This is part 7 in a series on what we know about how we learn and how this knowledge should inform how we teach. The series is intended for teachers, students, and developers of education technology who want to be more informed about their practice. Parts 1, 2, 3, 4, 5, and 6.
My trial advocacy teacher began the class by putting on an animal mask and bawking like a chicken. Then he asked all of the students to stand up and pretend to be their favorite barnyard animal.
The rest of the class involved fewer chicken dances. During class, we would focus on a particular aspect of trail advocacy — opening statements, cross-examining witnesses, qualifying an expert, submitting evidence. In between class meetings, we had small group sessions. Each group was led by an experienced trial attorney. Every student would practice one part of a trial in front of the group. The attorney would provide feedback to each student. And everything was videotaped. We all had to watch our own — sometimes cringe-worthy — performances and attend individual sessions with the experienced attorney, who would ask us to identify our own mistakes before giving us her take on what we did wrong and how to improve. The class proceeded this way every week, until we tried a mock case in front of a practicing judge and a volunteer jury.
Whether he knew it or not, my teacher’s approach is a fantastic example of implementing the principles of deliberate practice. Each week was focused on a part of the trial advocacy process. And we would have more opportunities to practice each part in the future. The class used mock cases derived from real ones, ensuring that our practice was realistic. We dressed for our roles. We never stopped if someone couldn’t think of what to say or mishandled evidence.
And we received lots of expert feedback from several different experienced attorneys (and a practicing judge and mock jury at the end of the semester). The feedback focused mostly on mistakes in our performance and how to fix those mistakes in the future — how to not lead the witness on direct examination, for example. It was an exhausting, but enormously fulfilling course.
What about the barnyard animals? The aim of that activity was to make the critical feedback we would be receiving in the coming weeks less threatening: we had all made fools in front of each other once, we’d all be looking foolish again — and that’s okay.
What it is.
Deliberate practice is a way of training designed to bring students to high levels of skill efficiently. The idea is to transform novice habits, movements, and ways of thinking into expert habits, movements, and ways of thinking.
But before getting into more detail about what deliberate practice is, it’s helpful to think about what deliberate practice is not.
Deliberate practice is not play.
On summer nights in high school, my friends and I would go out toss the Frisbee around until late at night. We would goof off — try ridiculous kinds of throws and catches — or just chat as we casually flicked the Frisbee to each other. Although we would have liked to get better, we also just were there for the fun of it. We didn’t consider it serious training.
That was play, not practice. Play is an immensely valuable part of learning and development, especially for young children. Deliberate practice, however, is about getting very good at a specific, well-defined skill.
Deliberate practice is not “regular” practice.
The concept of deliberate practice was created to explain an early mystery in expertise research. One person can practice the violin for twenty years and never really get that good at it, while another person can practice the violin for twenty years and reach a professional level. What is it about the professional that distinguishes her from the life-long amateur?
Although there are lots of potential explanations, deliberate practice offers a particularly striking one: it’s not just about the quantity of practice, it’s about the quality of practice. “Regular” practice is about putting in the time. I sit down at the piano, play the pieces I’m working on for an hour, and clock out. I might be practicing things I already know how to do well. I might not pay particular attention to my mistakes. I might have no one to give me feedback about how I’m doing. This is regular practice.
Deliberate practice, by contrast, is practice that resolutely focuses on skill improvement. It consists of several components:
- Linking the student’s practice to how experts perform.
- Providing training situations that are appropriately challenging.
- Providing time for reflection.
- Providing feedback that makes it clear how to improve or what to do next time to make the performance better.
- Providing many opportunities to practice again (with all of the other elements of deliberate practice at play).
The point of all this is to “break through” skill plateaus — times when you’re still practicing, but you’re not getting any better. We all plateau without dedicated effort.
How it works.
Each component of deliberate practice is important for contributing to the overall success of the training regime.
Linking student practice to well-defined aspects of expert performance ensures that the student is actually practicing the right skills — that they’re moving toward expertise. Without this, students might very well get better at doing something, but not necessarily the right thing. You see this sometimes in sports. Occasionally, the guy with the really weird swing ends up batting .350. But more often, that guy just created some bad habits that he has to work really hard to change.
Providing challenging training situations ensures that student practice is efficient. If the student isn’t being challenged, she’s not going to improve very rapidly. Usually, this means focusing on the weakest parts of the student’s performance.
Without feedback about how to improve, students won’t improve. Many domains, such as sports, provide automatic feedback of a sort: the player scored a point or missed a shot. Scoring a point tells the player something like, “you probably did something right with your shot.” Missing a shot means, “you probably did something wrong with your shot.” It’s the same kind of right/wrong feedback that students get when they get their graded test back.
But this feedback alone doesn’t tell you whether you should open your racket more, follow-through less, or be lighter on your feet. It’s the expert who can understand why what happened happened. And provide good advice on what to do differently.
One of the long-term goals of deliberate practice is to make the students better able to train themselves. Time for reflection builds this skill. This is almost a “meta-feedback” step: if I review my own trial advocacy video, try to find my mistakes, and figure out what to do differently next time, I can compare my assessment to an expert’s assessment. In the long-run, I’m beginning to match my thinking towards theirs, which makes me better at assessing my own skill.
Finally, abundant practice opportunities provide the necessary opportunities for long-term skill growth. No one becomes an expert in any domain without lots of practice. Expert brains are particularly structured to meet the demands of the domain, but the brain does not instantly reorganize itself. And it’s not only cognitive benefits that accrue over time. Physical ones do, too. Heart size, capillary growth, and several other physical attributes respond to physical training (of course, many others do not: I can’t grow taller regardless of how much I stretch).
How we know it works.
The concept of deliberate practice rests upon a broad foundation of early research that explored systematic differences between novices and experts.
Everyone can see that some people become experts and others don’t. One popular explanation was that differences in innate ability (talent, IQ, etc.) lead to differences in expertise. Early research in what eventually became “expertise studies” challenged this view. No one is born knowing how to play chess or manage important physics experiments or release exquisite music from the violin. Even those who seem to quickly advance in such domains still require years and years of dedicated practice to perform at expert levels.
One of the most striking studies explored whether professional chess players have better memories than amateur chess players or non-chess players. The answer: not really. Professional chess players don’t have better memories generally, they just have better memories for chess. Playing so much chess changed how their minds understand chess patterns.
The classic approach to exploring deliberate practice is via retrospective study. Researchers identify at least two groups — experts and amateurs. They ensure there’s a reliable way to measure expertise through performance. Then researchers survey and interview members of both groups, carefully looking for differences that would explain why one group only made it to the “skilled amateur” level and another made it to the professional level.
This is how the concept of deliberate practice was originally created. Professionals accumulated many more hours of deliberate practice than skilled amateurs.
Each domain brings a slightly different spin to the idea. Researchers have to figure out how to reliably measure expertise in rugby players or ballet dancers. They have to figure out what kinds of activities have the features of deliberate practice in that domain, too. Deliberate practice in rugby probably involves realistic scrimmaging, detailed analyses of games, tracking performance on specific skills, etc. Deliberate practice in essay writing probably involves a lot of solitary work (writing), along with intensive review sessions with other writers, among other things. The exact nature of deliberate practice depends on the domain.
But there are some common characteristics across domains. Deliberate practice is mentally (and, depending on the skill, physically) exhausting. It’s typically perceived as “effortful”. In solitary domains, like music and spelling, it’s also just not as much fun as other kinds of activities. Part of this, however, seems to be the nature of the activity. In team sports, players report deliberate practice activities as being both effortful and fun.
The upshot of all this research is that deliberate practice over long periods of time (such as years) predicts performance levels pretty well.
A second, more recent approach to deliberate practice research is to compare training environments that incorporate aspects of deliberate practice with those that don’t. This is particularly prevalent in medicine.
Here, again, the major trend of this applied research is that training regimes that incorporate the elements of deliberate practice improve learning outcomes more than their counterparts. Surgeons-in-training become better surgeons when given ample opportunity to practice using simulations through a deliberate practice approach.
There is also some controversy over how much of expert performance can be attributed to deliberate practice (as opposed to factors like variation in working memory or lung capacity). More recent research, for example, has linked basic measures of working memory to small distinctions in performance at very top levels.
This is compounded with measurement issues. In most cases, it’s very hard to accurately measure the amount of deliberate practice. Retrospective studies rely on interviews and questionnaires that ask people to estimate time spent in various practice activities over the course of all of the proceeding years that they have been in the domain. These self-reports introduce some level uncertainty.
But the larger problem is that researchers have imperfect information about how research participants are practicing. This leads to impoverished measures of deliberate practice. For instance, in some cases, deliberate practice is operationally defined as simply “time spent practicing alone”. There are arguments why, in some cases, this might be appropriate. But it’s a far cry from the classic definition I provided above. It’s very easy to include “regular practice” or total study time in measurements of deliberate practice, which dilutes the validity of the findings. The whole point of deliberate practice is that it’s not just any kind of time on task.
Even the harshest critics, however, agree that deliberate practice is necessary to develop expert-level skill. There is simply no way for someone to reach high levels of skill without it.
How to implement it.
What does the research on deliberate practice imply about how we teach school skills — writing, reading, mathematics, science, history? I think there are at least four lessons here:
Understand what experts do.
It’s impossible to guide students towards expert skills if you don’t know what those expert skills are. A lot of times, students can do something that superficially looks like what experts do, but they end up completely lost when actually trying to do what experts do. The student ends up frustrated, because in “practice” she felt like she could do it, but turns out she can’t. The problem, however, is usually caused by the teacher, who didn’t really understand what the expert skills are.
Keith Devlin has written about this in math — students who practice “procedure following” in grade school face a reckoning when asked to actually explore mathematical patterns in upper-level math courses. Carl Wieman talks about this in physics — what students tend to do in science labs is fundamentally different than how professional physicists conduct research.
It can look like “students spent all this time practicing, but never learned what the heck they were doing.” But they never actually got the right kind of practice in the first place.
Carefully structure the feedback students receive.
Deliberate practice requires useful feedback. And the best example is a personal trainer or coach who’s observing, advising, and challenging the student. Even a little bit of that kind of one-on-one attention can be invaluable, but teachers only have so much time. So what are the alternatives?
Peers can provide surprisingly good feedback. And the process of providing feedback is also valuable for learning. There’s also automated feedback — math programs that evaluate a learner’s math mistakes, writing programs that evaluate essays. These can provide low-level feedback that teachers can supplement later.
The key question is: how can the student improve, so that the next time they will be better at the skill? This makes mistakes feel less threatening: they are signposts telling students how to improve.
Provide more practice opportunities.
The kind of feedback discussed above is only really helpful if students get the opportunity to do something again. One way to do this is to let students resubmit work for a grade after receiving feedback. The student is being rewarded for learning from her mistakes, which is exactly what mistakes are for. High-stakes “final projects” or “final exams” (without sufficient practice opportunities beforehand) are not conducive to this kind of iterative process.
Consider the level of challenge.
Finding the right level of challenge is intertwined with understanding the expert skills. If we want a kid to learn to ride a bike, just putting them on a real bike would be challenging, but also terrifying and probably not that helpful. Giving him training wheels reduces the challenge, but also prevents him from practicing the key skill, which is balancing. Giving him a balance bike is a way of reducing the challenge and having him practice the key skill.
Looking for these opportunities — ways of practicing the skills at the appropriate level of challenge — is essential.
Of course, it takes iteration to create truly effective deliberate practice activities. My trial advocacy teacher had refined the course over many years before I took it. And, if he’s still teaching, I bet he’s still making adjustments today.