Never Forget What You Learn: 4 Reasons You Should Be Using Anki in Medical School and Beyond

Anki, a spaced repetition software

Anki, a spaced repetition software

Even with well-made memory palaces and tricks for stronger images (e.g., I shoot for ≤3 images per locus to minimize congestion and speed recall), fading images are an inevitable problem. When I first experimented with memory palaces in medical school, this frustrated me to no end. What's the point of doing all this work if I'm going to forget my images in a week?! The solution: spaced repetition—a review strategy that spaces out reviews over progressively longer time intervals. For example, you learn the material on Day 1, review on Day 4, review a week later, then a month later, etc.). The good news is that after 2-3 reviews, the images and ideas tend to stick really well. There's good learning science to back up spaced repetition too. One of the learning literature's seminal papers marked "distributed practice" (spaced repetition, essentially) as a high-efficacy learning strategy. The only other strategy to receive that designation was practice testing (also called "retrieval practice"—essentially, having to bring a fact to mind from memory). An easy way to implement both spaced repetition and practice testing (more concisely, "spaced retrieval practice")? Anki.

Anki refers to a free, flashcard-based spaced repetition software available here. Spaced repetition is a powerful tool for making things stick long-term, and I can't imagine learning without it—even with the aid of memory palaces. As I've said previously, if I had to choose between memory palaces and spaced retrieval practice, I'd choose spaced retrieval practice every time. Spaced retrieval practice without memory palaces beats memory palaces without spaced retrieval practice. Of course, you don't have to pick one or the other. The combo of spaced retrieval practice supplemented with as-needed memory palaces beats both. There are several scientific studies which support this using more general mnemonics, like the keyword method (essentially, image-based mnemonics without memory palaces), but check out one study which used memory palaces combined with spaced retrieval practice here

I take just about all of my notes using Anki (for medical school, language learning, etc.). It’s not perfect, but I think it can be a valuable learning asset for anyone when used correctly. I do, however, accept that different approaches work for different people. That said, based on the literature, I'd argue that almost every learning researcher would recommend SRS electronic flashcards. Similarly, just about every comprehensive "how to score well in medical school"  resource I've come across (see a few below) uses SRS electronic flashcards as the core of its learning system. Despite its rigid structure, I find that Anki helps de-constrain me in many ways. In addition to its automated spaced repetition algorithm, here are the main things I like about Anki:

The Pros...

  1. Active learning, the testing effect, and spaced repetition: As I've described, Anki capitalizes on the testing effect (i.e. the practice testing/retrieval practice strategy mentioned above). Research shows that active recall aids retention much more than passive recall like rereading notes. Its spaced repetition algorithm means it also capitalizes on the benefits of spaced practice. It's hard to overemphasize these two points. Practice testing + spaced repetition = the two most powerful and scientifically validated learning strategies from the last century of cognitive psychology research.

  2. Clear daily goals: I like the structure of Anki, since it theoretically optimizes the amount of time needed to get info into long-term memory. I know exactly what I need to do each day, and I feel a sense of completion when I finish. Reviewing for a half-hour or more each day isn’t always fun, but spaced review is unquestionably important for building long-term understanding. Spaced repetition is the most efficient way to do that. In addition, I prioritize info I've missed while doing question banks (using a "missed" tag), so I know I'm hitting the material I've tripped up on while actually practicing the format in which I'll be evaluated (eg, USMLE Step exams).

  3. Easy method for implementing and cataloging mnemonics: I can easily attach my mnemonic images using an extra field. See this article about my personal Anki setup for more specifics. This makes the integration of memory techniques into note-taking quite seamless. The coupling of memory techniques and spaced repetition makes for a synergistic combination.

  4. Study anywhere (and during otherwise wasted time): Its mobile app enables me to knock out reviews no matter where I am. It's $25 for iOS, pricey for an app but well worth it. Also, because it's all digital, I can easily search and find anything I want to look up. That's huge. 

And a Few Words of Caution...

While Anki is helpful for making things stick, it should come as no surprise that it’s not the end-all-be-all of learning. If you’re learning a language, get out and practice with a fluent speaker. If you're learning medicine, you need to test yourself with question banks like UWorld (or with actual patients on rotations). My recommendation is to make these vignette-based questions the bulk of your study time. Once I've made an initial pass through the material, I aim for a ~4:1 ratio of QBank:Anki. I time myself during Anki reviews and Qbank use to gauge this split. Using Anki as an adjunct is fantastic, but don't get swept up in Anki reviews to the detriment of QBank time. Although research says that practice testing of any kind does lead to transferrable skills, there are many skills necessary for doing well on vignette-based multiple choice exams for which Anki just isn't enough: properly interpreting questions, being able to pick out multiple ways of saying particular facts, understanding both nuances and whole facts you may have missed with your Anki cards, recognizing which details are actually most often tested, etc. Even the best-made Anki decks are no substitutes for these more "real world" experiences. We believe that an optimal learning strategy contains not just book learning/review, not just practical experience, but a mix of the two.

Making effective flashcards is an art in itself, and it can be easy to overdo it. My experience: Be very selective when it comes to creating cards. Do not get sucked into making and reviewing cards for every fact under the sun. You'll soon find it's unsustainable and not particularly effective. I’ve accepted I won’t have every fact stored away in Anki. My preference is to stick to high-yield sources only (eg, First Aid, Pathoma, UWorld, OnlineMedEd, Emma Holliday, etc). If I need to learn something specific for a particular professor or situation, I might make cards and then suspend or delete them afterward. I won't get into too many specific tips here, but for general advice about making effective Anki cards, I've found Yousmle's and Hacking Chinese's tips to be concise and helpful. TLDR: keep your cards simple (don't jam more than 2-3 discrete facts into a card), and focus on mechanisms (eg, explain why pathogenesis x leads to presentation y). For books discussing Anki and SRS approaches for medical students, check out Learning Medicine: An Evidence-Based Guide and Medical School 2.0.

spaced-retrieval-practice-anki-medical-school-learning-memory-palace

You can read more specifics about my personal Anki setup here.

For more on spaced repetition and Anki: Check out First Aid for the USMLE Step 1, the most popular USMLE Step 1 review book for medical students. Cathy and I wrote a section for the book entitled "Learning Strategies," which overviews which learning strategies are supported by research (eg, spaced repetition, practice testing) and which aren’t (eg, rereading, cramming).