Learning

Context-Dependent Memory: Study Where You'll Test?

You tend to remember more when the setting at recall matches the setting where you learned. That's context-dependent memory, and it's part of why an old song can bring back a whole summer. But the famous diver study behind the idea failed a 2021 replication, and in real college exams the effect of the room barely shows up. Here's what the research supports, and how to study with it.

Key Takeaways
    • Context works as a retrieval cue: Recall tends to be better when the conditions at retrieval match the conditions at learning. Endel Tulving and Donald Thomson formalized this in 1973 as the encoding specificity principle.
  • The diver study is famous, not settled: Godden and Baddeley (1975) found that divers recalled words best in the environment where they had learned them. A 2021 replication with 16 divers found no sign of that pattern.
  • The average effect is small: A 2001 meta-analysis of 93 effect sizes found d = 0.28, and a 1985 study found no reliable cost in 21 comparisons of college exams taken outside the usual lecture room.
  • Imagining the room can stand in for being there: In a 1979 experiment by Steven Smith, picturing the original learning room just before recall eliminated the cost of being tested somewhere new.
  • Studying in more than one place can help: In a 1978 experiment, students who studied a word list in two different rooms later recalled 24.4 words, against 15.9 for students who studied it twice in the same room.
  • Bring your cues with you: Quiet study, practice tests and notes in your own words travel into any exam room. The room itself doesn't.

What Is Context-Dependent Memory?

Context-dependent memory is the finding that people tend to recall information better when the context at retrieval matches the context at learning. The context can be the physical place, the background sound or your own internal state. When those conditions come back, they act as cues that help pull the memory up.

You've probably felt it. You visit your old school and remember the names of classmates you hadn't thought about in years. You run into a coworker at the supermarket and can't place their face, because you've only ever seen them at work.

Psychologists usually split the idea into a few kinds:

  • Environmental context-dependent memory: the external setting, such as the room, the building or the background noise.
  • State-dependent memory: your physiological state, such as being sober or intoxicated.
  • Mood-dependent memory: your emotional state, such as feeling happy or sad.

The theory behind all three is the encoding specificity principle, which Endel Tulving and Donald Thomson set out in a 1973 paper in Psychological Review: "Specific encoding operations performed on what is perceived determine what is stored, and what is stored determines what retrieval cues are effective in providing access to what is stored."

A memory isn't stored as a bare fact. It's stored with traces of how you took it in: what you were thinking about, where you were, what you could hear. At retrieval, a cue helps to the extent that it matches what was stored. If the room was part of what you stored, the room becomes a cue. If you never noticed the room, it isn't one.

Psychology courses often teach the flip side as context-dependent forgetting, a kind of retrieval failure: the memory is there, but the cue that would reach it is missing.


The Diver Study: Godden and Baddeley (1975)

The experiment psychology textbooks use to illustrate context-dependent memory was run in open water near Oban, on the west coast of Scotland. Godden and Baddeley gave a group of divers lists of words to learn, either on the shore or underwater, and then asked them to recall the words either in the same environment or in the other one. Every diver did all four combinations: learn on land and recall on land, learn on land and recall underwater, and so on.

The result, in the words of their abstract in the British Journal of Psychology: "Lists learnt underwater were best recalled underwater, and vice versa." A second experiment tested an obvious objection, that getting in or out of the water was simply disruptive, and concluded that "the disruption of moving from one environment to the other was unlikely to be responsible for context-dependent memory."

Two later findings complicate the story.

Recognition didn't show the effect. In 1980, Godden and Baddeley ran a follow-up in which divers learned lists of 36 words on land or underwater, then had to pick them out of a list of 72 words in the same environment or the other one. "In contrast to an earlier recall study," they reported, "no trace of context dependency was observed." A common explanation is that a recognition test hands you the word itself, a far stronger cue than any setting, while free recall gives you almost nothing to work with.

A direct replication failed. In 2021, Jaap Murre of the University of Amsterdam published a replication in Royal Society Open Science. Sixteen divers learned lists of 36 two-syllable Dutch words, on land or underwater in a heated indoor pool, and recalled them in each environment. There was no matched-context advantage. Words learned on land were recalled about equally well in both places, and words learned underwater were recalled better on land. "Our results, thus, not only fail to replicate Godden & Baddeley," Murre wrote, "but do not even have a trend in the same direction."

The two studies weren't identical: a heated pool in Amsterdam is a milder change of setting than the open sea off Scotland, and the original took four days of diving where Murre's took one. Still, the best-known demonstration of context-dependent memory rests on one small experiment from 1975, and the one published direct replication didn't find the effect.


State-Dependent and Mood-Dependent Memory

Context doesn't have to be a place. The same logic predicts that your internal state while learning becomes part of the memory, and that getting back into that state should help you recall it.

Alcohol. In a 1969 study in Science, Donald Goodwin and colleagues had male volunteers do four memory tasks either sober or under the effects of alcohol, then tested them 24 hours later in the same state or a different one. On the recall tasks, people who learned while intoxicated did better when they were intoxicated again than when they were sober. Recognition wasn't significantly affected by the change of state. None of this means alcohol helps you learn. It means the state you learned in became part of the retrieval cue.

Mood. In his 1981 paper "Mood and memory," in American Psychologist, Gordon Bower reported mood-dependent recall of word lists, diary entries and childhood experiences in people put into happy or sad moods through hypnosis. It proved hard to repeat. In 1985, Bower and John Mayer published a paper titled "Failure to replicate mood-dependent retrieval," which concluded that "mood-dependent retrieval must be judged to be an unreliable phenomenon in the laboratory."

Music. Background sound can act as context too. In a 1985 study, Steven Smith had 54 students learn 40 words while a Mozart piano concerto, a jazz recording or nothing at all played in the background. Two days later, students tested with the same background recalled 15.75 words on average, against 13.46 for students tested with a different one, an improvement of 17%. Students who had learned in quiet showed no effect of what played at the test. (Whether music helps you learn in the first place is a separate question, covered in our guide to whether music helps you study.)

Chewing gum. A 2004 study by Baker and colleagues in Appetite reported that switching between gum and no gum from learning to recall led to poorer performance, which looked like a neat context effect you could carry in your pocket. It didn't hold up well. A 2012 review by Tucha and Koerts counted two studies that found context-dependent effects of chewing gum and four that "failed to replicate" them, including one in which chewing gum at learning significantly impaired recall.

Type of contextWhat changes between learning and recallClassic studyHow well it holds up
Physical environmentRoom, building, land or waterGodden and Baddeley (1975)Reliable but small on average; the diver study failed a 2021 replication
Background soundMusic, noise or silenceSmith (1985); Grant et al. (1998)Matching helped in these studies; people who learned in quiet were unaffected in Smith's
Physiological stateSober or intoxicatedGoodwin et al. (1969)Shown for recall, not for recognition
MoodHappy or sadBower (1981)Hard to reproduce; Bower and Mayer called it unreliable in 1985
Chewing gumGum or no gumBaker et al. (2004)Four failed replications against two positive studies

How Strong Is the Context-Dependent Memory Effect?

Single studies make memorable stories, but the pooled evidence is less dramatic.

The best summary is a 2001 meta-analysis by Steven Smith and Edward Vela in Psychonomic Bulletin and Review. They pooled 93 effect sizes from 75 studies in 41 articles published between 1935 and 1997. Environmental context effects were reliable: across all the studies, people remembered more when tested in the environment where they had learned. But the average effect was modest, d = 0.28, which counts as small by the usual benchmarks.

Three details from that analysis matter for students:

  1. The effect grew with delay. At delays of a day to a week, the average effect was about 0.63, compared with 0.22 to 0.28 at shorter delays.
  2. Recall and recognition looked alike. Despite the 1980 diver result, recall and recognition tests showed similar average effects across studies (0.29 and 0.27).
  3. The effect shrank when the surroundings stopped mattering. Smith and Vela argued that the more a task absorbs people in their own thoughts, the less their memories depend on the room. Mentally reinstating the original context shrank the effect, and so did absorbing tasks that kept the room from being stored (overshadowing) or used (outshining). In their words, context effects "are less likely to occur under conditions in which the immediate environment is likely to be suppressed."

Some studies found nothing at all. In 1985, Fernandez and Glenberg reported eight experiments under a blunt title, "Changing environmental context does not reliably affect memory." None produced a reliable same-context advantage, including one that repeated a published study that had found the effect, with more than twice as many participants.

The study closest to a student's life came out the same year. Saufley, Otaka and Bavaresco compared college students who took their regular exams in a room other than the lecture room with classmates who stayed in the lecture room. Across 21 comparisons in seven courses, they found "no statistically reliable effects." Their explanation was that "college classes induce students to decontextualize information." When you meet the same material in lectures, in the textbook, in your notes and in problem sets, it stops being tied to any one room.

StudyWhat it testedWhat it found
Godden and Baddeley (1975)Divers learned and recalled words on land and underwaterRecall was best in the matching environment
Godden and Baddeley (1980)The same comparison with a recognition testNo context effect
Saufley, Otaka and Bavaresco (1985)College exams in the lecture room or in a different roomNo reliable difference in 21 comparisons
Fernandez and Glenberg (1985)Eight attempts to find a same-room advantageNone reliable
Smith and Vela (2001)Meta-analysis of 93 effect sizesReliable but small, d = 0.28
Murre (2021)Replication of the diver study in a poolNo matched-context advantage

Should You Study Where You'll Take the Test?

Usually it isn't worth the trouble, and most of the time you can't anyway. The average room effect is small, and the best-known study of real college exams found no reliable cost to taking a test somewhere new. If you can study in the exam room and it's convenient, fine. Don't rearrange your week around it.

The context worth matching is the context you can carry with you.

Study in quiet, because exams are quiet. In a 1998 study titled "Context-dependent memory for meaningful material: Information for students," Grant and colleagues had 40 participants read an article in silence or with background noise, then tested them in silence or with noise. On both short-answer and multiple-choice questions, people did better when the conditions matched. Their practical advice: "students may perform better on exams by studying in silence." Smith's music study points the same way, since the students who learned in quiet weren't affected by what played at the test.

Practice the way you'll be tested. The match that matters most isn't the room but the task. If the exam asks you to write answers from memory, practice writing answers from memory: closed book, timed and on paper if that's how the test runs. Practice testing was one of only two techniques rated "high utility" in a 2013 review of ten common study techniques, and our guide to active recall shows how to do it.

Don't lean on props you can't bring. Your playlist, your desk and your favorite café all stay behind when you walk into the exam. Even gum, which you could bring, has weak evidence as a cue.

Keep your state in mind. The alcohol studies aren't a study tip, but they show that your internal state is part of the context. A reasonable, if less tested, extension: do some of your review alert and timed, the way you'll be at the test, not only late at night and half asleep.


Mental Context Reinstatement: Rebuild the Room in Your Head

If you can't return to where you learned something, you can go back in your imagination, and in one of Smith's experiments that worked about as well as the real thing.

In Steven Smith's 1979 experiments, being tested in a new room cost people some recall. But in his second experiment, the cost disappeared when people tested in the new room were told to recall the original learning room just before trying to remember the words. When Smith made the original room harder to remember, the benefit of imagining it shrank. That's what the encoding specificity principle predicts: imagining the room only helps if the room is part of the memory to begin with.

Smith and Vela's meta-analysis found that mental reinstatement reliably shrank environmental context effects, and the same move is built into the cognitive interview, a method for questioning eyewitnesses. One of its instructions asks witnesses to mentally recreate the physical and personal context of an event before they describe it. A 2010 meta-analysis by Memon and colleagues found that the cognitive interview produced "a large and significant increase in correct details and a small increase in errors."

A 2022 study by Joey Essoe and colleagues at UCLA found the same benefit in virtual reality. Participants learned words from two foreign languages, Swahili and Chinyanja, inside virtual environments. On test trials where they first imagined the environment in which they had learned a word, their average recall score was 52%, against 47% on trials where they imagined the other environment (each answer was scored by the share of the word's sounds it got right).

Here's how to use it in an exam when you blank:

  1. Stop forcing the answer. Put the pen down for a few seconds.
  2. Picture where you studied it. The desk, the light, the screen or the page, the time of day.
  3. Picture the material itself. Where the idea sat on the page, the diagram next to it, the note you wrote in the margin, the example you used.
  4. Walk forward from there. What came before this topic in your notes? What came after it?

The third step is the one most people skip, and it's often the most useful. The page and your own notes are context too, and they're context you created.


Why Studying in Different Places Can Help

If memories lean on context, tying a memory to one context makes it fragile, and tying it to several makes it sturdier.

In a 1978 experiment, Steven Smith, Arthur Glenberg and Robert Bjork had students study a 40-word list twice, three hours apart. One group studied in the same room both times. The other studied in two very different rooms. Three hours later, everyone took a surprise test in a third, neutral room. Students who had studied in two rooms recalled 24.4 words on average, against 15.9 for those who had studied in one. It was a small study (16 students), and more than the room changed between sessions, but the size of the gap made it a classic.

A classroom version followed in 1984. Smith and Ernst Rothkopf taught an 8-hour statistics course as four videotaped lessons, in one room or in four different rooms, in one day or across four days, and tested students five days later in a new room. Spacing the lessons helped, and varying the room improved test performance for field-dependent students (people whose perception leans more on their surroundings).

A 2021 study by Imundo and colleagues, including Elizabeth and Robert Bjork, added a qualification. Participants came back for a second session 48 hours after the first, either in the same setting (same room, same computers, same people nearby, same experimenter) or in a completely different one. For participants who simply restudied, the new setting never hurt, and in the second experiment it nearly doubled final recall (26% against 14%). For participants who tested themselves, the new setting lowered recall on that test, and the cost carried over to the final test. In the first experiment, people tested in the new setting recalled about half as much during the second session (17% against 36%), and words missed then mostly stayed missed.

The VR study by Essoe and colleagues shows a different benefit: less interference. Among participants who felt strongly present in the virtual worlds, those who learned each language in its own environment retained, a week later, an average of 92% of the recall score they'd reached at the end of training, against 76% for those who learned both languages in one environment. The two-environment group also made 38% fewer intrusions, recalling a word from the wrong language.

Put together:

  • Review in more than one place. Reread and review your notes at your desk, on the bus and in the library. Each setting gives the memory another route back and makes it less dependent on any one of them. Our guide to desirable difficulties explains why variety that makes studying feel harder often makes learning last.
  • Give confusable subjects their own places. If two courses keep blurring together, study them in different rooms, or at least at different spots.
  • When you test yourself somewhere new, check your answers right away. A failed attempt in an unfamiliar place can carry over, so make sure every miss ends with the right answer in front of you.
  • Use places on purpose. The memory palace technique turns the same principle into a tool by hanging new information on locations you already know well.

How to Use Context-Dependent Memory With Your Highlights and Notes

Half a century of studies suggests that the room is a weak cue and that the cues you build yourself are stronger ones. Your highlights, margin notes, questions and summaries are all context. They record what you noticed, how you phrased it and what it reminded you of. Unlike a room, they go wherever you go.

A few ways to put that to work:

  • Write the cue, not just the mark. A bare highlight records where you looked. A note in your own words records what you thought, and that's the part you'll need to retrieve. With Glasp's web highlighter, you can highlight web pages and PDFs and attach a note to any highlight.
  • Review the same highlights in different settings. Go through them on your laptop at home, then on your phone somewhere else. It's a cheap version of the two-room condition in Smith, Glenberg and Bjork's experiment.
  • Keep your cues together. If your book highlights live in one app, your articles in another and your lecture notes in a third, your review is scattered too. In Glasp, your Kindle highlights can sit next to your web highlights and notes.
  • Summarize lectures from memory. YouTube Summary puts a lecture's transcript beside the video: pause, write down what you remember, then check it against what was said.
  • Turn highlights into practice questions. Testing yourself usually beats rereading, especially when you check your answers. Glasp's AI chat answers from your own saved highlights, so you can ask it for questions on a source and answer them before you look.
  • Rehearse the reinstatement. When you review a highlight, take a second to remember where you read it and why you marked it. You're practicing the move you'll need if you blank in an exam.

Frequently Asked Questions

What is context-dependent memory in simple terms?

It's the tendency to remember something better when you're in the same situation you were in when you learned it. The situation can be a place, a sound, or your own physical or emotional state. Those conditions act as reminders that help bring the memory back.

What is an example of context-dependent memory?

The classic example is Godden and Baddeley's 1975 study, in which divers recalled words best in the environment, on land or underwater, where they had learned them. An everyday example is failing to recognize a coworker you run into at the supermarket because you've only ever seen them at work.

What is the difference between context-dependent and state-dependent memory?

Context-dependent memory usually refers to the external environment, such as the room or the background noise. State-dependent memory refers to your internal state, such as being sober or intoxicated, as in Goodwin and colleagues' 1969 alcohol study. Mood-dependent memory is the emotional version of state-dependent memory.

Does studying in the same place as the exam help?

A little, on average, but less than most people think. A 2001 meta-analysis found a small average benefit for matching environments (d = 0.28), and a study of real college exams found no reliable difference across 21 comparisons. Studying in quiet and practicing under test-like conditions matter more.

Is it better to study in different places?

Often, yes. In a 1978 experiment, students who studied a list in two different rooms recalled 24.4 words, compared with 15.9 for those who studied it twice in one room. A 2021 study found that changing settings helped or did no harm when people restudied, but lowered recall when they tested themselves, and the cost carried over to a later test.

Why do I forget things when I walk through a doorway?

Gabriel Radvansky and colleagues found that memory declined when people moved from one location to another, and that going back to the original room didn't restore it. This is often called the doorway effect. The evidence is mixed, though: a 2021 series of four experiments by McFadyen and colleagues found no significant effect of doorways on forgetting, apart from more recognition errors when people were also holding other information in mind.

Did the Godden and Baddeley diver study replicate?

Not in the one published direct replication. In 2021, Jaap Murre ran the experiment again with 16 divers in a pool in Amsterdam and found no advantage for recalling words in the environment where they were learned. The original study still appears in most textbooks.

Does chewing gum help you remember on a test?

Not in any way you can count on. A 2004 study found that switching between gum and no gum from learning to recall hurt performance, but a 2012 review counted four studies that failed to replicate a context-dependent effect of chewing gum, against two that found one.


Conclusion: The Room Matters Less Than the Cues You Bring

Context-dependent memory is real. What you learn carries traces of where and how you learned it, and matching those conditions at recall can help. But the popular version of the story, that you should study in the exam room or chew the same gum on test day, rests on effects that are small on average, didn't show up in real college exams and didn't survive the diver study's one published replication.

The more useful lesson is about which cues you depend on. A memory tied to one room, one playlist or one mood is fragile. A memory you've studied in several places, practiced retrieving and attached to notes in your own words is sturdy, and if you blank anyway, you can rebuild the context in your head.

Glasp is built for that kind of studying: highlight on the web and in PDFs, keep your Kindle highlights alongside them, write down why each passage matters, and come back to quiz yourself wherever you are.


References

Baker, J. R., Bezance, J. B., Zellaby, E., and Aggleton, J. P. (2004). Chewing gum can produce context-dependent effects upon memory. Appetite, 43(2), 207-210. doi:10.1016/j.appet.2004.06.004

Bower, G. H. (1981). Mood and memory. American Psychologist, 36(2), 129-148. doi:10.1037/0003-066X.36.2.129

Bower, G. H., and Mayer, J. D. (1985). Failure to replicate mood-dependent retrieval. Bulletin of the Psychonomic Society, 23(1), 39-42. doi:10.3758/BF03329773

Dunlosky, J., Rawson, K. A., Marsh, E. J., Nathan, M. J., and Willingham, D. T. (2013). Improving students' learning with effective learning techniques: Promising directions from cognitive and educational psychology. Psychological Science in the Public Interest, 14(1), 4-58. doi:10.1177/1529100612453266

Essoe, J. K.-Y., Reggente, N., Ohno, A. A., Baek, Y. H., Dell'Italia, J., and Rissman, J. (2022). Enhancing learning and retention with distinctive virtual reality environments and mental context reinstatement. npj Science of Learning, 7, 31. doi:10.1038/s41539-022-00147-6

Fernandez, A., and Glenberg, A. M. (1985). Changing environmental context does not reliably affect memory. Memory and Cognition, 13(4), 333-345. doi:10.3758/BF03202501

Godden, D. R., and Baddeley, A. D. (1975). Context-dependent memory in two natural environments: On land and underwater. British Journal of Psychology, 66(3), 325-331. doi:10.1111/j.2044-8295.1975.tb01468.x

Godden, D., and Baddeley, A. (1980). When does context influence recognition memory? British Journal of Psychology, 71(1), 99-104. doi:10.1111/j.2044-8295.1980.tb02735.x

Goodwin, D. W., Powell, B., Bremer, D., Hoine, H., and Stern, J. (1969). Alcohol and recall: State-dependent effects in man. Science, 163(3873), 1358-1360. doi:10.1126/science.163.3873.1358

Grant, H. M., Bredahl, L. C., Clay, J., Ferrie, J., Groves, J. E., McDorman, T. A., and Dark, V. J. (1998). Context-dependent memory for meaningful material: Information for students. Applied Cognitive Psychology, 12(6), 617-623.

Imundo, M. N., Pan, S. C., Bjork, E. L., and Bjork, R. A. (2021). Where and how to learn: The interactive benefits of contextual variation, restudying, and retrieval practice for learning. Quarterly Journal of Experimental Psychology, 74(3), 413-424. doi:10.1177/1747021820968483

McFadyen, J., Nolan, C., Pinocy, E., Buteri, D., and Baumann, O. (2021). Doorways do not always cause forgetting: A multimodal investigation. BMC Psychology, 9, 41. doi:10.1186/s40359-021-00536-3

Memon, A., Meissner, C. A., and Fraser, J. (2010). The Cognitive Interview: A meta-analytic review and study space analysis of the past 25 years. Psychology, Public Policy, and Law, 16(4), 340-372. doi:10.1037/a0020518

Murre, J. M. J. (2021). The Godden and Baddeley (1975) experiment on context-dependent memory on land and underwater: A replication. Royal Society Open Science, 8(11), 200724. doi:10.1098/rsos.200724

Radvansky, G. A., Krawietz, S. A., and Tamplin, A. K. (2011). Walking through doorways causes forgetting: Further explorations. Quarterly Journal of Experimental Psychology, 64(8), 1632-1645. doi:10.1080/17470218.2011.571267

Saufley, W. H., Otaka, S. R., and Bavaresco, J. L. (1985). Context effects: Classroom tests and context independence. Memory and Cognition, 13(6), 522-528. doi:10.3758/BF03198323

Smith, S. M. (1979). Remembering in and out of context. Journal of Experimental Psychology: Human Learning and Memory, 5(5), 460-471. doi:10.1037/0278-7393.5.5.460

Smith, S. M. (1985). Background music and context-dependent memory. American Journal of Psychology, 98(4), 591-603. doi:10.2307/1422512

Smith, S. M., Glenberg, A., and Bjork, R. A. (1978). Environmental context and human memory. Memory and Cognition, 6(4), 342-353. doi:10.3758/BF03197465

Smith, S. M., and Rothkopf, E. Z. (1984). Contextual enrichment and distribution of practice in the classroom. Cognition and Instruction, 1(3), 341-358. doi:10.1207/s1532690xci0103_4

Smith, S. M., and Vela, E. (2001). Environmental context-dependent memory: A review and meta-analysis. Psychonomic Bulletin and Review, 8(2), 203-220. doi:10.3758/BF03196157

Tucha, L., and Koerts, J. (2012). Gum chewing and cognition: An overview. Neuroscience and Medicine, 3(3), 243-250. doi:10.4236/nm.2012.33028

Tulving, E., and Thomson, D. M. (1973). Encoding specificity and retrieval processes in episodic memory. Psychological Review, 80(5), 352-373. doi:10.1037/h0020071

Start building your knowledge library

Highlight what matters as you read across the web. Save insights from articles, books, and YouTube videos in one place.

Get Started Free

Or highlight this page as you read it