Here’s a burning question: What strategies can actually improve classroom learning? One promising strategy—pretesting—may be the answer for both instructors and students. Certainly, the notion of “evidence-based teaching” is becoming entrenched in the education lexicon, as instructors search for answers. Institutions are spending lots of money, resources, labour, and time to develop websites, offer workshops, and provide consults on effective teaching and learning strategies. Further, instructors are being encouraged to incorporate new strategies into their classrooms. As well, a surge of translational research from cognitive psychology has identified simple strategies that can have positive effects on learning when implemented successfully.
Notable examples of such strategies include things like attempting to recall information from memory rather than simply rereading the answer (the testing effect) or distributing repetitions over time instead of sequencing them one right after the next (the spacing effect). These strategies are “desirably difficult” because they engage students more effortfully in the learning process—they can initially generate more errors, thereby seemingly slowing learning, making them “difficult”; but they are “desirable” because using such strategies typically yields better long-term retention of information. However, the new and promising strategy is pretesting, in which students take a test on material they don’t yet know. Let’s take a look at how this works.
What is the pretesting effect?
Surprisingly, taking a test before learning new information in a subsequent lecture leads to better memory for that information than not taking such a pretest. Ostensibly, this pretesting effect is counterintuitive, as trying to respond to or guess the answer to a question one has not even learned yet would seem to be unproductive. There’s no question that it almost always leads to errors—but it’s these errors that create the opportunity for students to improve their learning. In fact, pretesting has relatively dramatic effects on students’ learning, such as students scoring around 12 percentage points higher on a later exam. Nonetheless, students and instructors surveyed on their attitudes and beliefs about learning from errors indicate that they usually want to avoid them.
How is the pretesting effect measured?
In a typical pretesting study, all students learn information on a topic. However, half the students answer questions on that topic before learning it (pretest condition), whereas the other half do not (control condition). On a final test, whether that happens immediately or later, all students answer questions that assess their memory and understanding of the topic. Some of these questions test for information that was on the pretest (pretested questions) and some test for information not included in the pretest, but still part of the topic (non-pretested questions). Most studies reveal that the students who take the pretest get higher marks on the pretested questions, compared to those students who don’t.
Is the pretesting effect observed in the wild?
We’ve seen evidence of the pretesting effect across diverse materials like text passages and videos, and in several domains, such as history, statistics, and psychology. It also works both in lab settings and in classroom environments. For example, Carpenter and colleagues demonstrated the pretesting benefit in an undergraduate psychology course, revealing that those students who took a daily pretest on that day’s lecture material performed higher on final test questions specifically related to the pretested information.
In another study, de Lima and Jaeger pretested fourth- and fifth-grade students on a science text passage about rockets. The pretest was structured so that students had to fill in some information missing in the text, even though other information was intact. Later, they read the same text again with all the information provided. One week later, they took the final test. The students did much better on questions about the information they filled in during the pretest compared to the other intact information they had read; they also felt more confident about their learning.
Several other researchers have reported similar pretesting benefits in classrooms. In an upcoming paper, Veronica Yan and I also examined the benefits of pretesting on long-term retention of lecture material in two undergraduate psychology courses, through two studies. We asked students to take weekly pretests prior to attending lectures on half of the to-be-taught concepts. In Experiment 1, we made the pretest questions short answer and factual in nature; in contrast, the Experiment 2 pretest questions were multiple choice and conceptual in nature. Our results show that pretesting led to better learning for the pretested concepts—with a 13 percentage point advantage on end-of-week quizzes (Exp 1) and a 12 percentage point advantage on midterm exams (administered up to seven weeks after initial learning; Exp 2).
Based on these findings, the pretesting effect seems robust, even in the face of various factors: the pretest question format; the duration between the pretest and the final test, known as the retention interval; and question type, such as whether it is factual or conceptual. Generally, most classroom studies on the pretesting effect support this impressive learning benefit, at least for pretested information. Ongoing research is also investigating when or how pretests may provide a broader learning benefit for the learning of non-pretested lecture content.
Why does pretesting promote student learning?
Some researchers have theorized that pretesting encourages students to engage in feedback-seeking behaviours. In other words, when they can’t correctly answer the pretest questions, they start looking for ways to “fill in” the particular knowledge gaps. This practice bolsters students’ motivation to learn and makes them more likely to pay attention during the subsequent lectures.
Another way to understand the pretesting benefit is that pretesting affects students’ capacity to pay attention. Specifically, pretesting directs attention to key and potentially testable information, so students notice it more when it does appear in the lecture. To explore this supposition, Lewis and Mensink used eye-tracking technology to investigate students’ attention during reading that followed a pretest. These participants focused more on the pretested information, spending more time re-reading and looking back at relevant sentences; they also recalled about twice as much of the pretested information on a final test compared to participants who didn’t take the pretest.
Pretests may also work because they improve learner’s ability to stay focused during lectures. In one of my studies, we explored students’ mind-wandering while watching a video, which showed that pretesting reduced this mental behaviour so that they did better on a delayed final test. However, the positive impact on “attentional processing” during subsequent lectures does not explain everything. In another one of my studies, we divided participants into three groups: one group took a pretest before learning some concepts; a second group took the same pretest questions, but with the correct answers highlighted; and a third group didn’t engage in any activity before learning the concepts. At the end of the experimental session, all participants took a final test on the taught concepts. The surprising outcome, especially in light of attentional processing theory, was that the first group—those who took the pretest and didn’t see the right answers—did better than the other group that did see the answers. Still, as anticipated, both groups did better than the non-pretest/control group.
The main takeaway from this finding is that pretests encourage deep processing above and beyond that of simply orienting attention to target information. Indeed, all theoretical accounts suggest that pretesting modifies students’ cognitive processes engaged during the subsequent lecture. This shift, in turn, affects students’ ability to remember the lecture content.
What is the recent progress on the pretesting effect?
As noted above, when learners take pretests, their final scores on non-pretested questions may also improve. This impact of enhancing memory for the lecture more generally is perhaps the most striking finding from this pretesting literature. Nevertheless, this begs the question of why some studies only show specific benefits, whereas others show both specific and broader benefits of pretesting.
The structure of pretest questions may be a key moderating factor here, one that can significantly impact the information that subsequently gets attended to and ultimately retained. Most pretesting studies use simple, factual pretest questions. But some research now shows that the better pretesting approach is through integrative questions that encompass broader content and concepts from lecture and textbook material, rather than focusing on discrete or isolated details. This strategy is what improves the overall learning of course material.
Another potential moderating factor is the timing of the non-pretested information within the subsequent lecture. Shana Carpenter and I have recently proposed that pretesting triggers an attentional window that opens during the lecture and closes after the student has identified the pretested information. Any information, including non-pretested information, benefits from being in this window because it is more likely to be processed. We observed that participants better remembered non-pretested information when it came early (i.e., prior to the pretested information) in the lecture as opposed to the end (after the pretested information). This highlights that the presentation order of pretested vs. non-pretested information is an important consideration that contributes to the broader memory benefit associated with the pretesting effect.
How can instructors incorporate pretesting into their teaching?
We still have some ways to go to understand the pretesting dynamic in classroom learning. But by now, there’s no doubt that taking pretests enhances the learning of pretested information (and sometimes non-pretested information). This means that instructors can begin each lecture or any class assignment with a short pretest on the upcoming content. The most efficient way to approach this would be to create questions about key concepts students will encounter in the lecture or to convert learning objectives into pretest questions. We have shown that this latter approach improves how well students do on a delayed final test.
Here are a few other useful pointers for using pretest questions in your classroom: you don’t need to mark or provide answers to the pretest questions; you could take or derive the questions from a study guide or a textbook; the questions themselves can be factual or more complex, but should likely be integrative; questions can be in presented in various ways—on paper, via PowerPoint slide, on a digital blackboard—whatever is convenient for you and your students; and students can use the questions to guide their note-taking.
Pretesting offers a straightforward approach to learning. We must remember that pretesting will typically result in mistakes, but these very mistakes are inherent to the learning process. By embracing mistakes, reflecting on errors, and actively correcting misconceptions, we unlock students’ full potential to improve and deepen their understanding.