GUEST POST: How Does Testing Affect Related, Untested Content?
By Michelle Rivers
Michelle Rivers is a PhD student studying Cognitive Psychology at Kent State University in Ohio. Her research applies theories of learning and memory to enhance educational practice. She is particularly interested in investigating factors that contribute to students’ metacognitive judgments and how learners understand and manage their own learning. She can be found on Twitter at @meta_michelle and her website is https://sites.google.com/view/michellerivers/. Michelle previously contributed a guest blog on how expecting a test promotes learning.
You’re doing everything right as an instructor. You encourage your students to study using flashcards and other forms of retrieval practice (i.e., recalling information from memory) and regularly administer quizzes during class. But, due to practical constraints, you can’t possibly test your students on all the content they need to know. Similarly, students who study using retrieval practice are unlikely to test themselves on every course concept in preparation for an upcoming exam.
Fortunately, you do not need to test your students on all of the course content for all of the course content to benefit from retrieval practice. However, if you do test your students on just a subset of course content (or if your students engage in selective testing on their own), the way testing is structured for that content matters. As educators, what can we be doing to ensure related, untested content still benefits from testing?
The Problem: Retrieval Can Cause Forgetting
Testing can be a powerful memory enhancer. But it has a dark side: Retrieving some content from memory can cause forgetting of other, related content; a phenomenon called retrieval-induced forgetting (1). For example, if participants study facts about two different islands, so called “Tok” and “Bilu,” and then practice retrieving half of the facts about Tok island, their recall performance on a final test is lower for initially untested facts about Tok compared to facts about Bilu (that never received retrieval practice; 2). This forgetting effect has been demonstrated across a wide variety of material and contexts, but can be particularly problematic in educational settings: If students are tested on a subset of content, their recall of related, untested content may be susceptible to impairment in the future, such as on final exam.
Simplified schematic of the method used by Macrae and MacLeod’s (1999; Experiment 2) retrieval-induced forgetting study. Final test performance was lower for facts that were related to facts that received retrieval practice (e.g., The capital of Tok) compared to facts about the island that did not receive retrieval practice (e.g., Bilu’s only major export).
...But Not Always!
Other research has identified conditions in which the fate of untested, related content following retrieval practice is not so bleak (3,4,5). For example, Chan (2009) found that when participants were presented to-be-learned information as a coherent passage, as opposed to presenting the same information as a series of disconnected facts, they were protected from retrieval-induced forgetting on a final test of the information (4). In particular, participants studied information (either as a coherent passage or a list of facts), practiced retrieving some of the content from the passage, and then completed a final test 20 minutes later. This final test assessed (1) content that initially received retrieval practice, (2) content that was related to the content that received retrieval practice (i.e., untested, related content), and (3) unrelated, untested content. Performance on the final test was higher for content that underwent retrieval practice compared to unrelated, untested content – consistent with research showing the benefits of retrieval practice for learning. For the participants who studied the information as a series of facts, final test performance was lower for related, untested content compared to unrelated, untested content – a retrieval-induced forgetting effect. Encouragingly, participants who studied the information in a coherent passage did not forget the related, untested information. More encouraging still, participants who studied the information in a coherent passage and received the final test after a 24-hour delay demonstrated a retrieval-induced facilitation effect for related, untested content – final test performance was higher for both initially tested content and untested, related content compared to unrelated, untested content. These findings suggest that presenting content as a coherent, integrated passage can protect related, untested content from being forgotten and at times even extend the benefits of retrieval practice to related, untested content. Furthermore, whereas the cost of retrieval practice (i.e., retrieval-induced forgetting) tend to disappear with a delay, the benefits are more long-lasting (5).
Retrieval-induced forgetting can be eliminated and even reversed when content is integrated and when delays exist between retrieval practice and a final test. However, these conditions may be difficult to achieve in educational settings, where students study in non-integrative ways (e.g., using flashcards) and engage in cramming (thus, practice retrieving content immediately before a test). So, what can educators do to minimize forgetting and foster facilitation of untested, related content during retrieval practice? Here are my three recommendations, based on learning science research (6).
Encourage students to integrate their learning material.
Students who are able to integrate to-be-learned content with knowledge they already possess seem to be protected from retrieval-induced forgetting (e.g., 7). One effective way to encourage integration (beyond presenting content as a coherent passage) is to have students elaborate on the material they are learning during testing; for example, by having students generate examples from their own lives and connect the new content to what they already know (8).
Use multiple-choice tests with competitive alternatives.
Multiple-choice tests can help protect related content from forgetting because they lead students to consider all of the alternative answer choices in an attempt to eliminate competitive but incorrect responses (e.g., 9,10). Consider what a student might do when answering the following multiple-choice question:
What process is used when a cell needs to take in a substance that is higher in concentration inside the cell than outside and requires the cell to use energy to complete this process?
A. Passive transport
B. Active transport
In order to arrive at the correct response (active transport), the student might retrieve information about the incorrect alternatives, for example, that passive transport, osmosis, and diffusion do not require cellular energy, or that diffusion involves movement of particles from regions of high concentration to regions of low concentration. By engaging in retrieval practice for information associated with both the correct and incorrect answer choices, future recall for these concepts is enhanced. (See this guest blog post by Dr. Andrew Butler and this weekly digest for helpful tips on designing multiple-choice tests for your own classes.)
Offer students a chance to restudy content after retrieval practice.
When students are given an opportunity to review untested content following retrieval practice, forgetting of related, untested content can be eliminated (11, 12). In fact, Storm, Bjork, and Bjork (2008) found that material that was subjected to retrieval-induced forgetting actually benefitted more from restudying compared to material that was not subjected to retrieval-induced forgetting (11). Thus, incorporating retrieval practice and additional review into regular classwork can benefit learning of both tested and untested content. In the classroom, this can be implemented by reviewing the correct answers to low-stakes quizzes after administering them, and connecting the tested content to related content during review.
Bottom line: When the semester comes to a close, we hope our students will remember more than just the content that was tested in the classroom. Fortunately, integrating content during learning, using multiple-choice quizzes, and following retrieval practice with restudy opportunities can promote a win-win situation for both tested and untested content!
(1) Anderson, M. C., Bjork, R. A., & Bjork, E. L. (1994). Remembering can cause forgetting: Retrieval dynamics in long-term memory. Journal of Experimental Psychology: Learning, Memory, and Cognition, 20, 1063-1087.
(2) Macrae, C. N., & MacLeod, M. D. (1999). On recollections lost: When practice makes imperfect. Journal of Personality and Social Psychology, 77, 463-473.
(3) Chan, J. C., McDermott, K. B., & Roediger III, H. L. (2006). Retrieval-induced facilitation: Initially nontested material can benefit from prior testing of related material. Journal of Experimental Psychology: General, 135, 553-571.
(4) Chan, J. C. (2009). When does retrieval induce forgetting and when does it induce facilitation? Implications for retrieval inhibition, testing effect, and text processing. Journal of Memory and Language, 61, 153-170.
(5) Chan, J. C. (2010). Long-term effects of testing on the recall of nontested materials. Memory, 18, 49-57.
(6) Storm, B. C., Angello, G., Buchli, D. R., Koppel, R. H., Little, J. L., & Nestojko, J. F. (2015). A review of retrieval-induced forgetting in the contexts of learning, eyewitness memory, social cognition, autobiographical memory, and creative cognition. In Psychology of Learning and Motivation (Vol. 62, pp. 141-194). Academic Press.
(7) Carroll, M., Campbell-Ratcliffe, J., Murnane, H., & Perfect, T. (2007). Retrieval-induced forgetting in educational contexts: Monitoring, expertise, text integration, and test format. European Journal of Cognitive Psychology, 19, 580-606.
(8) Endres, T., Carpenter, S., Martin, A., & Renkl, A. (2017). Enhancing learning by retrieval: Enriching free recall with elaborative prompting. Learning and Instruction, 49, 13-20.
(9) McDaniel, M. A., Thomas, R. C., Agarwal, P. K., McDermott, K. B., & Roediger, H. L. (2013). Quizzing in middle‐school science: Successful transfer performance on classroom exams. Applied Cognitive Psychology, 27, 360-372.
(10) Bjork, E. L., Little, J. L., & Storm, B. C. (2014). Multiple-choice testing as a desirable difficulty in the classroom. Journal of Applied Research in Memory and Cognition, 3, 165-170.
(11) Storm, B. C., Bjork, E. L., & Bjork, R. A. (2008). Accelerated relearning after retrieval-induced forgetting: The benefit of being forgotten. Journal of Experimental Psychology: Learning, Memory, and Cognition, 34, 230-236.
(12) Little, J. L., Storm, B. C., & Bjork, E. L. (2011). The costs and benefits of testing text materials. Memory, 19, 346-359.