Last week I talked about the nature of short-term, or working memory and how it helps it build expertise. This week, I want to talk a bit more about how that short-term memory builds long-term memory, and what that means for the age-old debate between proponents of “rote memorization” and those of “conceptual understanding.” If you’re interested in all of this, I’d refer you to Breakthrough Leadership in the Digital Age by the brilliant Bror Saxberg (and the far less brilliant Rick Hess).
Long-term memory is where we store the things that we’ve mastered. It’s really long-term memory that defines “who we are"—given that it determines our immediate, instinctive reactions, and our ability to handle complex tasks. Long-term memory is not conscious; we typically don’t even realize what we’ve stored there. Capabilities stored in long-term memory just seem “easy” and “natural.”
Long-term memory gives us instant, fluid, genuine responses to events occurring around us. As Bror so distinctively captures it:
A kindergarten student was watching the news on television with his grandfather when the initial bombing raids on Iraq were announced in 2003. As soon as a map showing the targets in Iraq came on the screen, the little boy turned to his grandfather, and said, "Is that where they're bombing? Are they taking care of the old things?" His grandfather, bemused, asked him what old things he was talking about. "That's the Fertile Crescent—we've been studying it. It's where civilization started—there are many old things there. Grandpa, are they taking care of the old things?"
Repeated exposure to the maps and history of the Fertile Crescent wired the child’s long-term memory to recognize those features and feed them to his working memory, even in an unexpected situation. That’s the point—what’s stored in long-term memory affects how one views the world. By the way, this is the intuition underlying Don Hirsch’s work on “cultural literacy.” The more a student has committed major historical and literary touchstones to long-term memory, the easier it is to make sense of new information and process new connections. Conversely, students who haven’t mastered this knowledge will have a tougher time as they struggle with the limitations of working memory.
Things already stashed in long-term memory seem “simple,” or “obvious,” because our minds can access them without conscious effort. When such mastery is lacking, however, working memory has to work overtime, making everything slower, more frustrating, and error-prone; like running a computer with too many applications open.
Consider how this plays out in a typical classroom. The range of skills that different students have mastered in long-term memory varies widely: at any given point, some students are still learning what nouns are while others are onto the work of crafting paragraphs. If a teacher is teaching how to write a topic sentence for a paragraph, some students will find this a simple question of mastering one new chunk of information while others will be swamped by a confusing mess of words.
As students tackle more complicated tasks and concepts, those who haven’t absorbed key building blocks into long-term memory will find it hard to keep up. Thus, a core challenge for instruction is building long-term memory so as to make it both accessible and useful for working memory. We do this by having students work on tasks and make decisions that require working memory, optimally using repeated feedback and practice to convert that learning into long-term memory.
Understanding the relationship between working and long-term memory helps illustrate the false dichotomy between “conceptual understanding” and “rote memorization.” Learners have to develop certain fast, fluent capabilities if they are to prepare for the next tier of learning—and, in many instances, “rote learning” is an essential tool for building fluency. Students need enough repeated practice and feedback to build long-term memory in order to successfully work towards deeper, more conceptual mastery. Compare the fluency of a teenager typing on a smartphone, steeped in repeated practice, with her less practiced parents, who use such a device in a much more occasional, stumbling manner—as they struggle to “think through their fingers” in this tiny medium.
Ironically—and especially plaguing to advocates of deeper learning and conceptual understanding—experts tend to forget just how much they’ve absorbed into long-term memory and thus take most of their foundational mastery for granted. Research shows, for instance, that when experts train novices, they tend to leave out a large amount of important information—70 percent or more of what’s required to complete a task! Experts have lost track of vast amounts of building block knowledge, because it long ago moved to long-term memory. They haven’t had to think much about these things for years, making it easy to overlook important chunks of their own expertise.
Experts wind up struggling with counterintuitive limitations when it comes to teaching or designing instruction. This is why superstar ballplayers, for whom fluid mastery came more naturally, frequently are less effective as coaches or instructors than are less-gifted former teammates who had to master their craft much more deliberately and haltingly. Those who have had to struggle to achieve mastery are more conscious of what they do, what mistakes learners make, and what kinds of demonstration and practice may help.
When it comes to schooling, most teachers have been successful learners themselves, but, over time, many of the study skills, habits, and academic foundations became invisible to them. One of the values of mentors, tutors, or similar resources is that they can help students focus on skills or knowledge that teachers might take for granted.
That’s plenty for today. I’ll wrap this all up in the next couple posts by talking a bit about what this means for the role of education technology and what it says about the limits of expertise.