Book Study: The Science of Learning Physics
The Science of Learning Physics: Cognitive Strategies for Improving Instruction by Mestre and Docktor published in 2021 is essential reading for all physics teachers, high school, AP, dual enrollment, and beyond. In this book, the authors clearly describe the relevant research in physics education, unpack its implications for our classrooms, and provide straightforward, actionable interventions for applying the findings to improve the learning of our students.
Teach the way you do research.
The overarching view that undergirds the book is that, as physics teachers, we should teach the way we do research. We shouldn’t try to reinvent the wheel, rather we build on an existing foundation of research, conduct investigations, analyze the data, and iteratively improve our practice through the conclusions we draw. This inquisitive approach can allow us to verify if our teaching is actually working, based on data, rather than just relying on intuition or tradition.
Each of the main chapters is organized into three parts:
What does the research say?
What are the implications of that research?
How do we apply that to our physics classrooms?
I strongly recommend reading it to see their dive into the research, but for this article (and in the podcast) I mostly want to focus on the implications and applications.
Concept Formation
The authors spend some time explaining the constructivist view of learning, which (to oversimplify) is the idea that people actively build knowledge in their brains as they connect it to their existing ideas. All knowledge is built on prior knowledge, and successfully using knowledge relies on it having strong connections to other bits of knowledge.
Our students 14-18 years of existing in the world means they have collected lots of knowledge about how the physical world behaves. However, as you have doubtless observed, a lot of this knowledge is either incorrect, incomplete, or misapplied. Our task, therefore, is to uncover and remedy these misconceptions.
Unfortunately, as the authors describe, misconceptions are extraordinarily hard to remedy. They present possible methods:
Conflict/Resolution: you create a situation in which the erroneous conception fails, creating dissatisfaction with that conception, then provide a more accurate conception that explains the situation.
“Knowledge in pieces”: you make students aware of their misconceptions, show them they are insufficient, provide the new conception, then teach the appropriate ways to apply the new conception.
The conflict/resolution method is less supported, because there is no evidence we can actually fully uproot the misconception; it continues to exist in their memory. The knowledge in pieces approach on the other hand, requires the student to recognize the misconception, which exists in their mind alongside the new conception, and hopefully, alongside knowledge about applicability. To make this work, the students must be repeatedly exposed to the new conception and apply it multiple conceptual experiences.
Regardless of how well they seem to get the new conception, however, students are still very likely to revert back to their misconceptions after some time.
Interventions for Concept Formation
Utilize active learning strategies (see below) that require students to engage with and manipulate the new information.
Elicit students misconceptions early and regularly so they can be addressed.
Expert-Novice Research
To no one’s surprise, experts and novices process and apply knowledge differently. Specifically, experts not only have a large set of knowledge, they have well-organized and well-connected schemas for that knowledge. When approaching a novel problem, they are quick to see the underlying principles at work in the situation. Novices obviously have less knowledge, and that knowledge is less connected and organized, meaning when encountering novel problems, they are more likely to look at the surface features of the problem, leading to the misapplication of the knowledge they do have.
For example, when given a problem with a curved ramp, the novice sees “ramp”, digs into their brain for other times they’ve seen a ramp, then will likely pick an equation they’ve seen in a ramp problem before, like kinematics or Newton’s Second Law, and try to make it work. The expert instead sees that the acceleration will be non-uniform, leading them to use a conservation of energy approach. The novice sees the surface feature (ramp), the expert sees the principle (conservation of energy).
Expertise takes time and effort, and I doubt any of us expect our students to become experts in a single course. However, some of the traits of experts can be modeled and taught in a way that improves students’ problem solving.
To do this requires teachers to be mindful of the “expert’s blind spot”; we often lose sight of all the tacit knowledge we have and apply in a problem that the students do not yet have. We must take great care to be explicitly and observably meta cognitive, so that students can “see under the hood” of how our brains are processing information.
Interventions based on Expert-Novice research
Employ problem categorization, in which students must identify the underlying concepts of a problem before solving it.
Emphasize the importance and interconnection of concepts by assessing it (you get what you measure!), for instance through error analysis problems.
Support knowledge organization, such as through concept maps and flowcharts and deliberately highlighting connections between topics.
Make your tacit knowledge clear when you model problem solving.
Try strategy writing, in which students must describe how they would approach a problem, in sentences, rather than just solving it.
Have students justify their problem solving decisions.
Problem Solving
Students LOVE to plug and chug. See v and x in an equation? Hmmm, let’s try a kinematic equation. Oh wait, that was a work-kinetic energy theorem problem. Oh well, I got a number! Moving on! We all know how this goes.
In order to get students to problem solve better, and more importantly, understand what the physical meaning of their solutions are, we should require modeling and multiple means of representation. This must be taught explicitly! Moreover, if you want students to do it, you have to assess it. You get what you measure. If you do not grade students on thorough, well-organized solutions, you will not get them.
Implications for Problem Solving
Problem solving needs to be explicitly taught and modeled.
You get what you measure, so assess students on good problem solving.
Select your problems carefully to encourage the type of processing you want students to do.
Encourage and model metacognition (such as through strategy writing, explained above)
Active Learning
People will throw down over whether or not we should be lecturing. I will defend the use of some direct instruction to the death, but we know now that lecture alone doesn’t produce long term knowledge, or perhaps more importantly, knowledge that can transfer and be generalized.
The implications and interventions here are pretty straightforward: students must actively DO things to learn things. Students often don’t like this and will be more critical of classes with more active learning DESPITE the fact they perform better and learn more.
Interventions for Active Learning
Have students engage regularly with peers, through think-pair-shares, small group discussions, and structured group work.
Make labs authentic to the nature of science and require students to do more of the investigations, rather than following a “recipe” to get a known outcome.
Use frequent formative assessment that requires students to engage with the material, such as through clickers, forms, verbal response, etc.
Why Students Suck at Studying
Just like how students are resistant to active learning in class, they are resistant to using study strategies that actually work. Students are very often very wrong about what methods of studying work, how well they’re learning, and how well they will perform on assessments. What’s worse, even when they get concrete feedback that their strategies aren’t working (such as a big ol’ F on a midterm), they are unlikely to change strategies. And EVEN WORSE, the poorest performers have the most inaccurate ideas about how well they are doing.
Yikes.
The issue is, students tend to choose passive strategies that build fluency because they feel easy. Rereading, highlighting, and summarizing make them feel familiar with the material, so they get confident, but they can’t actually retrieve the info from their own memories, let alone apply it. Then, they often use subjective measurements, such as how confident they feel, to know when they’re ready.
Active strategies that have desirable difficulties, such as quizzing yourself, writing about what you’re learning, interleaving content, spaced practice, taking practice tests, feel way harder, so students are unlikely to use them, even though they make the knowledge way “stickier” long term. Using these strategies also give objective feedback about how they’re doing: you either got the question right, or you didn’t.
Interventions for Studying
Tell students about the research on studying: passive stuff does not work, active stuff does, and they are likely to not believe you, but they really should.
Give students practice exams and instructions to take them in a realistic exam setting. As you’ll see in the next section, testing along is hugely beneficial for learning, and it gives an objective measure of progress.
Show that you value effective study strategies in the types of assignments you assign, for example by using interleaving on homework, spacing out assignments, etc.
The Testing Effect
This chapter is really straightforward: being tested on information makes you more likely to remember it in the future. Therefore, we should be quizzing students often, and encouraging them to quiz themselves.
Our quizzing should require students to produce the information from memory, rather than relying or just recognition (hello active learning!). It also helps if the quizzes are routine and low stakes.
Interventions using the Testing Effect
Tell students about the testing effect.
Give practice tests.
Quiz students regularly and encourage them to quiz themselves.
Teach students how to use their self-quizzing to identify and remedy their own misunderstandings.
Please Read this Book!
A lot of this stuff is just good teaching; we know to do formative assessment, we know lecture alone doesn’t work, we know students have misconceptions. This books not only outlines all of these good teaching practices, it really goes in depth into why and how these practices work. I strongly encourage every physics teacher to read this and see the research the authors cite, and hopefully test out these strategies in their classrooms.
