (and sometimes teachers’) willingness to use tests for learning. The other barrier is the fact that retrieval practice is harder to engage in, it is not always successful, and it challenges learners to face their knowledge gaps and mistakes. It is therefore important to frame retrieval practice as a learning activity, with positive approach and consequences. The way we address the activity (e.g. quiz vs. review), how we frame it (e.g. surprise quiz vs. morning review routine),  the stakes (high vs. low or none), and how much room we leave for making mistakes and correcting them are essential factors. In this blog you can see an example of a simple class activity that also develops students’ meta-cognitive skills.

It is noteworthy that explanations of retrieval importance may be valuable in some settings but they are certainly not enough, because when students are left to choose by themselves, they often choose the easier way – i.e re-reading (it is human nature to choose easy and immediately-rewarding pathways). It is therefore highly recommended to build retrieval practice activities into the learning ROUTINE in the classroom, rather than suggesting that students will use it at home by themselves. It is sometimes challenging to make time for a questionnaire or review during lessons, but it is probably worthwhile: you may find this example from secondary history teacher, and this detailed application from a higher-education psychology professor very informative, compelling and applicable. 

Another good reason to apply retrieval practice in the classroom rather than at home is the ability to make it available for ALL LEARNERS. There are learners who are less inclined to test themselves than others. It can be attributed to their previous learning experience or their self-confidence. When we suggest to learners that they will practice by retrieval independently, we are most likely encouraging only the already inclined students. When retrieval practice is applied in class, as a learning activity and by means that allow everyone to answer individually and discretely, we are actually promoting the learning of all students. The familiar routine of a teacher asking a question, only a few students are volunteering to answer and only one actually speaking is beneficial only for the active student(s), but not for all the rest (even if they were listening). There are many simple methods by which “on the fly” teachers’ questions can support the learning of all the students in the classroom:  cold- calling, using student response cards, or personal mini-whiteboards are all simple low-tech examples that can make a huge difference in your classroom. There are also various technological solutions that enable the added value of storing the information for formative purposes, like kahoot!, socrative or notably plickers – that allows individual and discrete, digitally registered answers without the use of students’ personal devices!

Luckily in recent years,  more and more teachers learn about retrieval practice, and come up with new and creative ways to use it in their classrooms. The best part is that they share their method, so many other students and teachers can benefit. Here are two last examples of short and focused blogs by two teachers that describe and demonstrate specific ways to encourage retrieval among their students: one is a simple yet sophisticated way to make the most out of multiple-choice questions, and the second is an idea to make sure students engage in retrieval practice of recent as well as older material.

To summarize, retrieval is highly effective as a method of practice, it has a strong continually-growing body of evidence, and most importantly - it can be applied in any classroom to enhance the learning of all students, to close gaps, and supply students with effective learning-tools for the rest of their journey.

Resources & additional information:

Specifically about retrieval practice, you can find a collection of teachers’ implementations and a summary of the many benefits of this method. 

References

1       Dunlosky, J. (2013). Strengthening the student toolbox: Study strategies to boost learning. American Educator, 37(3), 12-21  [PDF]

2       Karpicke, J. D., & Roediger, H. L. (2008). The critical importance of retrieval for learning. science, 319(5865), 966-968.‏ [PDF]

3       Blunt, J. R., & Karpicke, J. D. (2014). Learning with retrieval-based concept mapping. Journal of Educational Psychology, 106(3), 849.‏ [PDF]

Referenced Blogs:

How to Create Retrieval Practice Activities for Elementary Students by Megan Sumeracki (formerly Smith) via 'The Learning Scientists' http://www.learningscientists.org/blog/2017/4/6-1

Color Coding Recall Attempts to Assess Learning by Blake Harvard via 'The Effortful Educator' https://theeffortfuleducator.com/2017/09/14/color-coding-recall-attempts-to-assess-learning/

Nothing new, it’s a review – on why I killed my starters by Ben Newmark https://bennewmark.wordpress.com/2017/11/13/nothing-new-its-a-review-on-why-i-killed-my-starters/

Retrieval Practice: The What, Why, and How for Classroom Instruction by Pooja Agarwal via NOBA Blog http://nobaproject.com/blog/2017-11-08-retrieval-practice-the-what-why-and-how-for-classroom-instruction

Technique 22: Cold Call via Teach Like a Champion  http://teachlikeachampionch4.blogspot.com/p/technique-22-cold-call.html

Students Response Cards via The teacher Toolkit http://www.theteachertoolkit.com/index.php/tool/student-response-cards

THE Number 1 Bit of Classroom Kit: Mini-whiteboards by Tom Sherrington via teacherhead https://teacherhead.com/2012/08/28/the-number-1-bit-of-classroom-kit-mini-whiteboards/

Learn, Teach, Model RETRIEVAL PRACTICE by Tricia Taylor via 'The Learning Scientists'  http://www.learningscientists.org/blog/2018/2/28-1

RESEARCH BITES – ELABORATIVE INTERROGATION via Durrington Research School blog   https://durrington.researchschool.org.uk/2018/01/16/research-bites-elaborative-interrogation/

Retrieval Practice Challenge Grids for the classroom by Kate  Jones  https://lovetoteach87.com/2018/01/12/retrieval-practice-challenge-grids-for-the-classroom/

Maximizing the Effectiveness of Multiple-Choice Qs by Blake Harvard via 'The Effortful Educator' https://theeffortfuleducator.com/2018/05/15/maximizing-the-effectiveness-of-multiple-choice-qs/