BrainFinch
Articles

Memory training: four techniques with evidence behind them

Memory training with direct evidence means techniques for memorizing particular material: spacing your reviews, testing yourself instead of rereading, placing items along a familiar route, and grouping items into chunks. Each makes the material you apply it to easier to recall later. None has been shown to raise memory in general, and neither have memory games.

The game above is Sequence Tap. We use it further down as a place to try one of the four techniques, and we say plainly what practicing it does.

What does spaced repetition do?

Spaced repetition means spreading your reviews over time, and it leaves you remembering more than the same number of reviews done in one sitting. This is one of the oldest results in the study of memory.

Hermann Ebbinghaus reported it in 1885, using himself as the subject. For a series of 12 syllables, 38 repetitions spread over the three preceding days worked as well as 68 repetitions done the day before. He concluded that with any considerable number of repetitions, distributing them over time “is decidedly more advantageous than the massing of them at a single time.”

Cepeda and colleagues (2006) pulled the later research together. Their review covered 839 assessments of spaced practice in 317 experiments from 184 articles. One pattern stood out: the gap that gave the best retention got longer as the time until the final test got longer.

How far apart should the reviews be?

The gap should grow with how long you need to remember. Cepeda and colleagues (2008) taught facts to more than 1,350 people, gave one review after a gap of up to 3.5 months, and tested them up to a year after that. The best gap was about 20 to 40 percent of a one-week test delay, and about 5 to 10 percent of a one-year delay.

We turned those two published ranges into days.

You need the material in Best gap before the review (Cepeda et al., 2008) In days
1 week (7 days) 20 to 40 percent of the delay 1.4 to 2.8
1 year (365 days) 5 to 10 percent of the delay 18 to 37

That study used a single review. For several reviews, a common way to apply the same idea is to let each gap be longer than the last. The schedule below is our own worked example with gaps that double, for something first learned on November 18, 2026. We computed the dates with a short program. It illustrates the pattern, and the studies above do not show that doubling is the best rule.

Review Gap since the last session Days after first learning Date
1 1 day 1 Nov 19, 2026
2 2 days 3 Nov 21, 2026
3 4 days 7 Nov 25, 2026
4 8 days 15 Dec 3, 2026
5 16 days 31 Dec 19, 2026
6 32 days 63 Jan 20, 2027
7 64 days 127 Mar 25, 2027

Seven sessions cover four months.

What is retrieval practice?

Retrieval practice is recalling material from memory instead of looking at it again. On a delayed test it beats rereading, even though rereading feels more effective at the time.

Roediger and Karpicke (2006) had students study prose passages. Some then took one or three free-recall tests with no feedback, and others restudied the passage the same number of times. When the final test came five minutes later, the restudy group did better. When it came two days or one week later, the tested group remembered substantially more. The restudy group had been the more confident of the two.

In practice this means closing the book. Read the page once, put it away, write down everything you can recall, then check what you missed. Spacing and retrieval fit together: each session in the schedule above should be a test.

How does the method of loci work?

The method of loci has you picture each item at a spot along a route you know well, such as the rooms of your home, and then walk the route in your head to recall the items in order.

Dresler and colleagues (2017) tested it on 51 people with no experience of mnemonics. One group practiced the method of loci for 30 minutes a day over 40 days. A second group spent the same time on n-back working memory training, and a third did no training. Everyone memorized lists of 72 concrete nouns before and after. The loci group improved significantly more than both other groups, and the difference was still there four months later.

The n-back group trained for the same 40 days on a working memory task and did not get the loci group’s gain on the word lists.

What is chunking?

Chunking is grouping items so that several are held as one unit. George Miller wrote in 1956 that the span of immediate memory “is about seven items in length”, and that it is limited by the number of items, whatever each item contains. He called the regrouping “recoding” and described how Sidney Smith learned to recode binary digits into larger units and reached a span of about 40 of them.

The ten numbers below are a chunking exercise set. We generated them with a short program that follows the rules of our Digit Flash game (no leading zero, no digit twice in a row), and the program checked that each grouped form joins back into the original number. Cover the first column, read one grouped form once at speaking pace, look away and write the number.

Number First grouping Second grouping
451036 451 036 45 10 36
676230 676 230 67 62 30
4805034 480 5034 4805 034
6198362 619 8362 6198 362
49317630 4931 7630 493 176 30
82849265 8284 9265 828 492 65
145962645 145 962 645 none (three threes is the even split)
239467282 239 467 282 none
2936768046 293 676 8046 293 6768 046
2830501748 283 050 1748 283 0501 748

Try each number both ways and keep the grouping you find easier to say. The article on how a digit span test works has a table of groupings for every length from 6 to 10.

What does practicing a memory game do, and what does it not do?

Practicing a memory game makes you better at that game, and that part is well supported. Gains on other tasks are much less certain. The study by Dresler and colleagues above is one example: 40 days of n-back practice did not give that group the word-list gain the loci group got.

So we make no claim that Sequence Tap improves your memory. Practice gains are mostly specific to the practiced task, and broad transfer is not established. Our longer piece on what brain training research shows goes through the reviews.

The game does give you a quick place to try chunking on positions instead of digits. Tiles light one at a time and you tap them back in order. We read the numbers from the game code: a new player starts with three tiles on a 3 by 3 grid, lit 0.62 seconds apart. Each sequence you finish adds a tile, each miss removes one, and the range is two to ten. From six tiles up the grid is 4 by 4. No tile lights twice.

Chunking here means seeing shapes. Four tiles can be held as one square, or as a line with a hook on the end. Six tiles split well into two groups of three. If your score in Sequence Tap goes up after you start doing this, you have learned a strategy for that game.

Flash Grid asks for positions without order, and Reverse Sequence asks for the same sequences backwards. Both take under a minute a round.

Common questions

What is the most effective memory training technique?

For keeping material over weeks or months, the best supported pair is spacing your reviews and testing yourself instead of rereading. Cepeda and colleagues (2006) reviewed 317 spacing experiments, and Roediger and Karpicke (2006) found that testing beat restudying on recall two days and one week later.

How far apart should spaced repetition reviews be?

It depends on how long you need to remember. Cepeda and colleagues (2008) found the best gap before a single review was about 20 to 40 percent of a one-week delay and about 5 to 10 percent of a one-year delay.

Do memory games train your memory?

People reliably get better at the game they practice. Broad carry-over to other tasks or to everyday life is not established, so a game is practice for that game.

What is chunking in memory?

Chunking is grouping items so that several of them are held as one unit, the way a ten digit phone number is read as three groups. Miller described it in 1956 as recoding.

Sources

  1. Ebbinghaus, H. (1885, English translation 1913). Memory: A Contribution to Experimental Psychology, chapter 8.
  2. Cepeda, N. J., Pashler, H., Vul, E., Wixted, J. T., & Rohrer, D. (2006). Distributed practice in verbal recall tasks: A review and quantitative synthesis. Psychological Bulletin, 132(3), 354-380.
  3. Cepeda, N. J., Vul, E., Rohrer, D., Wixted, J. T., & Pashler, H. (2008). Spacing effects in learning: a temporal ridgeline of optimal retention. Psychological Science, 19(11), 1095-1102.
  4. Roediger, H. L., & Karpicke, J. D. (2006). Test-enhanced learning: taking memory tests improves long-term retention. Psychological Science, 17(3), 249-255.
  5. Dresler, M., et al. (2017). Mnemonic training reshapes brain networks to support superior memory. Neuron, 93(5), 1227-1235.
  6. Miller, G. A. (1956). The magical number seven, plus or minus two: some limits on our capacity for processing information. Psychological Review, 63, 81-97.

Games in this article

Take all 62 games with you

The BrainFinch app plays offline, keeps your streak and reminds you when the daily challenge is ready.

Get it on Google Play