How to Make it Stick: The Psychology of Learning and Memory

Apologies to my readers for the long delay. I recently started teaching a pre-calculus class at a private school in Brooklyn, through which I’m putting into practice many of the pedagogical theories and methods that I’ve observed and studied over the last few years; the process is exciting, but the extra planning required to implement so many new ideas takes up most of the time that I previously devoted to this blog. Further retarding my posting schedule, the post that I’ve been working on intermittently for the past few weeks defies my best efforts to tie it down to finitude and linearity. In the interim, I wanted to post a couple links & a “brief” discussion.*

The Research

My friend Sam Gershman, who studies the neuroscience of memory was thoughtful enough to pass on to me a bunch of cognitive-psych research (see links below) on the conditions that facilitate long-term learning retention. The broad principle around which this research coheres is a theory, strongly supported by experimental data, about the relationship between long- and short-term memory: learning conditions that facilitate short-term recall hamper long-term storage, and vice versa.

The relationship isn’t as counter-intuitive as it first sounds. The more you’re forced to struggle to recall a piece of information, the more likely you are to remember it weeks, months, or years later. To take a familiar example, studying a small set of facts for a long time at a stretch leads to good short-term recall but poor long-term retention—thus, cramming for tomorrow morning’s exam is an effective strategy for improving performance on the exam but a lousy one for mastering the course material. Intermittent studying, by contrast, is not as efficient at improving short-term recall, but it is much better for building retention. (The term “drilling” seems an unfortunate one, in light of this research—it implies, erroneously, that heavy drilling will drive facts deep in a student’s head, whereas the opposite appears to be the case.)

Researchers have identified several other so-called “desirable difficulties”—conditions that impede short-term recall but improve long-term retention: varying the context (location, time of day, etc.) and even formatting (as in page layout) of study sessions and materials; interleaving different topics and different types of study questions so that different pieces of knowledge are constantly interfering with one another; and having students attempt to perform procedure without first seeing them modeled by a teacher. All of these conditions force students to work harder to recall information or carry out procedures, thus improving retention. Perhaps the simplest result of this principle is the fact that practice tests are more effective than studying sessions.

Of course, the retrieval process must ultimately succeed. If conditions are made so difficult that the student is entirely unable to recall information from one study sessions to the next, then the benefit of these desirable difficulties diminishes or disappears entirely. The educator’s task, it seems, is to create a balance wherein students will struggle to recall facts, but ultimately succeed in recalling them.

One interesting side-point: according to some of the articles linked below, learning retention is actually a type of learning transfer. Learning transfer refers to the ability of students to take information and skills learned in one context and apply them to another. Obviously that’s a big issue in education, and one that’s of particular concern in inner-city schools, where kids often have trouble transferring knowledge. Long-term retention, according to these researchers, is not so much a matter of keeping memories from decaying (they won’t) as making the memories accessible at a later date—which is largely an issue of making the knowledge transferable to the new physical, psychological, and contextual conditions under which it will appear at that later date. Thus, instructional methods that improve long-term retention are precisely those that increase learning transfer.


Implications for Schools

These results may sound familiar to anyone who has acquired good study strategies, but they’re not obvious to everyone, and they run counter to many standard instructional practices—especially those used in a number of inner-city charters that I’ve observed. As discussed in the first link below, instructors are often seduced by strategies that reduce long-term retention precisely because they improve short-term performance: if the students recall the entire lesson by the end of the period, we think, then they are more likely to recall some of it tomorrow—but in fact, the opposite is the case. The more difficult the classroom conditions, the more tempting is this fallacy. In the inner city, where students struggle with the most basic facts, where retention is leaky, confidence weak, and mastery rare, teachers are inclined to do everything they can to get kids to answer questions correctly. The long-term detriments of such strategies may seem obvious to some external observers, but it’s difficult to maintain that perspective in a classroom full of eighth graders who are still counting on their fingers.

High pressure testing creates further incentives for instructional methods that favor recall over retention. In order to better prepare students for state exams, for example, many schools design classroom materials to mimic the formatting, phrasing, and problem types found on those exams. More generally, the knowledge that the measure of one’s teaching is the result of an exam, whether one approves of that measure or not, biases one towards strategies that seem to produce a maximum of correct answers, though precisely that preponderance of correct answers increases knowledge loss from day to day, week to week, and year to year.

The relevance of this research is not limited to inner-city education. Last week, after reading the articles below, I altered the materials I had created for my pre-calc class. Though my materials already contained many conditions that would challenge students’ recall and application of skills and strategies on more complex tasks, for facts that I wanted them simply to memorize (the sines and cosines of common angles, for example), I had created materials that were intentionally repetitious and uniform, in order, I thought, to accelerate memorization. Based on this research, I altered that approach.

It is this instantaneous movement from theory to practice that makes teaching this class so gratifying, and it is for this that I allow it to siphon time from my beloved Dewey to Delpit.


The Links

“Assessing Our Own Competence: Heuristics and Illusions,”
Robert A. Bjork
A literature review focused on learning conditions that impair people’s ability to judge their own level of competence and knowledge, this also contains a lot of information about conditions that improve recall at the expense of performance; there is naturally, a lot of overlap between the two types of conditions.

Spacing Effects in Learning: A Temporal Ridgeline of Optimal Retention
Nicholas J. Cepeda, Edward Vul, Doug Rohrer, John T. Wixted, and Harold Pashler
A study seeking to determine the optimal time interval between study sessions, as a function of the desired length of storage time—i.e. whether you want to retain information for a week, a month, a year, etc. helps to determine the optimal study schedule.

“The Critical Role of Retrieval Practice in Long-Term Retention”
Henry L. Roediger III and Andrew C. Butler
A study showing that, with regard to long-term retention, repeated testing is more effective than repeated studying—even if you don’t provide feedback on the test, surprisingly.

“Recent Research on Human Learning Challenges Conventional Instructional Strategies”
Doug Rohrer and Harold Pashler
A lit review geared specifically towards educators. I actually haven’t read through it, as it reiterates a lot of the conclusions of the other articles linked here, but if anyone wants further reading, have at it.


* ^ This is how I always describe these kinds of posts to myself before I actually write them.