*Quick aside: Sometimes this study is used to argue that reading comprehension skills don’t matter. That’s not our interpretation. Rather, this study was set up very carefully to show the power of prior knowledge in particular.
Part of this effect is due to chunking. For folks who know a lot about baseball, they likely can process a description of a double play as one thing. That makes sense to them. For me, a double play would involve a series of distinct events that I process individually, because I know very little about baseball. (What is a double play anyway?) Another classic study looks at how chess masters process chess boards (2). Where I see many distinct pieces on the board (a horse, a castle…), they see something like, “ah, classic… that’s the queen’s gambit”.
Novices, beginners, folks with little prior knowledge literally think about concepts differently than those with prior knowledge. It’s not just that they can build on those connections, but the thinking is qualitatively different. Novices need more working memory to process all those distinct chess pieces, but experts need more complex tasks to feel challenged and reach the desirable difficulty needed for effective learning.
Therefore the strategies that work best for novices differ from those that work best for experts. Novices tend to learn best with guidance – with direct instruction of some kind. They need structure because their working memories are taken up with individual pieces of new information. Experts, on the other hand, learn best when they are left to their own devices and engage in problem-solving or inquiry. They can handle the added challenge, which in turn leads to better learning.
This general idea that what works best for novices is the opposite of what works best for experts is called the expertise reversal effect and explains why scaffolded instruction and gradual release is often very successful in the classroom (3). As students go from novice to expert in an area they will learn most effectively if they go from direct instruction to maybe worked examples or instructor-supported practice to independent, problem-based, or inquiry-based approaches.
