Study Skills

What I used to do
When I went to college, I realized that the one thing I never really learned was how to learn.

You'd think with 13 years of school behind me, in that time, someone would have shown me how to properly amass knowledge. That was a lesson I never got. Rather, like most everyone else, I just concocted my own habits, largely by copying everybody else.

So, for many years I went to class and took notes, but never did anything with those notes after the class was over. I rarely even kept the notes. I was operating under the idea that writing notes in class is a good way to learn because just the process of writing things out will implant the knowledge firmly in mind and trap it there. Wrong!

My main mechanism of obtaining information was reading books. I did truly like to read, especially later in high school. That was how I absorbed information, and at least for the level of high school coursework, reading a textbook once over was sufficient to allow me to perform reasonably well. I remember cracking open my AP World History book for the first time a few hours before my AP World History test and just cruising through it. I remember almost nothing of that futile exercise except that I got the highest grade possible on that AP test. So, lesson is, cramming works, but the information you 'learn' - if it can even be learning - is so fleeting. Its in one ear and out the other. Does this sound familiar to anyone?

As a freshman in college, I realized that I would need to amend my methods, but what could I do? I went to class diligently, took notes, reviewed them sometimes. I read the textbook before class and again after. Before big exams, I'd re-read whole chapters and do more problems. I had abandoned the days of cramming, and I was much more attentive to the time frame in which I needed to operate. I was studying as well as anyone could, I convinced myself. My grades were quite good, so I figured this was the best I could do.

The summer after freshman year, I was taking a general chemistry lab course at Columbia and I was struggling to recall much the stuff I learned just a few months earlier in Gen Chem course. Did I really spend all that time and money just to forget most of what I learned? Sure, I remembered the broad concepts and could apply them, but what about the finer details? One might argue that for the fine details, you can go to a book. But that's inefficient, especially the way I was accustomed to doing it, re-reading whole chapters in a futile attempt to get that information to stay put.

There had to be a better way. I needed to learn how to learn.

Learning How to Learn - Cornell Notes

In the summer of 2006, a few weeks before sophomore year was about to begin, I was in the local library scanning the stacks for books on study skills. I had never thought to read about studying before, but I wanted to see what kind of methods others had devised. My goal was to learn for the long-term. How depressing it was to think about the time, money, and effort I would put into the rest of my college career only to lose most of that to forgetting.

I remember the day vividly. I looked on the book shelf and there was a title calling me.

How to Study in College by Walter Pauk (HTSIC)


I immediately went to the nearest Starbucks and devoured the 200-something pages of that book in one sitting. Yes, I'm a huge nerd.

So what was inside that was so wonderful?

A couple of things. Several chapters were dedicated to time management. I had never managed time. Not methodically at least. I would plan in my head for the next few days ahead, but I never ever followed any kind of set schedule. That was the first thing HTSIC told me I needed to do. And so I did. I found the useful program iCal on my computer and began arranging my life.

6 AM: Wake up
6:15 AM: Breakfast
7:00 AM: Study Organic Chemistry
9:00 AM: Go to class
....
6 PM: Free time (yes, I scheduled free time)
7 PM: Gym
8 PM: Dinner
9 PM: Study class notes
11 PM: Sleep

I found this process of planning my day just as liberating as the book said it would be. I didn't need to waste time thinking about what I'd do next, and I didn't need to worry about not having enough time because I could plan well enough in advance to prevent crises.

Did I stay exactly on schedule? Well, I tried. In life one has to be flexible. I often fudged my schedule and maneuvered blocks of time to conform to what I was actually doing. Study sessions would go longer, dinners longer, sleep shorter, gym completely tossed. But on the whole, I was able to stick to the broad schedule +/- an hour or two. So this was really helpful.

But the best and most important thing from that book was not time management. It was Cornell Notes and the Q method.

The author of HTSIC, Walter Pauk, was a study skills instructor at Cornell. Based on evidence from educational psychology in the early post-war era, Pauk came up with the Cornell Notes method.

Here's a summary of what the method consists of.

(1) Take a plain piece of paper and make it look like the one above. That's Cornell Note paper. I generated mine on MS Word and then kept that template for the rest of college. Pretty simple.

(2) In class, take notes in the note taking area. Keep them concise and clear.

(3) Immediately after class (or at your earliest convenience, but no later than 24 hours after the class) take the most important bits from the body of notes and write 2-3 sentences in the summary box.

(4) Questions in the margin. You write questions pertaining to the details in the note box. Check out the example below.

The questions should be as direct and clear as possible. In essence, you're making mini-tests on each page of your notes.

(5) RECITATION. This is the most important part and what makes the Cornell Notes so effective. Cover up the details with a blank piece of paper but leave the question exposed. Proceed to recite aloud the answer to each question. If you get it right, move to the next. If you get it wrong, do it again.

Does this mean you have to talk to yourself? You betcha! It's weird, yah, but the benefits of doing this far outweigh any embarrassment or uneasiness you might get from talking to yourself. Pretend you're teaching your imaginary friend.

(6) Review. After you're done reciting, skim over the notes again, eying the summaries.

(7) Reflect. When you're done reciting and reviewing, you should think about what you've just learned, connecting it with you already know, integrating the information into the conceptual framework you've already built up on related material.

I began using Cornell Notes sophomore year. I used it in class, when reading textbooks (modified method), when reading everything. And it worked so well. I was able to recall myriad details from my notes with ease, and not just in the first few days after learning them. Weeks, months later, the information was there still firmly implanted.

So why is Cornell Notes so effective?

A couple reasons I think. The most important part of all this is the recitation. That is what sets Cornell Notes apart. Devising and answering specific questions immediately after learning new material gives you instant feedback. You know if you know it and you know if you don't. If there isn't some testing mechanism to assess how much we've retained, we can very happyily read our notes or read a book chapter and feel that we've got it all in our minds. But then, when we are pressed to recall specific information, such as on an exam, you realize that you really didn't have the nitty gritty in there. The recitation and questions gets around this problem by pinpointing precisely what you know and what you don't.

The recitation also makes the learning an active rather than a passive process. You're not just receiving knowledge. You're repeating it, as if you were teaching. People often say that the surest way to learn something is to teach it to someone else. When I was doing my recitations, I felt as if I were teaching an imaginary audience.

I also think there is something about reciting aloud that really makes the stuff stick. You add another sense - hearing - to the learning process that solidifies the imprint that new information makes on your mind.

The anticipation of knowing that I'd have to answer questions after reading my notes or a book chapter made me more attentive to the details. I didn't just gaze over text on a page. I read with a purpose, and I think that primed my mind for maximum uptake.

Another major benefit of Cornell Notes is that now you have a systemic way of studying in the future for an exam or whatever. No more re-reading whole book chapters. Or going over all of your notes. With the questions and summaries, you've got all you need. The weekend before an exam, I'd go straight to the questions without looking at the details. In this way, I avoided having to read over all my notes again. Questions I got right I didn't need to review. I only looked at the details to questions I didn't get right. Rinse, wash, repeat.

I aced my exams, was able to recall most all of what I learned, and all this made me enthusiastic to learn more.

That's a lot of work

An objection that people have to studying this way is that it seems like a lot of work. Compared to what? It's a lot of work to read over all your notes a bunch of times because they just won't stick. It's a real time killer to go back an re-read textbook chapters before an exam. And it's even more time consuming to have to go back an relearn 75% of your course material a year or two down the road when you take a more advanced course that presumes you remember what you learned at the lower level.

You've go to put you're time in. Learning isn't easy. Cornell Notes is front loaded. You put the effort in at the beginning, but after you've made your notes, it's easy sailing. You just recite, review, recite, review. And best of all, it sticks for good.

I know I sound like a zealot about this, but I'm just so grateful to have learned how to learn early in college. I have a system that works, that helps me learn for a lifetime. So naturally I want to share it with others.

In Medical School

Given the effort needed to study like this, I was a bit concerned that I couldn't make it work in medical school. But, thankfully, I can report that it has worked remarkably well, just as it did in college. I've tweaked the process a bit to accommodate the fact that all our learning is now digital. I don't use the Cornell Note paper. Instead, I've been annotating PDFs from class directly and then covering the details and reciting. It works like a charm.

However, I have not been able to review as much as I should. In the past, for me the magic number was three times. That was how many times I needed to recite the information before it was firmly embedded in my memory. And, the greater the span of time over which those reviews were spread, the better the retention. That's something I realized from the very beginning. Three reviews in rapid succession (say, in three days), is not as good for long term retention as three reviews spread across say 3 weeks. Which brings me to my next point.

Spaced Repetition

Spaced Repetition is revolutionary study methodology based on a common sense insight: to remember information for the long-term, you need to review that information several times and those reviews need to be spread out at appropriate intervals. If you review several times in a short period (say, 5x in 1 week) you may remember that information for the short term, but not for the long term.

We all know that the key to memorization of facts and information is frequent review. But how frequently? And at what intervals? Are there optimum times to review such that you'll be most efficient with your time and most effective in embedding the material deeply in your long-term memory?

People have thought about this question for quite some time. Educational psychologists, beginning with Hermann Ebbinghaus in the late 1800's, have tried to understand why we forget and how to combat it. An excellent article in Wired Magazine about spaced repetition discusses Ebbinghaus' early experiments and the line of inquiry that has followed since that time.

[H]uman forgetting follows a pattern. We forget exponentially. A graph of our likelihood of getting the correct answer on a quiz sweeps quickly downward over time and then levels off. This pattern has long been known to cognitive psychology, but it has been difficult to put to practical use. It's too complex for us to employ with our naked brains.

So basically, we forget things exponentially. The time to forgetting after each review increases, and if we knew when forgetting was about to happen, we could review right before that time. If you review too early, you don't maximize the the time you get until the next time you need to review. The graph above summarizes the effect of appropriate spacing.

But how do I know when I should review?

That's where Supermemo and a string of spin-off computer programs come in.

SuperMemo is based on the insight that there is an ideal moment to practice what you've learned. Practice too soon and you waste your time. Practice too late and you've forgotten the material and have to relearn it. The right time to practice is just at the moment you're about to forget. Unfortunately, this moment is different for every person and each bit of information. Imagine a pile of thousands of flash cards. Somewhere in this pile are the ones you should be practicing right now. Which are they?

Supermemo was created by a remarkable computer scientist and learning afficionado named Pietr Wozniak. He, like so many of us, was totally distraught by the fact that he studied hard in college and as a graduate student, and he'd learn, but the stuff he learned would rapidly fade from memory after a period of time. He harnessed his frustration to try to find a way to not forget. To that end, he came up with a program that makes an algorithm that predicts when you need to review certain information. Basically, it's a flash card program, but with a twist. When you answer a flashcard, you rate how difficult it was for you to recall the answer. Based on your ratings, the Supermemo algorithm will schedule your next review of the card.

Supermemo has quite a following on the net and with some learning enthusiasts, but it hasn't caught on the way you think it might. This is such a game-changer, you'd think everyone would be all over it. Here is a way to learn things for good. The legions of people that have used Supermemo to learn just about everything - commonly languages, but a slew of other topics - swear by it. I didn't know about Supermemo or spaced repetition until this year when I learned about one of its spin-off programs, Anki.

The beauty of these programs and the spaced repetition method is that you are being maximally efficient. When we make notes or review cards, the reason most of stop looking at them is because we just don't have the time to completely review old stuff while new stuff just keeps coming. So we don't review the old stuff, even though that is what is necessary in order to not forget. Programs like Anki and Supermemo use their algorithms to only present to you cards that need to be reviewed, not every single card in the deck.

For example, let's say I have 100 cards in my deck that I've generated from class notes. Then I add more and more cards as I learn new stuff. So what happens to the 100 old cards? I need to review them, but how can I review 100 cards while adding more and more every day? The key is that you don't review them all on each day. Those 100 cards will be spread out in the future based on how you rated them the first time you used them. The key is daily review so that the cards don't pile up. So a year from now, when Duke 1st year is long over, I might have thousands of cards. I need to still review them, but I won't have time to review 100's of cards a day. Thankfully, I don't need to. Maybe 20 cards will be in my review deck every day, based on how I rated them. I'll easily be able to handle those many cards on a daily basis. If not for the algorithm that informs the spaced repetition, there'd be no way to review all those cards in a timely way and in a way that commits the info to memory.

So what about the Cornell Notes method?

The question-recitation component of Cornell Notes is still very effective. I'm not throwing the baby out with the bath water. I'm convinced that review and recitation is critical for learning. But in the past, I had no methodical way schedule for review. I'd review my notes and do the recitation arbitrarily - maybe 3 times before a test, at no particular interval. Anki and the spaced repetition algorithm will just help me know when to do my review and recitation. This is an enhancement to an already excellent system.

I'm very excited to experience the results of this new and improved study strategy. My goal is to build a permanent foundation of knowledge. I absolutely hate the idea of spending all the time and effort this year learning all this medical knowledge just to forget 75% of it by next year. And then, while doing clerkships, I'll need to constantly backtrack to re-learn everything. If I could just do it right the first time, that wouldn't be necessary.

Also, since I'm going to have at least four years during my PhD when I'll be away from the clinic, I need to have a way to continually review the material I will have learned by the end of second year. With Anki and the spaced repetition algorithm, I will be able to do that easily. How much work is it to study 10-20 cards a day, every day? 10 minutes max. Slow and steady wins the race I say, and that's how I like to study.

Making those cards takes a lot of time, you might be saying. And you're right. Making the cards does take time, but in my mind, it's totally worth it. The workload is front end, but then I can use the cards for years to come. And I can modify them if need be, and I can search them easily with the programs available.

If you don't want to make the cards but you want to reap the benefits, there are options.

What actually got me on the whole spaced repetition bandwagon is a USMLE Step 1 resource that my good friend Theo at U. of Washington sent me the other day.

It's called GunnerTraining (Funny name, I know). GT was founded by some Harvard residents in 2009 and has quickly become a favorite of med students across the country. I count myself as one of them. GT is a web-based knowledge database that uses the principles of spaced repetition to help students study for the Step 1. It's got flashcards with snippets of high yield knowledge.

The process goes like this:

(1) Study some flash cards
(2) Add pre-made review cards to your personal review bank
(3) Do an immediate review of the cards and rate their difficulty
(4) GT then schedules when you next need to review those cards



GT generates a daily schedule for you to study those cards. Just come back when you have review cards. At the same time, you can move forward and study more cards and put those in your bank, repeat the process. So you're review old cards and learning new ones at the same time, and you're doing it in a way that optimizes the use of your time and your ability to remember your knowledge.

The caveat for all of these SR-based programs is that they are for the long term. You could use GT's resources in a month, but you'd lose most of the benefits that SR offers in terms of long term storage. I'm so happy I found this in the beginning of 1st year. I can do a slow and steady review concurrently with my coursework. I can study the same topics so that I'm not only studying for Step 1 but I'm also reinforcing things I'm learning in class. Two years from now, hopefully I'll be a Step 1 machine : )

So that's what I'm up to these days. I'm passionate about finding new ways to learn and about sharing what I learn with others. I hope someone benefits from this post. If you have any of your own methods that you want to share, please go ahead and start a conversation on this post.

Happy Studying.




UPDATE (12/17/11): It's been more than  year since I wrote this post. Some things have changed. Most notably, I no longer use Gunner Training. I think it's a fine service for many people, especially those who don't want to make their own cards. But I find card-making a valuable experience, and Anki, in my opinion, is a far superior platform that can be used offline, on multiple devices. And I can add my own stuff. And did I mention it's free? And I can add media and such and not be worried about copyright issues. I'm not knocking GT, but I just want to be honest about what my studying actually consists of. There is also a level of redundancy too. There aren't enough hours in the day to do everything. My Anki load is demanding enough as it is.


I'm going to write a more thorough update in a couple days, but for now, that's all.


Crush away :)