How memory works
How learning works · Published · Evidence: Moderate · About this guide
How memory works, in brief: psychologists separate three processes. Encoding is taking information in while you study; storage is keeping it, made more lasting by consolidation, the slower settling of a memory after you stop studying [1]; retrieval is finding it again, and it changes the memory too [2]. A memory can fail at any of the three, so not recalling something does not necessarily mean it is gone [3].
Memory is not one event. Taking information in, keeping it and getting it back out are three processes that psychologists study separately, and a memory can fail at any of them. Telling them apart explains everyday puzzles: why a name comes back an hour too late, and why, on one influential theory, a little forgetting between study sessions is not wasted time.
What are the three stages of memory?
Psychologists describe three broad processes:
- Encoding: taking information in and making sense of it while you study.
- Consolidation: the slower settling of a new memory after you stop studying, which makes it more lasting [1].
- Retrieval: finding the memory again when you need it.
The processes overlap, though, and retrieval is not a neutral read-out: using a memory changes it, making what you retrieved easier to recall later [2].
Why doesn't seeing something often mean you remember it?
Because seeing something many times does not mean its details were encoded. In a classic study, 20 American adults drew both sides of a one-cent coin from memory, an object they had all seen many times. Of eight key features, the typical person recalled and placed only three. Shown 15 drawings of the coin's head side and told one was correct, only 15 of 36 students in a related test picked the real one as the most likely to be correct [4]. The authors suggested that the details of even a very familiar object are typically available from memory only to the extent that they are useful in everyday life [4].
What gets encoded depends on what you do with the material. People in a series of experiments answered quick questions about how a word was printed, how it sounded or what it meant. On a surprise memory test, words considered for their meaning were remembered better. In one experiment, a fiddly question about a word's pattern of vowels and consonants took about twice as long as a question about meaning, yet led to poorer memory: the kind of processing mattered more than the time spent [5]. Attention matters too. Doing a second task while studying a list of words cut later memory sharply. Doing it while recalling cost memory much less, though recalling still took attention: the second task slowed down more [6].
How much can you hold in mind at once?
Working memory, what you can keep active and work with at one moment, is small. You may have heard "seven, plus or minus two", from a famous 1956 paper [7]. A later review argued that this was a rough estimate, and that when people are kept from grouping items together, the limit is about three to five "chunks", averaging around four [8]. A chunk is a group of items you treat as one unit, such as a familiar word or a date, and what you already know helps you form them [8]. The review uses two names for this limit, short-term memory in its title and working memory in its conclusion, and places it in what you can hold in the focus of your attention at one time [8].
How much can long-term memory hold?
The small limit of working memory concerns how much you can juggle at once. Long-term memory, by contrast, can hold a great deal. In one study, 14 adults, told to try to remember every detail, looked at pictures of 2,500 objects over about five and a half hours. Afterwards, shown two pictures at a time, they picked the one they had seen 87–92 % of the time, even when the other showed the same kind of object, or the very same object in a different state or pose [9].
What happens after you stop studying?
The idea that new memories stabilise over time is more than a century old. Consolidation research describes processes that unfold over time, from hormones and brain systems to changes in cells and molecules [1]. Sleep has been identified as a state that favours this consolidation, although how much it helps depends on what was learned and when the sleep comes [10].
If you can't recall something, is it gone?
Not necessarily. In a 1966 experiment, high-school students studied lists of words from named categories. Given the category names as cues, they recalled more words than without them. The authors' reading: many words people failed to recall were still stored, available but not accessible without the right cue [3]. The article on why we forget looks at this distinction, and at what else makes a memory hard to reach.
Relearning points the same way. In Ebbinghaus's experiments on himself around 1880, and in a 2015 replication with a single subject, lists of nonsense syllables learned and then left alone took less time to learn again than they had the first time. This measure is called savings, and it generally shrank as the delay grew [11]. The article on the forgetting curve explains what savings do and do not measure.
What are storage strength and retrieval strength?
Robert and Elizabeth Bjork's theory gives every memory two separate strengths [12]:
- Storage strength: how well learned something is and how well connected to what else you know.
- Retrieval strength: how easily you can reach it right now, given the cues around you.
Their everyday examples [2]: an address where you lived for years is high on both. The address of a shop you first visited this morning is easy to recall today but barely stored. The name of a childhood friend you have lost touch with may be well stored but hard to reach. A lecture you did not understand may be low on both by evening.
Storage strength and retrieval strength: four everyday examples
A concept sketch, not data: Robert and Elizabeth Bjork's four everyday examples, placed by storage strength (how well something is learned and connected to what else you know) and retrieval strength (how easily it comes to mind right now). Nothing in it is measured, and that storage strength is never lost is the theory's assumption, not a finding.
Examples: Bjork & Bjork (2023), pp. 20–21. A sketch drawn for this guide. [2]
The grid as a table
| Storage strength: low | Storage strength: high | |
|---|---|---|
| Retrieval strength: high | The address of a shop you first visited this morning | An address where you lived for years |
| Retrieval strength: low | A lecture you did not understand, by evening | The name of a childhood friend you have lost touch with |
The theory assumes that storage strength, once built, is not lost, while retrieval strength fades as time passes and the cues around you change [2].
Why can a little forgetting help you learn?
The theory assumes that the two strengths interact unevenly. The easier something is to recall when you restudy or retrieve it, the smaller the gain in storage strength. So reviewing what you can still recall effortlessly adds less, and coming back once it has become harder to reach can add more [2]. The theory also holds that retrieving something yourself does more for both strengths than reading it again [2]. This is how the theory explains why spacing out study and testing yourself help long-term learning.
An experiment shows the pattern. Students read prose passages, then either reread them or tried to recall them, without feedback. On a test five minutes later, rereading came out ahead. Two days or a week later, the students who had practised recalling remembered substantially more, even though rereading had made students more confident [13]. The study did not test the Bjorks' theory, but the pattern fits it: what helps in the moment is not necessarily what lasts. A review of this "learning versus performance" distinction concludes that performance during practice is often an unreliable guide to learning, and that people tend to mistake one for the other [14].
What does this mean when you study?
- Work with the meaning. Ask what a new idea means and how it connects to what you know [5].
- Give it your attention. Divided attention costs your memory far more while you study than while you recall [6].
- Judge your learning later. How familiar a page feels today can mislead; trying to recall it after a delay tells you more [2][14].
- Expect some forgetting. According to the theory, it is part of what makes the next session count [2].
What the research does not say
- That everything you learn stays stored for good. "Storage strength is not lost" is an assumption of the Bjorks' theory. They base it partly on older findings that memories which fall out of use become hard to reach but can be relearned faster than they were first learned [2]. Learning is inferred from later performance; it cannot be observed directly [14].
- That deeper processing helps on every test. When a later test asked people to pick out new words that rhymed with the ones they had studied, they picked out rhymes more often for words they had studied for their sound than for words they had studied for their meaning, although studying for meaning still did better on an ordinary recognition test [15]. What helps depends on what you will need the memory for.
- That working memory holds a fixed four items. The three-to-five-chunk estimate comes from lab tasks designed to stop people grouping items [8]. It is not a number to apply to a page of notes.
- That memory has no limit, or that you can recall thousands of details on demand. The 2,500-picture study tested recognition, choosing between two pictures, not recall [9].
- That consolidation happens at one moment, or how much you should sleep. Consolidation unfolds over time [1], and sleep research describes how memory works, not what any one person needs [10]. For sleep or other health concerns, ask a health professional.
- Anything about memory problems. These studies describe typical memory, mostly in students. Worries about your own memory deserve a professional's assessment.
Related articles
-
The forgetting curve
Ebbinghaus's forgetting curve explained: why forgetting is fast, then slow, what his savings data show, and why "70% forgotten in a day" misreads them.
-
Why do we forget what we study?
Why you forget what you study: new learning can disturb fresh memories, similar ones compete, cues change. Forgotten often means out of reach, not gone.
-
Sleep and memory
What studies show about sleep after learning, what an all-nighter costs your memory, and what is known about sleep on the night before an exam.
-
Retrieval practice (active recall)
Retrieval practice, or active recall: why recalling and then checking the answer tends to beat rereading, how large the testing effect is, and its limits.
-
Attention and multitasking while studying
Why splitting attention tends to cost later memory, what laptop studies found for multitaskers and those nearby, and whether handwritten notes beat typing.
-
The illusion of competence: when studying feels better than it works
Familiar is not the same as remembered. Why rereading and highlighting feel effective but rate low in research, and how to check what you really know.
Sources
- [1] McGaugh, J. L. (2000). Memory—a century of consolidation. Science, 287(5451), 248–251. review · doi.org/10.1126/science.287.5451.248 ↩ Back to the text
- [2] Bjork, E. L., & Bjork, R. A. (2023). Introducing desirable difficulties into practice and instruction: Obstacles and opportunities. In C. E. Overson, C. M. Hakala, L. L. Kordonowy, & V. A. Benassi (Eds.), In their own words: What scholars and teachers want you to know about why and how to apply the science of learning in your academic setting (pp. 19–30). Society for the Teaching of Psychology. book chapter · teachpsych.org ↩ Back to the text
- [3] Tulving, E., & Pearlstone, Z. (1966). Availability versus accessibility of information in memory for words. Journal of Verbal Learning and Verbal Behavior, 5(4), 381–391. experiment · doi.org/10.1016/S0022-5371(66)80048-8 ↩ Back to the text
- [4] Nickerson, R. S., & Adams, M. J. (1979). Long-term memory for a common object. Cognitive Psychology, 11(3), 287–307. experiment · doi.org/10.1016/0010-0285(79)90013-6 ↩ Back to the text
- [5] Craik, F. I. M., & Tulving, E. (1975). Depth of processing and the retention of words in episodic memory. Journal of Experimental Psychology: General, 104(3), 268–294. experiment · doi.org/10.1037/0096-3445.104.3.268 ↩ Back to the text
- [6] Craik, F. I. M., Govoni, R., Naveh-Benjamin, M., & Anderson, N. D. (1996). The effects of divided attention on encoding and retrieval processes in human memory. Journal of Experimental Psychology: General, 125(2), 159–180. experiment · doi.org/10.1037/0096-3445.125.2.159 ↩ Back to the text
- [7] Miller, G. A. (1956). The magical number seven, plus or minus two: Some limits on our capacity for processing information. Psychological Review, 63(2), 81–97. review · doi.org/10.1037/h0043158 ↩ Back to the text
- [8] Cowan, N. (2001). The magical number 4 in short-term memory: A reconsideration of mental storage capacity. Behavioral and Brain Sciences, 24(1), 87–114. review · doi.org/10.1017/S0140525X01003922 ↩ Back to the text
- [9] Brady, T. F., Konkle, T., Alvarez, G. A., & Oliva, A. (2008). Visual long-term memory has a massive storage capacity for object details. Proceedings of the National Academy of Sciences of the United States of America, 105(38), 14325–14329. experiment · doi.org/10.1073/pnas.0803390105 ↩ Back to the text
- [10] Diekelmann, S., & Born, J. (2010). The memory function of sleep. Nature Reviews Neuroscience, 11(2), 114–126. review · doi.org/10.1038/nrn2762 ↩ Back to the text
- [11] Murre, J. M. J., & Dros, J. (2015). Replication and analysis of Ebbinghaus' forgetting curve. PLOS ONE, 10(7), e0120644. replication · doi.org/10.1371/journal.pone.0120644 ↩ Back to the text
- [12] Bjork, R. A., & Bjork, E. L. (1992). A new theory of disuse and an old theory of stimulus fluctuation. In A. F. Healy, S. M. Kosslyn, & R. M. Shiffrin (Eds.), From learning processes to cognitive processes: Essays in honor of William K. Estes (Vol. 2, pp. 35–67). Hillsdale, NJ: Erlbaum. book chapter ↩ Back to the text
- [13] Roediger, H. L., III, & Karpicke, J. D. (2006). Test-enhanced learning: Taking memory tests improves long-term retention. Psychological Science, 17(3), 249–255. experiment · doi.org/10.1111/j.1467-9280.2006.01693.x ↩ Back to the text
- [14] Soderstrom, N. C., & Bjork, R. A. (2015). Learning versus performance: An integrative review. Perspectives on Psychological Science, 10(2), 176–199. review · doi.org/10.1177/1745691615569000 ↩ Back to the text
- [15] Morris, C. D., Bransford, J. D., & Franks, J. J. (1977). Levels of processing versus transfer appropriate processing. Journal of Verbal Learning and Verbal Behavior, 16(5), 519–533. experiment · doi.org/10.1016/S0022-5371(77)80016-9 ↩ Back to the text