Hi, this is Ray.

Confession upfront: somewhere around 2014, I spent about $100 on a fancy "brain training" app subscription, fully convinced that if I just played their little reaction-time games for 15 minutes a day, I'd emerge from the year 30% smarter, faster, and better at remembering where I left my keys. I did this faithfully for about six weeks. I got dramatically better at the specific games. My ability to remember my keys did not improve. My ability to remember names at parties did not improve. My ability to absorb new study material did not improve. I had simply become really good at a particular pattern of clicking colored squares, which is, as far as career skills go, not a marketable one.

This experience left me cynical about memory training in general for years. I assumed it was all snake oil, that the only real way to improve memory was to study more and stop being lazy. Turns out I was half-right and half-wrong, in ways that the research has gotten much clearer about over the past decade. The "brain training" industry IS mostly nonsense. But there's a separate category of memory exercises (older, weirder, less marketable) that genuinely do work, and have been working for, in some cases, more than 2,000 years. The catch is that they don't promise to make you smarter at everything. They promise to make you better at memory specifically. Which is what we wanted in the first place.

Today's newsletter is the honest tour of that landscape. Which memory exercises actually improve your ability to learn things, which don't, and the six I'd actually recommend you start practicing this week. Let's get into it.

From our partners at The Code:

Staying behind on tech trends can be a career killer.

But let’s face it, no one has hours to spare every week trying to stay updated.

That’s why over 200,000 engineers at companies like Google, Meta, and Apple read The Code twice a week.

Here’s why it works:

Join 200,000+ engineers who read The Code to stay ahead of the curve.

Before we get into the techniques, I have to deliver some bad news that I wish someone had delivered to me a decade ago. There's a critical distinction in memory science between two things that sound similar but aren't:

Memory PERFORMANCE: how much you can remember on a specific task using specific techniques.

Memory CAPACITY: the underlying biological "size" of your working memory and overall memory system.

Most memory exercises improve performance dramatically without much affecting capacity. That sounds disappointing, but it isn't. What it means in practice is: you can become functionally MUCH better at remembering things using techniques, even though the underlying hardware doesn't really change. It's like the difference between buying a faster computer and learning to use the computer you already have more efficiently. Both result in faster work. Only the second one is available to you.

A meta-analysis examining working memory training in healthy adults found exactly this pattern. According to the researchers, adaptive working memory training and mindfulness training showed effects on working memory in healthy adults, but the field still struggles with the question of whether these gains transfer to broader cognitive functioning. Translation: the games make you better at the games. Whether that translates to better life-cognition is much less clear.

This is why I want to focus this newsletter on EXERCISES that improve practical memory performance for learning… not on "brain training" promises that you'll become generally smarter. The first category is real and valuable. The second category is mostly marketing. As one review of the field put it, mnemonic strategies improve memory performance but do not target underlying memory processes, and they typically fail to broadly generalize to untrained activities. That's actually fine. Generalization isn't the goal. Better learning IS the goal. The techniques deliver on that.

Let's start with the heavyweight champion of memory techniques. The method of loci is over 2,000 years old. The Greeks and Romans used it. Sherlock Holmes uses it. Every world memory champion uses it. There's a reason it's the technique that has refused to die for millennia: it actually works, dramatically, for almost everyone who learns it.

The technique: you take a place you know intimately (your childhood home, your apartment, your daily commute route), and you "place" the items you want to remember at specific locations along a path through it. Then, to recall, you mentally walk the path and "see" each item where you left it.

Why it works is fascinating. Your brain has dedicated neural circuitry for spatial memory that's vastly more powerful than your memory for arbitrary lists. Spatial memory evolved over millions of years to help you remember where the food is, where the predators live, and how to get home. By piggybacking on that ancient system, you essentially cheat the limited-capacity verbal memory system. As one study comparing memory techniques put it, the successful implementation of the method of loci depends on creating a familiar visuo-spatial mnemonic environment filled with landmarks, imagined navigation whereby locations are paired with to-be-remembered items, and "mentally walking" through the created environment to sequentially retrieve them. Spatial encoding plus visual imagery plus narrative path equals durable memory.

How to practice: pick a familiar place. Walk through it mentally, identifying 10-15 specific spots in a fixed order. Practice walking the path until you can do it from memory. Then start placing things at the spots… a grocery list, vocabulary words, key points from a study session. Make the images vivid, weird, exaggerated. Practice retrieving by walking the path again. The first time you do this, you'll be shocked by how well it works. By the tenth time, it'll feel like cheating.

If your work or studies require you to remember numbers (dates, statistics, formulas, phone numbers), the Major System is the move. It's a phonetic code that converts digits into consonant sounds, which you then turn into words and images.

The basic mapping: 0 = s/z, 1 = t/d, 2 = n, 3 = m, 4 = r, 5 = l, 6 = sh/ch/j, 7 = k/g, 8 = f/v, 9 = p/b. To memorize a number, convert each digit to its consonant, add vowels to make a word, and visualize it. The number 32 becomes "M-N" which becomes "moon." 47 becomes "R-K" which becomes "rock." You memorize the image, recover the consonants, decode back to digits.

This sounds insane. It is, in fact, mildly insane. It also works embarrassingly well once you've learned the base mapping. World memory champions use it to memorize hundred-digit numbers in minutes. You can use it to memorize the year of important historical events, key statistics for an exam, or your friends' phone numbers. The setup cost is high (learning the mapping takes a couple of weeks), but once it's wired in, your number memory becomes substantially better than it was. Permanent skill upgrade. Worth it.

I covered this in a previous newsletter on retention but it bears repeating because it's the most important memory exercise most people aren't doing. Active recall is the practice of trying to retrieve information from memory rather than passively reviewing it.

The drill: after reading something, close the book. Take a blank sheet of paper. Write down everything you remember. Compare. Note the gaps. Try again later.

Why it works has been one of the most robust findings in memory research for the past 100 years. Every time you successfully retrieve a memory, you strengthen the neural pathway to that memory. Recognition (seeing something familiar) does almost nothing for your retention. Recall (producing it from memory) is what builds durable knowledge. The struggle of trying to remember IS the workout.

Practical version: build active recall into every study session. After every 20-30 minutes of input, do 5 minutes of recall. Flashcards count, but only if you actually try to produce the answer before flipping the card. Self-quizzing counts. Teaching the material to an imaginary student counts. Reading the same notes over again does NOT count, even if it feels productive. Productive feeling = bad sign in this context.

Working memory is roughly limited to 4-7 "items" at a time, which is one of the harshest constraints on human learning. The trick that gets around this constraint is chunking: combining smaller items into larger meaningful units. The phone number 4155551234 is 10 items. (415) 555-1234 is 3 items. Same information. Different cognitive load.

Master chess players don't see 32 individual pieces on a board, they see configurations and patterns, each functioning as a single chunk. Expert readers don't process individual letters, they process whole words and phrases. The skill of chunking isn't innate; it develops with deliberate practice in a domain.

The drill: when learning new material, actively look for grouping patterns. Three definitions that share a theme? Group them. Five steps in a process? Combine them into a single labeled unit. Random list of 20 items? Find the 4 categories they fall into. Practice the grouping out loud or on paper. As one review of working memory training noted, strategy-based training that focuses on internal strategies like verbal rehearsal, chunking, grouping, associations, and visualization helps improve memory performance through encoding and retrieval enhancement. Chunking specifically expands what your limited working memory can handle by changing what counts as "one thing." Powerful, undertaught, very learnable.

If you struggle to remember names of people you meet (and almost everyone does) this is the drill that fixes it. Most name-forgetting isn't actually a memory failure. It's an attention failure. You don't remember the name because you didn't really encode it in the first place.

The drill works in three steps. When you meet someone, step one: actually hear and repeat their name out loud immediately ("Nice to meet you, Sarah"). This forces auditory rehearsal. Step two: link their name to a vivid mental image based on a feature of their face, an association with the name, or both. "Sarah with the big smile, big as the sun, like 'Sahara Sarah.'" Step three: use their name once more during the conversation. "What do you do, Sarah?" The triple exposure plus image association locks in the memory in a way that just hearing the name does not.

This sounds elaborate. It takes maybe four extra seconds per introduction. The hit rate jumps from "remember about 10% of names" to "remember about 80%" within a few weeks of practice. Tested on myself. Embarrassingly large effect for such a small intervention.

The single biggest leverage you have over memory is timing. Reviewing material once and never again leads to losing about half of it within a few days. Reviewing material on a spaced schedule leads to keeping almost all of it indefinitely. Same total time investment. Wildly different outcomes.

The drill: when you learn something you want to keep, schedule reviews at 1 day, 3 days, 1 week, 2 weeks, 1 month. Use Anki if you want this automated. Use a paper calendar if you want it manual. The exact intervals are less important than the principle of expanding gaps over time. Each review at the edge of forgetting strengthens the memory more than reviewing while it's still fresh.

The variation worth knowing about: interleaving. Instead of reviewing one topic for an hour and another topic the next, mix them. Review topic A, then topic B, then topic C, then back to A. Feels harder. Produces better long-term retention. Your brain has to actively work to retrieve each topic instead of cruising on contextual cues. Difficulty IS the workout. Easy practice produces weak memories. Slightly painful practice produces strong ones.

A quick rant, because this matters. Most "brain training" apps will not help you. The research on this is now pretty clear. As one major review put it, while these apps make you better at the specific exercises, they typically fail to broadly generalize to untrained activities. You'll get faster at clicking colored squares. You will not get measurably better at remembering anything that matters in your actual life.

Specific things that I'd put in the "skip" pile based on the research: most generic brain training apps with daily mini-games, "improve your IQ" subscriptions, dual n-back unless you specifically want to practice working memory tasks for their own sake, and any course or app that promises "photographic memory" (which is largely considered pseudoscience).

What's worth your time and money: practicing the actual mnemonic techniques above, using them on real material you actually want to remember, and building spaced repetition into your study routine.

Here's the protocol I'd suggest if you're starting from scratch:

Week 1-2: Build one memory palace using a familiar place. Practice placing and retrieving simple lists in it (grocery items, then word lists, then to-do lists). Get comfortable with the basic technique before scaling up.

Week 3-4: Add active recall to all your learning. Every study session ends with a blank-paper recall test. Make this a non-negotiable habit.

Week 5-6: Start the Major System for numbers if your work involves numerical material. Otherwise, focus on chunking practice with the material you're already studying.

Week 7-8: Add spaced repetition to your most important learning material. Anki for digital, a paper calendar for analog, whichever you'll actually maintain.

Ongoing: The name-face drill on every introduction. Memory palaces for any longer list you need to remember. Chunking habits applied to all new material.

The practice doesn't need to be huge. 10-15 minutes a day of dedicated technique practice plus integration of the techniques into your normal studying. Six months of this and you will be measurably, noticeably better at remembering things. Not because your brain has changed. Because you've finally learned to use the brain you already had.

Here's what I want you to take from all this. Memory isn't a fixed trait that some people have and others don't. It's a set of skills that can be deliberately trained, and the techniques have been around for thousands of years for the simple reason that they work. The reason most modern people have terrible memories isn't that human memory has degraded. It's that we've outsourced everything to phones, search engines, and AI, and stopped practicing the techniques that previous generations used routinely.

The good news: those techniques are still there, still effective, and still freely available to anyone willing to learn them. You don't need an app subscription. You don't need expensive equipment. You don't need special talent. You need 15 minutes a day, a willingness to feel slightly silly while you mentally place a watermelon on your kitchen counter to remember to buy fruit, and the patience to practice for a few weeks before judging the results.

The Greeks called this the "art of memory." Art, not science, not magic. A trained skill that improves with practice and rewards consistent application. You can have it too. Most of us have just never been taught.

Start with the memory palace. Build one. Use it. Tell me how it goes.

Keep learning (and keep remembering),

Ray