Not All Practice Hours are Created Equal: Deliberate Practice Versus Naïve and Structured Practice

 
By K. Anders Ericsson, PhD and Robert Pool, PhD

As the authors of a book about deliberate practice, we are delighted to see that the term has gotten to be so popular. Everywhere you look, it seems, you find “deliberate practice” this and “deliberate practice” that, and the idea that deliberate practice is the key to excellence in almost any field is now widely known.

At the same time, though, we think it is crucial that we remind people about our criteria for deliberate practice since some important details are getting lost. Too often we see people lumping all sorts of practice activities together and calling it “deliberate practice” when it really is not, or expecting other types of practice to have the same results as deliberate practice. And this is not just in popular accounts but, unfortunately, in the scientific literature as well. 

This sort of confusion creates several problems. It keeps individuals from getting the maximum benefit out of their practice hours, and it leads some scientists to make questionable conclusions about deliberate practice.  So it is worth taking a closer look at why not all practice hours are created equal.

Deliberate practice is a very specific sort of practice that was first identified in a study by one of us (Ericsson), of a group of violin students at a prestigious music academy in Berlin. A key finding of that study was that the very best students in their program for solo violinists had put in significantly more hours of violin practice in their lives than other students in that program. These two groups had put in significantly more hours of practice than the students in another program with lower admission standards (who were still very good violinists, just not as good as the best among their peers). 

But it was not just any sort of practice that these students had engaged in. It was a very specific type of practice that Ericsson and his co-authors, Ralf Krampe and Clemens Tesch-Römer, named “deliberate practice.” As they reported, the students—most of whom were destined to become professional violinists—had started their training at an early age, often well before 8 years old. From the beginning they were working with a teacher, who guided them to acquire the skills they would need to play the violin at a high level. 

The process is like constructing a house, where one starts by laying the foundation and then there is a recommended sequence of next steps, some of which must be finished before others can be started. In the case of learning to play the violin, a student is taught increasingly complex skills according to a well established curriculum and under the supervision of the teacher. 

Because the teacher has only limited time for instructing each student, typically 1–2 hours per week, music students must learn how to practice effectively by themselves, putting in 10–20 hours between lessons while making gradual improvements. Student need to have a clear idea—a “mental representation”—of what their practice goal sounds like even though they cannot yet produce this sound by themselves on the instrument, and they must be able to hear what it sounds like when they try to play the assigned practice exercise and identify the differences between their current performance and the desired performance. In this way they can make gradual adjustments in their playing until they have mastered that particular technique. Over time the students develop highly refined mental representations that they use to guide both their practice and their performance. A critical role of the teacher is to provide students with new, and steadily more challenging goals, that push them beyond the comfort of their current performance to new and higher levels. 

Now, contrast this with how most people develop a skill—driving a car, for example, or playing golf or performing a new technique in their jobs. Getting started usually involves some introduction from someone else who describes what to do, often by breaking the skill into a number of steps and demonstrating how it is done. Then you try it yourself, carefully thinking about what you’re doing at each step. You make mistakes and try again. Eventually you reach a level where your performance is “good enough.” You can drive without having an accident (most of the time). You can hit a series of golf strokes and eventually get the ball in the hole. You can carry out that new task at work. Then over weeks and months of regularly engaging in this activity your performance gets dramatically smoother and you are making fewer obvious mistakes. Eventually you get to a point where you can perform the activity without explicitly thinking about the detailed steps—that is, the skill has become internalized or automated. 

This general pattern is observed in the acquisition of a very wide range of activities, with practice improving performance immensely in the beginning but subsequent increases in performance becoming smaller. With extended practice individuals reach a relatively stable plateau of highly reliable performance, where they can repeat the same automated performance over and over with very limited need for attention. When we measure their performance as a function of the number of years spent playing golf, teaching, or typing, we don’t see any clear improvements after the first couple of years. There are even examples where the performance gets worse with more experience. For example, the ability of doctors in general practice to listen to heart sounds and detect abnormalities gets worse with the number of decades since graduating from medical school. These doctors encounter these abnormalities so rarely that they forget what they sound like over time. 

We call this approach to learning a skill “naïve practice,” and it is completely different from the deliberate practice that allows an individual to become an international concert violinist or chess grandmaster or Olympic champion gymnast. In particular, with naïve practice once individuals reach an acceptable skill level, most of their “practice” consists of actual performance—going to work and doing their jobs, playing games of golf with friends, etc. By contrast, a music student training to become an elite violinist spends very little time performing in front of an audience and instead engages in deliberate practice nearly all the time. Thus the music student doing deliberate practice steadily improves, while the amateur doing naïve practice reaches a particular level and stays there. This is why the number of years of practice is a good predictor of performance for those who use deliberate practice but not for those who engage in naïve practice.

Naïve practice and deliberate practice can be thought of as two ends of a practice spectrum which runs from the least effective to the most effective. In the middle are a variety of activities specifically created to improve performance—attending lectures, group discussions, practicing with a team or a choir, and so on. These activities are “deliberate” in the sense that they were designed with the goal of improvement through education and training, and they are sometimes confounded with deliberate practice.

This was the case, for example, in a recent scientific paper by Brooke Macnamara, David Hambrick, and Frederick Oswald. They collected data from a number of studies that had measured the total number of hours of “engagement in structured activities created specifically to improve performance in a domain”, but they referred to all those activities as “deliberate practice.” To minimize confusion we will refer to their definition as structured practice. They correlated the accumulated time spent in structured practice and attained performance and found that structured practice predicted around 25% of the variability in attained performance. They concluded that hours of “deliberate practice” (really structured practice) left most of the variability in one’s level of performance unexplained.

The basic problem with their paper is that they claim to be writing about deliberate practice but almost all of the studies they examined were looking at other types of practice that did not meet the criteria for deliberate practice. They included hours of attending lectures as structured practice, for example, but this is not individualized for each student, nor does it allow each student to generate the target behavior with immediate feedback. They also included practice time, when sports teams were playing against each other, where the activity is not individualized by a coach to help athletes improve their personal performance. In their meta-analysis Macnamara and colleagues added up hours spent in any structured practice activity and then looked for a correlation with attained performance. When we examined their analysis we found that if they had restricted their analysis to only those studies that involved actual deliberate practice, they would have seen a much stronger relation between the number of hours engaged of actual deliberate practice and attained performance.

The moral here is that there are many different types of practice and the closer you get to deliberate practice, the more effective that practice will be. For those interested in deliberate practice, the best option is to work with experienced teachers who have successfully developed individualized training and demonstrated that they can dramatically improve individuals’ performance. This is not always possible, however, and when there are no validated teachers available or if you can’t afford to work with one (they are often expensive), it is still possible to use ideas of deliberate practice to guide your training. We call that type of practice “purposeful practice,” and it essentially follows all of the principles of deliberate practice, just without a teacher. Specifically, it entails (1) carrying out exercises designed specifically to help you overcome weaknesses or develop new capabilities and (2) getting immediate feedback on your performance that you use to guide future training. 

Consider, for example, how the best chess players develop their skill: They study the games of chess grandmasters, seeking to predict what move the grandmaster made for each chess position in the game. If they do not come up with the correct move made by the grandmaster, they go back and reexamine the position, looking for what they missed. This idea can be applied in many different areas. To train radiologists, it is possible to assemble a library of X ray images from previous cases where the diagnosis is known; the radiologists in training examine each image, offer a diagnosis, and then compare that with the correct diagnosis; if they get it wrong, they go back and reevaluate. This immediate feedback is key to effective learning.

In any training it is crucial to get outside your comfort zone and push to the point that you are making mistakes. This is the only way to improve. The best way to increase your typing speed, for instance, is to set aside 15–20 minutes each day where you push yourself to type 10–15% faster than your best error-free speed. This challenge will point to letter combinations that you type more slowly than others or are more likely to make a mistake on. By working on these weaknesses you will be able to type them faster and more accurately and increase your overall speed. By doing this over and over again, people have been able to increase their typing speed by 30–40% after months of training.  

If you are going to spend the time and energy practicing at something, make sure your practice conforms as closely as possible to deliberate practice because not all practice hours are created equal.

About the Authors

K. Anders Ericsson, PhD, is presently Conradi Eminent Scholar and Professor of Psychology at Florida State University. With Robert Pool he published Peak: Secrets from the New Science of Expertise in 2016. Currently he studies the measurement of expert performance in domains, such as music, chess, nursing, law enforcement, and sports, and how expert performers attain their superior performance by acquiring complex cognitive mechanisms and physiological adaptations through extended deliberate practice. 

Robert Pool, PhD, is co-author of Peak: Secrets from the New Science of Expertise.