Case Study 8.1: Marcus Builds His Anki Deck
Month One of Medical School, and the Question of What Actually Works
Marcus entered his first year of medical school three months after failing anatomy in his post-baccalaureate program — a prerequisite course he'd needed to retake before his application could move forward. He'd passed on the second attempt, but barely.
He knew, with the certainty of someone who'd already learned it the hard way, that his study methods had been the problem. Not his intelligence. Not his work ethic — he'd studied for hours daily throughout that failed semester. The problem was method.
He arrived in medical school having done his research. He knew about spaced repetition. He'd downloaded Anki during the summer and made a few test decks. He understood the theory.
What he didn't yet understand was how to make good cards.
The First Two Weeks: The Wrong Approach
During orientation week and the first days of anatomy lecture, Marcus made Anki cards the way most first-time users do: he summarized.
His early cards looked like this:
Front: "Describe the brachial plexus." Back: "The brachial plexus is a network of nerve fibers running from the spine through the neck, into the armpit, and down the arm. It originates from spinal roots C5-T1. It is divided into roots, trunks, divisions, cords, and terminal branches. The five terminal branches are the musculocutaneous nerve, axillary nerve, radial nerve, median nerve, and ulnar nerve."
On the surface, this looks comprehensive. In practice, it was a disaster.
When Marcus attempted to review this card, he faced an unanswerable question: what exactly does "correct" look like? If he remembered four of the five terminal branches, was that right? If he described the structure correctly but forgot the spinal root levels, did he know it?
Worse: the card tested five or six distinct pieces of information simultaneously. A card you can only be "sort of right" on doesn't fit well into a two-value review system (right/wrong). And reviewing it took two minutes minimum.
He also made cards for concepts he hadn't yet understood. There's a term in anatomy — the dorsal scapular nerve's course through the middle scalene muscle — that he added to Anki before understanding it. He reviewed it daily for a week. He could eventually reproduce the words. He had no idea what they meant in any functional sense.
By the end of week two, his deck had 200 cards, his reviews were taking 90 minutes a day, and he was starting to dread opening the app.
The Crisis and the Course Correction
At the end of the second week, Marcus sat down with his anatomy professor during office hours and laid it out: "I'm spending more time on Anki than on everything else. I don't think it's working."
The professor — who had been teaching anatomy for fifteen years and had watched the Anki revolution unfold among his students — asked to see Marcus's deck.
He scrolled through a few cards and nodded. "Your problem is you're trying to test understanding with a recall card. That's not what Anki is for. Anki is for retention of specific facts. You need to understand the content first, then make small cards about the parts you need to retrieve reliably."
He showed Marcus what a well-designed anatomy card looked like:
Front: "What spinal roots form the brachial plexus?" Back: "C5, C6, C7, C8, T1"
One fact. Unambiguous right or wrong. Reviewable in thirty seconds.
Then a second card: Front: "Name the five terminal branches of the brachial plexus." Back: "Musculocutaneous nerve, axillary nerve, radial nerve, median nerve, ulnar nerve"
Then a third: Front: "Which terminal branch of the brachial plexus is most commonly injured in a 'Saturday night palsy'?" Back: "Radial nerve (compressed against the spiral groove of the humerus during prolonged arm hanging over a hard surface)"
Each card was specific. Each tested exactly one piece of information. The clinical card — connecting anatomy to pathology — was particularly valuable because it contextualized the fact within a scenario Marcus would eventually encounter in clinical practice.
That night, Marcus rebuilt his deck from scratch.
The New System
Over the next three days, Marcus deleted 140 of his 200 original cards. Of the surviving 60, he revised most to reduce scope.
His new card design principles: 1. One fact per card — never more 2. Understand before adding — if he couldn't explain a concept to himself in plain language, it didn't go in Anki yet 3. Clinical anchors whenever possible — connecting anatomy to a clinical scenario 4. Test in both directions for critical structures — location → name AND name → function → clinical relevance
He also changed when he added cards. Instead of adding cards as he reviewed lecture notes, he began a two-step process: first, understand the material through lecture notes and his anatomy atlas. Then, close the notes and try to write down the key facts from memory (a brief retrieval practice session). Finally, make Anki cards for the things he knew he needed to retain but that had required effort to recall.
This last step meant his Anki cards were being created at the intersection of two signals: "this is important enough to test" and "this is hard enough for me to need repeated practice."
Month One Results
By the end of month one: - Card count: approximately 620 cards (adding roughly 20-25 new cards per day, on average) - Daily review time: 20-25 minutes - Review schedule: consistently, every morning before breakfast
The daily review time was half of what his early sessions had taken, with three times as many cards in the deck. The efficiency gain was entirely from card quality: well-designed single-fact cards review in 20-30 seconds; his original multi-fact summaries had taken 2+ minutes.
His first anatomy practical exam — in which students had to identify structures on cadaveric specimens — placed him in the top third of the class. His written anatomy exam placed him in the top quarter.
In his journal:
"The difference isn't that I studied more. I actually studied fewer hours than in my post-bacc. The difference is that I'm not studying the same things repeatedly just to feel like I know them. I'm studying my gaps, and then I review everything on a schedule that makes sure nothing slips away."
What Month Two and Beyond Looked Like
The power of spaced repetition became more visible over the subsequent months. While classmates who'd crammed for anatomy began to find their anatomy knowledge fading as pharmacology and histology competed for attention, Marcus's anatomy knowledge held.
The mechanism: his daily 20-25 minute review session was continuing to bring anatomy cards up for review at their optimal intervals, maintaining the memories with minimal additional investment. New subjects — pharmacology, histology, biochemistry — each got their own Anki sections and their own daily card additions.
By the end of month three, his daily reviews were taking 35-40 minutes, but they covered three subjects. The time per subject was actually decreasing as his cards matured and review intervals lengthened.
He described the feeling at the end of the first block:
"I'm not afraid of forgetting things anymore. I know that if I review my Anki every day, the material stays accessible. That's a completely different relationship with studying than I used to have. It used to feel like pouring water into a bucket with a hole in it. Now it feels like actually filling the bucket."
Principles Illustrated by This Case Study
- Card quality matters more than review consistency. Perfect reviews of bad cards produce poor results. Good cards reviewed imperfectly produce better results. Start by learning to make good cards.
- Understand before you encode. Anki is a retention system, not a learning system. Cards for ununderstood content are almost worthless.
- The one-fact rule is not a suggestion. Multi-fact cards create ambiguous review outcomes and slow down review sessions. They are a common reason people find Anki frustrating.
- Clinical/application anchors add depth. Connecting facts to scenarios or applications makes them more meaningful, easier to retrieve, and more durable. Always ask: what is this fact for?
- The efficiency of spaced repetition compounds. The real payoff is not week one — it's months later, when knowledge is maintained automatically while new material is being added. The long game is where SRS wins decisively.