Deliberate Practice
The summer of 2014 was a busy time for online chatter about deliberate practice, the specific type of practice that is designed to improve performance of a complex skill. In July, the journal Psychological Science published a paper (summary, full text) by Brooke Macnamara and colleagues, arguing that practice, even deliberate practice, plays only a small role in explaining the difference between novice and expert performance. This contradicts the argument, made in an influential 1993 paper by K. Anders Ericsson and popularized in Malcolm Gladwell’s Outliers, that even talented people make it to the top of their fields mainly by spending enough time practicing in a particular way. The new result prompted a flurry of articles and blog posts. For example:
- How Do You Get to Carnegie Hall? Talent (Benedict Carey): despite the headline, a balanced summary of the deliberate practice controversy and how it relates to the never-ending nature vs. nurture debate.
- Here’s The Problem With ‘Practice Makes Perfect’ (Erin Schumaker): one of many articles that takes the paper at face value and doesn’t include an opposing viewpoint.
- Talent versus Practice: Why Are We Still Debating This? (Scott Barry Kaufman): an argument that the nature vs. nurture debaters are too focused on their respective sides of the debate, and that the truth is somewhere in the middle.
- Challenges for the Estimation of an Upper-Bound on Relations between Accumulated Deliberate Practice and the Associated Performance of Novices and Experts (K. Anders Ericsson): a reply from Ericsson himself, which questions whether the studies referred to in the Macnamara paper were sufficiently strict in their definition of deliberate practice.
Choosing Sides
What does this discussion among psychologists mean for people who are interested in getting better at a complex skill? The Macnamara paper argues that deliberate practice does not explain most of the difference in performance between experts and novices. Therefore, there must be other factors that do explain why top performers are better than others. The authors provide examples of potential factors: general intelligence, working memory capacity, and the age at which an individual starts serious training. Can an aspiring expert use this list to get better at a skill? Only if they can control these factors. Starting age is under an individual’s control, but only in a limited sense: unless you have access to a time machine, you can’t decide at age 20 that you’re going to start practicing violin at age 6. Whether general intelligence and working memory capacity can be improved is tied to the nature vs. nurture debate: someone who doubts the efficacy of deliberate practice isn’t likely to also claim that there’s a technique to increase general intelligence, since that technique would essentially be deliberate practice by another name. One could make a similar argument about working memory. There are also competing claims about whether that can be improved.
It you’re keeping up with the latest learning research, you could decide to pick a side:
The mostly nature side, as argued by Macnamara: Innate ability (“talent”) is a prerequisite for high levels of expertise, with deliberate practice helping at the margins for some skills. To become an expert using this approach, you need to pick a skill that you have a natural talent for, preferably one that you started practicing at a young age. You’ll still have to put the practice time in, but you’ll be refining something that you’re already good at.
The mostly nurture side, as argued by Ericsson: Expertise is mostly attained through deliberate practice, with innate ability providing a faster start, easier progression, or in some cases a necessary prerequisite (the canonical example is minimum realistic height for becoming a professional basketball player). To become an expert using this approach, you can pick essentially any skill, but you’ll expect to spend significant time (on the order of thousands of hours) in deliberate practice to reach elite levels of performance.
Getting Better
When listening to scientists discuss their specialty, it can be easy to get caught up in the technical details. Let’s consider a question that the mostly nature and the mostly nurture sides disagree on:
Can any person with average physical and mental abilities reach a world-class level of expertise in any skill by following a practice system designed by an expert instructor for a sufficient length of time?
Even after making the usual caveats, like excluding 5.5-ft NBA hopefuls, this question will generate opinionated responses. But while it’s an interesting scientific topic, it’s often not the question that people really care about when they’re trying to improve their abilities. The more relevant question for this purpose is:
What is the most efficient way for me to get better?
The advantage of this question is that regardless of which side of the nature/nurture debate you’re on, once you choose a skill, the answer is the same: deliberate practice.
An Example: Computer Programming
If you search the Web for the phrase “deliberate practice,” you’ll discover a lot of information on the subject. It’s a popular topic for blog posts and magazine articles. My goal over the coming weeks and months is to take the general principles of deliberate practice and explore how to use them to improve performance in a particular area of expertise, computer programming. In fact, I’m going to narrow it down even more than that, since it’s more effective to practice a specific skill than a general area of expertise. Here’s the skill definition that I’ll be using:
Write correct, efficient, and maintainable code for a software component given well-defined requirements.
Notice that this definition does not include every aspect of a software developer’s job. It includes some characteristics of good software: correctness, efficiency, and maintainability. But it doesn’t mention many other skills that are important to building software: gathering and interpreting requirements, coordinating work with other developers, designing large systems, and so on. This is intentional. Just as developers break down a problem into components that can be assigned to multiple team members, the components of a developer’s job can also be practiced one at a time.
Why use Deliberate Practice?
If you’re doing any kind of programming (or any other skill) on a regular basis, then you’re getting some kind of practice in that skill. However, practice as it is usually carried out leads to a plateau of expertise. When you’re at a low level of skill, almost any kind of practice works. But before long, just repeating the skill doesn’t make you any better at it. For someone with natural talent, the initial plateau may be high, but it’s still lower than that person’s ultimate potential (even in the mostly nature model). Consider a child who is naturally good at math — the ninety-something percentile type. They might breeze through all of their pre-college classes without much studying. But if they continue to pursue the subject, they’ll eventually run into similarly talented students and more challenging topics. This is where they have to make a decision. They could continue to rely on their innate ability, and see how far that takes them. Or they could decide to find out where their real limits are. Deliberate practice is a way to find those limits in an efficient way.
What’s Next?
This is the first in a series of posts exploring how software developers can use deliberate practice techniques to get better, and why they should. In this article, I summarized what two groups of psychologists think about the implications of the deliberate practice process, and why it’s a critical tool for aspiring experts regardless of the state of the debate. Next, I’ll be going into more detail on what it means to write correct, efficient, and maintainable code for a software component given well-defined requirements.
(Image credit: Jim Larrison)