How can I use memory techniques to study in medical school?

Our learning strategy in medical school consists of three main pieces: memory techniques, [spaced repetition][10], and practice questions. If you haven't already, we'd recommend watching our [Getting Started][11] and [Refine Your Technique][5] series first. We also strongly encourage you to read ["Do memory palaces hinder learning?"][1] It summarizes our key dos and don'ts re: using memory techniques for learning. Here are medical examples of the techniques in practice: [tetracyclines][12], [acute pyelonephritis][13], [opioid analgesics][14] and [the trigeminal nerve][15]. In [our interview for Luis Angel's podcast][16], we also walk through a few specific anatomy, biochem, and pharm examples.

We use the software [Anki][17] to take notes and to actively recall concepts in a spaced repetition-based manner.

Also see: "Do you write or draw out your mnemonics?", "What does your Anki setup look like?"

How well do memory palaces work in real-time situations?

Cathy: As we usually say, these memory techniques are not the final solution. We always recommend reviewing via spaced repetition, training with questions, or drilling with friends to help integrate real-time problem solving with mental navigation of the memory palaces, which is really just a stepping stone to making the knowledge completely yours.

This practice is what allows you to quickly name the Chinese restaurants in every single town you've lived in. The same for pizza restaurants, Italian food, trendy places... In fact, you probably have no problem mentally jumping around, and could even list the entrees you like to eat at each and describe the look/feel of the restaurants. Similar to training with practice questions or drilling with friends, you've probably had a lot of practice with quickly analyzing the restaurants by situation (time to eat, need to cater a party) and specific criteria (my date is vegetarian, my budget is $5-10, I need something light on the stomach, must be within 5 minutes drive etc.).

Lastly, in my personal experience, for things that I'd like to retrieve quickly under pressure, using a rhyming or phonetic technique for creating the image allows me to verbalize the information faster.

I'm having problems converting medical terms to images. I find myself using lots of people (e.g. an angry-faced man for increased blood pressure, an amputee for diabetes), so my palaces are full of arms, legs, and also things like blood and urine tanks, which gets confusing. What should I do?

Generally I try to pick specific people; e.g. Dr. Cox from Scrubs for hypertension instead of an "angry-faced man." Imagining individuals cues you into their unique presences and personalities. So doing that may help in differentiating people. If you feel like your palaces are filled with people, I'd make a more concerted effort to imagine more objects. But it happens. A solid chunk of my images are people. They're just generally more memorable.

I'd add that you shouldn't necessarily feel that your images need to be "true to life." It may work to simply imagine a pair of lungs, for instance, but strange characters or nonsensical objects are often more interesting. For instance, I use Spongebob to represent the lungs, Hagrid from Harry Potter to represent blood cells, a bike pump to represent dialysis, Dr. Cuddy from House to represent UTI, Ron Swanson to represent the central nervous system, etc. They may seem like trivial connections, but they quickly become ingrained with a few repetitions.

If you're filling up your palaces with things like blood and urine tanks, you're probably "over-memorizing" to some extent. These types of associations are better learned from an intuitive standpoint, rather than brute-memorized with mnemonics. Following our tips for encoding efficiently^ and [choosing loci][5] (e.g. spacing them out) should minimize palace congestion.

^See: "Should I try to encode or 'memorize' everything using memory techniques?"

Also see [Refine Your Technique #3][5] for general tips on creating images.