Bryan Caplan  

Learning and Retention in Medical School

Coming ObamaCare Challenges... Brennan's Libertarianism: W...
Peter Wei, a medical student at Duke, has some interesting thoughts on my post about learning and retention.  Here's Wei, reprinted with his permission:

You're right, there's a substantial literature on learning loss, yet this doesn't seem much lamented and educators don't seem to tackle this specifically as a problem.

You may be aware of spaced repetition software, electronic "flashcards" that exploit this exponential decay.  They essentially predict when you're about to forget each fact, and schedule you to review that fact just when you're about to forget it.  A friend and I in medical school have used this with considerable success in and it's gained some acceptance here at Duke Med; we're now planning to write an instructional ebook for other med students.

One interesting feature of spaced repetition is that while it's great for long-term retention, it's not the best strategy for doing well on short-term exams. If you want to ace a test next week, the time-tested "cram and forget" strategy is much better.  So, if it were more widely known, the adoption of SRS could be a proxy for whether students are studying mainly for short-term credentialing or long-term use.  SRS was first developed for self-study language learners, who simply care about whether or not they're learning.  Med students have an incentive to retain their learning long-term.  But we wouldn't expect, say, employees doing the more pro forma sort of corporate education to find it a good use of their time.

Peter Wei

Comments and Sharing

COMMENTS (12 to date)
Floccina writes:

It is interesting to try to think about how school would be different if it were about education. The long term would be more important. You might have a model were students are tested on what they learned a year ago and refreshed on what they forgot.

When I look at how things actually work, it looks like everyone is acting as if it is all about signaling but few on the supply side admit it.

MingoV writes:

I taught pathology to medical students for eleven years. The courses in medical school can be approached in two different ways. The ideal approach is the pyramid method: the student masters the basic concepts in the early courses to build the foundation layer and then adds concepts from subsequent courses to build more layers. This approach makes it easier to learn advanced topics because the student remembers the underlying fundamentals. This approach also improves the higher levels of learning: application, analysis, synthesis, and evaluation.

Unfortunately, only a minority of medical students use that approach. The alternate approach is memorize and forget. Medical schools make it easy to use this inferior approach by providing simple handouts that include the facts needed to answer the multiple choice exams, by providing previous years' exams, and by guaranteeing that most of the upcoming exam questions will be drawn from the previous exams. This approach addresses only the lowest level of learning: knowledge.

In later years, the medical students who used the memorize and forget approach never master the concepts underlying all of medicine. Instead of understanding diseases and integrating that knowledge with information on the various ways patients present when sick, they memorize the commonest patterns of diseases. When a pattern memorizer encounters a patient that doesn't fit the pattern, the patient is in trouble. The physician will shoe-horn the patient into a wrong pattern or request many diagnostic tests or consults or refer the patient to another physician (who may be equally clueless). The lack of thorough understanding of health and disease underlies the mediocre performance of many physicians.

Thomas DeMeo writes:

Most of us can't retain what we've mastered unless we use it regularly. "Learning" something in school is done as a form of practice and testing. We do a dry run to see if we can master something. Retention is largely irrelevant, and can even be counter productive if we obsess over it. The capacity to relearn efficiently on demand is what is important.

The idea that medical students need to retain more is silly. They need better research skills.

gwern writes:

I feel it's relevant to point out that one of the areas I list studies using spaced repetition in is medical education, largely thanks to the efforts of Kerfoot; besides what I listed, is a good starting point for finding material on the topic 'spaced repetition in medical education'.

jseliger writes:

One interesting feature of spaced repetition is that while it's great for long-term retention, it's not the best strategy for doing well on short-term exams.

Med schools appear to reward both, to some extent—there are short-term tests, but Steps 1, 2, and 3 appear to reward long-term retention.

Still, my default mode when hearing or talking about med school is now to argue that students are better off being nurses or P.A.s, not doctors.

Shiv Gaglani writes:

Interesting discussions and post. I'm a med student at Johns Hopkins and would like to know more about what's helping Duke students - Peter, can you touch base? Thanks.

jseliger writes:

@Shiv Gaglani -- take a look at "Want to Remember Everything You'll Ever Learn?" and Anki, the latter being a cross-platform version of SuperMemo.

Alex writes:

Excellent post Peter! I'm the other classmate who uses Anki at Duke Med. We've had tremendous success with spaced repetition and our hope is that more medical people will adopt it.

@ shiv Check out this link for more on how to use Anki in medical school.

Steve Z writes:

I used (and continue to use) a SRS system to learn a hard foreign language (think ideographs). They really are excellent, and the community decks you can find should count as evidence against Tyler's "great stagnation." If school were about learning, then one would have to re-take exams at spaced intervals to demonstrate recall. The fact that we instead have a cram system is surely weighty evidence for the signaling camp.

Shiv: I'd recommend that you get either Anki or the equivalent and load up one of the pre-fab anatomy decks (or somesuch -- from speaking with doctor friends anatomy is the course that requires the most hard memorization).

MingoV writes:

"We've had tremendous success with spaced repetition and our hope is that more medical people will adopt it."

I certainly hope not. Repetition is a way of memorizing facts. It is not how one understands concepts. If being a physician is all about memorizing facts, then physicians should be replaced by computer expert systems.

gwern writes:

MingoV: actually, that has been studied. Learning facts via spaced repetition fosters underlying learning; see in my article (the link in the OP) the 'Abstraction' section.

Medford writes:

This article discusses the possible benefits of utilizing a spaced repetition software for learning instead of the “cram and forget” method. This new flashcard repetition method appears to be gaining success in the medical field and could see wider spread success in the upcoming future. Being a “cram and forget” student on many occasions, I see this new educational tool being very valuable to lifetime learners and for career oriented coursework. The key would be to implement this style of learning at an early age and probably integrate it with the pyramid style of developmental learning.
This style of learning could see a great increase in popularity with the way students are learning more on computers and less by classroom lecture. The ease and affordability of a spaced repetition software could make it accessible to learners of all ages and subjects. I could see learners sitting at home on the couch learning on their tablets or laptops. This could cause the shift in learning to be more in line with lifelong education rather than passing a short term test which a lot of students today care more about. Currently I see a lot of classmates that use the flashcard system to help them study for tests, but then again this is just for the short term exam, not for lifelong retention. Also, the time it takes for students to have to make these handwritten flash cards can be time consuming, so a prefabricated, smart system would be very beneficial.
This shift in studying could be very effective for students in middle school and high school. These students probably have the greatest focus on passing the test rather than learning for their future. If the makers of this system could adjust it for different subjects and learning levels, I could see this becoming very popular in elementary learning. The only deterrent I see to this style of studying is the daily time it takes to study, which would not be suitable for a “cram and forget” type of studier since they usually wait until the last minute. This system is probably most effective when used daily so it can more accurately adjust to topics you need to review and will most likely forget.
Some of the previous comments mentioned potential drawbacks to this style of studying could be the lack of implementation and understanding the concepts that can build to lifelong learning. This is also the case for "cram and forget" students though. Most learners do not understand the concepts they learned in school until they utilize the knowledge in their respective fields of work. If the students could though retain the information better with the spaced repetition software, I think they would in turn be better physicians and engineers, etc. The potential of a self study and retention system could change the future of learners for the better although I do not see all employers investing in these systems. I see the more hands on learning approach being beneficial to workers already in the field and using the repetition system for refreshing their education. The spaced repetition software demands a greater level of dedication to learning and it will be interesting to see how well it will be accepted across different industries and students.

Comments for this entry have been closed
Return to top