Retrieval Practice Is Costly For Some Students

Reading time: 2


Ross Morrison McGill founded @TeacherToolkit in 2010, and today, he is one of the 'most followed educators'on social media in the world. In 2015, he was nominated as one of the '500 Most Influential People in Britain' by The Sunday Times as a result of...
Read more about @TeacherToolkit

How do teachers ensure retrieval practice has notable benefits for all learners?

New research suggests the effect of working memory using retrieval practice only emerges when the task challenges WM capacity.

Using retrieval practice with different learners

Retrieval Practice Is CostlyIn a new 14-page paper published by Zheng et al. (October 2022), research explores the inconsistency regarding the effect of working memory capacity on the testing effect – otherwise known as retrieval practice.

The typical finding is that in the final test (exam), items practised in the test condition (e.g. mock exam) are better than those in a restudy condition (day-to-day classroom).

This research poses an interesting conundrum: it is unclear whether testing should be uniformly applied in the classroom and whether testing help certain subpopulations more than others.

This raises some important questions for teachers working with various students (gifted, pupil premium, special educational needs etc) who are more successful than others in certain learning conditions.

While many teachers are excited about the benefits of retrieval practice, if we stop to consider the various needs of our students in our classrooms, we should not be surprised to learn that students with a lower working memory capacity will benefit less from retrieval practice in certain scenarios.

For example, Monday morning or Friday afternoon? During detention or high-stakes exam? In English, maths or an Art lesson they love or loathe?

Methodology used

This research lasted four weeks, with each participant completing a visual search task over 9 sessions in the first 3 weeks of the trial. Spaced practice was used at various intervals (more details can be found on page 3-5).

Thirty-five undergraduate students – note, not classroom pupils – took part in the study, with 33 participants required to achieve 80% success. All students spoke Chinese as their first language, and two participants were excluded because of low accuracy in the final test.


The research suggested that “people with abundant working memory resources benefited from retrieval practice regardless of the stimulus frequency.”

Conversely, people with a lower working memory tended to have a “negative testing effect” and “demonstrates that retrieval practice is a costly learning technique.” The research suggests that a bottleneck of working memory emerges only when the demands of working memory exceed the working memory capacity.

Older people benefited from testing when feedback was provided but learned better through restudying than testing when no feedback was provided (Tse et al., 2010).

Two hypotheses are offered with my questions for teachers to consider below:

The first, that that the only benefit of retrieval attempt is for memories to be fully strengthened as long as the correct targets are retrieved. The second? That learners can benefit from re-encoding correctly retrieved information. The research writes that “successfully recalled associations may not be fully strengthened if WM resources are depleted during the retrieval attempt.”

Recommendations for teachers

  1. How should teachers adapt retrieval practice techniques for disadvantaged pupils?
  2. When should retrieval practice be used?
  3. Why should a retrieval practice quiz be adapted for different abilities?
  4. Do online quizzing platforms consider the working memory of the pupils using the software?
  5. What in-house research can schools conduct to learn how retrieval hinders or supports various learners?

Successfully retrieving the item does NOT guarantee that the memory is more effectively strengthened than restudying the item. (Zheng et al., 2022) 

One thought on “Retrieval Practice Is Costly For Some Students

  1. This is really interesting – however I’m not sure about some of your conclusions:

    You said “people with a lower working memory tended to have a “negative testing effect” and “demonstrates that retrieval practice is a costly learning technique.” – This was only true in the low frequency condition, in a test design specifically designed to overload the working memory of participants: “participants were required to decide whether the current stimulus matched the stimulus presented N pictures before”. This by no means suggests that people with limited working memory don’t benefit from retrieval practice, and it was not a ‘negative testing effect’, just a slightly lower improvement compared to re-study. Furthermore, the disadvantage of testing vs re-study in this condition was lower than the advantage of testing in the other three conditions, including those with lower working memory when given high frequency practice.

    You also stated “The typical finding is that in the final test (exam), items practised in the test condition (e.g. mock exam) are better than those in a restudy condition (day-to-day classroom)”. I think this is a straw man or misunderstanding of what retrieval practice means: retrieval practice should be about focusing in on retrieval, not on exam performance. Answering mock exam questions requires a range of skills and knowledge to be applied. Retrieval practice should be simple recall of key information, ideally with recognition questions progressing to recall questions as knowledge improves. Once that knowledge is secure you can then progress to applying the knowledge in final test conditions.

    Online quizzing tools, used correctly, are a great way to cut out those working memory demands and focus in on retrieval alone. In my reading of it, this study supports the use of online quizzing systems, or other activities like flashcards, matching cards etc that give the maximum frequency of recall tasks with the minimum working memory requirements. It just emphasises further the need to develop questions / tasks that minimise the need for working memory or other cognitive tasks and focus in on repeated retrieval practice with immediate feedback to make the most of the testing effect.

Leave a Reply

Your email address will not be published. Required fields are marked *

This site uses Akismet to reduce spam. Learn how your comment data is processed.