Why Time Flies in the Moment—and Stretches in Memory

The paradox: one hour, two answers

Imagine two Saturdays. On the first, you answer messages and move through the familiar rooms of home. On the second, you take a train to a new town, miss a turn, find a market, and arrive home after dark. Both contain the same number of clock hours. They are unlikely to occupy the same amount of psychological time.

The stranger day may pass almost without notice. There is always another street to choose or detail to absorb. Yet, on Sunday, it can seem much longer than the quiet day beside it. The quiet day may have felt slow at three in the afternoon and nearly empty in retrospect. Time apparently flew and stretched in the same experience.

This familiar contradiction becomes less mysterious once we stop treating “How long did it feel?” as a single question. Psychology distinguishes judgments made while an interval is unfolding from judgments reconstructed after it has ended. Researchers call these prospective and retrospective duration judgments. The terms sound as if one concerns the future and the other the past, but the important difference is simpler: did you know you were attending to time, or did you only consider duration afterward?

Those situations ask the mind to consult different evidence. One leans heavily on attention. The other leans heavily on memory.

One hour, two readings

The mind does not consult the same evidence twice

Time while living it

1. Attention turns toward time
2. ↓
3. More temporal signals are noticed
4. ↓
5. The interval can feel longer now

Time while remembering it

1. Memory retrieves event boundaries
2. ↓
3. Distinct segments are reconstructed
4. ↓
5. The same interval can feel fuller later

The clock you watch

Suppose you are waiting for water to boil and have decided not to look away. The clock is visible. The sound of the kettle has not changed. Each glance confirms that less time has passed than you hoped. In a prospective judgment, time itself is part of the task, so attention repeatedly returns to it.

A major meta-analysis of prospective and retrospective duration judgments found that prospective estimates tend to be longer and less variable than retrospective estimates. The authors concluded that attention-based accounts are needed to explain judgments made with advance knowledge, while memory-based accounts are needed for judgments made afterward. This does not mean attention is a literal stopwatch in the brain. It means that noticing temporal information changes the evidence available when you judge an interval.

Attention is limited. When a demanding activity occupies it, fewer resources remain for monitoring time. A difficult puzzle, an absorbing conversation, or an unfamiliar route can therefore leave little room for asking, “How long has this been going on?” When the activity is thin and the ending matters—a delayed train, a loading screen, the final minutes of a meeting—attention can keep returning to duration.

Even here, “duration” is not identical to the felt speed of passage. A 2022 Scientific Reports study examined both verbal duration estimates and judgments about how quickly time seemed to pass. The measures did not behave as interchangeable versions of one experience. An interval can be estimated in minutes while separately feeling as if it dragged or flew.

The time your memory rebuilds

Now remove the clock. At the end of the day, someone asks how long lunch seemed to last. You cannot return to the interval and read its duration directly. You have to rebuild an answer from what remains: arriving at the café, waiting for a table, the first course, a story your friend told, the rain beginning, the walk home.

Memory is not a continuous recording stored at full resolution. A review of temporal compression in episodic memory describes recollection as a succession of remembered “slices,” with portions of an experience omitted between them. The density of recalled segments helps determine how compressed an event becomes when we mentally revisit it.

This gives retrospective time its peculiar logic. An hour containing several recoverable episodes can appear substantial because memory has many landmarks from which to reconstruct it. A routine hour can collapse because one moment predicts the next and relatively little distinguishes the middle from the beginning. The remembered duration is not necessarily accurate, but it is not random either. It is an inference built from the structure that survived.

The process also changes with delay. Details fade, events merge, and later knowledge can reorganize what seemed separate at first. Yesterday’s trip may be reconstructed from individual encounters; last year’s trip may be summarized as three places and one missed flight. When memory compresses an episode, it often compresses its time as well.

This is why the same busy afternoon can feel short while you live it and long when you remember it. Absorption reduces active time monitoring in the moment. Distinct episodes give memory more structure afterward. The two judgments can move in opposite directions because they were never measurements from the same instrument.

Event boundaries: where memory draws the cuts

Continuous experience does not arrive with chapter headings, yet the mind segments it into events. A change in place, goal, action, or situation can mark the end of one event model and the beginning of another. Leaving the office, entering the station, boarding a train, and sitting down for dinner may become separate units even when no pause exists between them.

Recent work makes this connection to remembered duration unusually concrete. In a study of naturalistic everyday videos, participants identified boundaries between subevents and later estimated duration. The number of subevents they recalled predicted their judgments of the whole event’s duration, both immediately and after a seven-day delay. Recalled details and general gist did not have the same relationship for the event as a whole. For smaller subevents, however, content memory mattered differently.

The nuance is important. It is tempting to reduce the result to “more memories make time feel longer,” but memories have levels of structure. The number of distinct chapters may matter for a complex event, while the texture inside a single chapter may matter at another scale. Duration judgments depend on what is being judged and how the remembered experience is organized.

Why “add novelty” is only half the story

Advice about slowing down life often ends with one word: novelty. New environments are less predictable and can produce more distinguishable events than a repeated routine. But novelty is not a lever that reliably stretches every kind of subjective time.

First, novelty can have opposite effects at different moments. An engaging new experience may reduce clock-watching and feel fast as it happens. Its distinct features may make it expansive in memory. Second, novelty can overload attention rather than enrich recollection. A frantic itinerary can blur into transit, queues, and photographs, while one quiet conversation becomes the clearest event of a trip.

Third, findings from short laboratory intervals should not be inflated into rules about whole years. A 2024 study found that more memorable images were perceived for longer during brief visual presentations, and that perceived duration and later memorability influenced one another. This is fascinating evidence that visual processing, perceived time, and memory are linked. It is not proof that collecting novel images will make a summer—or a life—objectively feel longer.

A better principle is meaningful differentiation. A day becomes easier to reconstruct when it contains changes that matter: a new goal, a different setting, a conversation with a clear before and after. Random stimulation creates change, but not necessarily coherent event boundaries. Memory needs something it can organize.

Three ways to work with both clocks

The research does not prescribe a perfect schedule, but it suggests three useful experiments.

1. Replace repeated checking with one external cue

If an ending matters, constant clock-checking keeps attention attached to time. Set one external alert, then let the activity—not the remaining minutes—hold attention. This will not make every wait pleasant, and an alert cannot remove uncertainty about a cancelled train or medical result. It simply removes the need to perform the same temporal check again and again when the ending is predictable.

2. Give important periods meaningful boundaries

If you want a day to remain distinct, create transitions that memory can recognize. Change location for a walk, close one task before opening another, or pause to name what just ended. The aim is not to fill every hour with novelty. It is to prevent an important stretch from becoming one undifferentiated block. A boundary can be quiet: leaving a desk, writing a final sentence, or taking a different route home.

3. Record elapsed time when accuracy matters

Memory is a poor billing system. A task with many interruptions may feel long in retrospect; a fluent task may be underestimated because its middle contains few memorable cuts. When planning, cooking, training, or estimating work, record objective start and finish times. Use recollection to understand the experience, not as the only source for its duration. Comparing estimates with records gradually reveals your own recurring distortions.

These are applications, not direct experimental conclusions. They translate attention and memory findings into low-risk practices that can be tested and discarded. The point is not to control subjective time perfectly. It is to notice which clock a situation is asking you to use.

A seven-day observation

For one week, choose a single activity each day: a commute, focused task, meal, workout, or conversation. Before it begins, note the start time and arrange one signal for the end so you do not need to monitor the clock.

Immediately afterward, answer three questions before checking the elapsed time:

1. How many minutes do you think passed?
2. Did time seem to move slowly or quickly, on a scale from one to seven?
3. How many distinct phases of the activity can you name?

Then record the actual duration. In the evening, estimate the duration once more and list the event boundaries you still remember. Do not treat the exercise as a clinical test. Seven observations cannot establish a personal law, and expectations will influence what you notice. It is simply a way to separate measures that ordinary language blends together.

You may find that estimated minutes, felt speed, and remembered structure disagree. A run can feel slow but be underestimated. A conversation can fly and later occupy half the day in memory. A routine commute can be judged accurately while leaving almost no retrievable episodes. The disagreements are the interesting part: they show that “How long was it?” conceals several questions.

One clock, two questions

Clocks are indifferent to attention and memory. An hour remains an hour whether it is watched, ignored, enjoyed, or forgotten. Human beings need something more flexible. We must decide when an event will end while it is happening, then later use incomplete memories to understand where our time went.

Prospective judgment asks, in effect, “How much duration am I noticing now?” Retrospective judgment asks, “How much structured experience can I reconstruct?” Neither answer is a defect. Each supports a different job.

The next time an afternoon vanishes and later feels enormous, there is no need to choose which impression was real. Both were real responses to different evidence. Time did not change speed. The mind changed questions.

Sources

1. Block, R. A., and Zakay, D. Prospective and retrospective duration judgments: a meta-analytic review. Psychonomic Bulletin & Review, 1997.
2. Martinelli, N., and Droit-Volet, S. Judgment of duration and passage of time in prospective and retrospective conditions and its predictors for short and long durations. Scientific Reports, 2022.
3. D’Argembeau, A., Jeunehomme, O., and Stawarczyk, D. Slices of the past: how events are temporally compressed in episodic memory. Memory, 2022.
4. Yue, W. W. Y., Liu, J., Yao, Z., Zhang, Y., Mu, Z., and Hu, X. Retrospective duration judgments of naturalistic events depend on memories of event boundaries. Psychonomic Bulletin & Review, 2026.
5. Ma, A. C., Cameron, A. D., and Wiener, M. Memorability shapes perceived time (and vice versa). Nature Human Behaviour, 2024.