Lecture playback speed: does 1.5x hurt learning?

Lecture playback speed is how fast a recording plays against real time. At 1.5x a 60-minute lecture takes 40 minutes; at 2x it takes 30. Whether the time you save costs you marks is the question behind almost every discussion of studying from recordings, and the research answers it in two directions at once.

What the studies on lecture playback speed found

Merhavy and colleagues (2023, BMC Medical Education) tested second-year medical students at one university. The students watched lectures at normal speed, 1.5x and 2x. Concentration and long-term retention showed no significant difference between the speeds. The authors' recommendation was blunt: students may watch at their preferred speed without worrying about their learning.

Murphy, Hoover and Castel (2023, Memory) showed a pre-recorded video lecture to younger and older adults at manipulated speeds. Younger adults watched faster without a significant memory deficit. Older adults performed worse at faster speeds. Faster playback also reduced mind-wandering.

Massoud and El Khoury-Malhame (2026, BMC Psychology) ran a within-subject experiment with 40 university students and speech clips of about 23 seconds. Normal speed produced roughly 27% better cued recall than time-compressed speech.

A scoping review by Silva and colleagues (2026, CoDAS) mapped nine studies of speech speed in audiovisual material for university students. Four found better learning at speeds up to 2x, three found no difference, and two found worse learning. Above 2x the review reports that learning is limited. It calls 1x to 1.75x the most favourable range.

The numbers get less kind in a meta-analysis. Huang, Du and Yang (2025, Frontiers in Psychology) pooled 12 experimental and quasi-experimental studies. Accelerated playback raised cognitive load (g = 0.59) and lowered retention-test scores (g = -0.41), transfer scores (g = -0.50) and satisfaction (g = -0.24).

Liang and colleagues (2026, British Journal of Educational Psychology) tested a different arrangement. In two studies of 100 and 96 students, one group could set its own speed. That group mind-wandered less and performed about the same as students kept at a fixed speed.

So the short answer is not "1.5x is fine" or "2x ruins you". Two studies find no harm in a memory test, two find harm on transfer, and one review says the answer depends on the material. The useful skill is knowing which situation you are in.

Why the studies disagree about playback speed

Four differences explain most of the gap.

The first is the outcome measured. A retention test asks whether you remember facts from the lecture. A transfer test asks whether you can use them in a new problem. Acceleration tends to survive the first and fail the second, which is exactly what the meta-analysis found.

The second is the stimulus. A 23-second clip is not a 60-minute lecture. Short clips remove the hardest demand of a real lecture, which is holding attention for an hour. Studies built on short clips will always look more encouraging than studies on full lectures.

The third is who controls the speed. A preset 2x is not the same experience as choosing 2x, dropping to 1x for a hard passage and speeding back up for a familiar one. The self-paced studies test the second thing. Most of the others test the first.

The fourth is the listener. A young adult with strong working memory tolerates speed better than an older adult, and the Murphy study measured that gap directly. A student hearing a second language, an unfamiliar accent or completely new vocabulary is not in the same condition as a medical student revising a subject seen twice.

What 2x does that 1.5x does not

At 1.5x a lecturer speaking 150 words a minute reaches you at 225. At 2x the same voice arrives at 300 words a minute. Nothing in the sound disappears. What disappears is the time to process each sentence.

Working memory holds what you just heard long enough to give it meaning. Cut that window in half and more sentences arrive before the previous one is resolved. Cognitive load rises, which is the g = 0.59 in the meta-analysis.

Two things break first. Note-taking becomes a choice between writing and listening, and unfamiliar terms arrive before you can place them. Both failures are silent during the lecture. You see them a week later, when you close the transcript and nothing comes back.

The speed that breaks your notes

If you write notes while a recording plays, the notes are the fragile part of the session. At normal speed you can copy a definition and still hear the example that follows it. At 1.5x the definition arrives while you are still writing the previous line.

The practical rule is to decide what the notes are for before you choose a speed. A lecture you will quote in an essay is a 1x lecture with pauses. A lecture you only need to recognise on a multiple-choice paper can run at 1.5x, because recognising a term is easier than reproducing it.

If your notes at 1.5x are a string of nouns that made sense at the time, you did not lose the lecture to the speed. You moved the work to a review session you will not schedule.

Lecture playback speed compared

How much time each speed actually saves

A 60-minute lecture, at each speed:

• 1x: 60 minutes.
• 1.25x: 48 minutes.
• 1.5x: 40 minutes.
• 1.75x: 34 minutes.
• 2x: 30 minutes.

Those numbers are the honest argument for speeding up. 1.5x returns 20 minutes per lecture, and two lectures return 40. The question is what happens to the 40 minutes. If the answer is "another lecture at 1.5x", the saving cancels itself. If the answer is "close the transcript and answer five questions", it compounds.

How to choose a speed for one lecture

• Start at 1x for five minutes. If every sentence carries a new term, stay there.
• Move to 1.25x when the topic is familiar or the speaker is slow. That is where most students stop, and the evidence supports stopping there.
• Use 1.5x on a second pass, or on a lecture you already attended.
• Use 2x to find a moment you remember, not to learn a passage you do not.
• Never study above 2x. The review finds learning limits there, and the time gain is smaller than it feels.
• If the recording has timestamps, jump to the passage instead of speeding through everything between.
• Every ten minutes, state the last point aloud without looking. If you cannot, drop one step.

Test yourself instead of speeding up

Roediger and Karpicke (2006, Psychological Science) gave students prose passages, then had them either restudy the material or take free-recall tests without feedback. A final test followed after five minutes, two days or one week. After five minutes the students who restudied recalled more. After a week the students who had been tested recalled more. The restudy group also felt more confident, which is the least reliable signal in the whole experiment.

The application to a recording is simple. Pause every 10 to 15 minutes, write three questions, close everything and answer them out loud. Then continue. That pause does more for retention than any speed below 2x does against it.

If you have no time to pause, you have no time to learn, and 1.5x is only a way to finish sooner.

Why 2x can feel easier and still be worse

Faster playback reduced mind-wandering in the Murphy study, and self-pacing reduced it again in the Liang study. Attention is a real benefit, not an illusion.

The illusion sits in the step from attention to knowledge. A fast lecture keeps you alert, and alertness feels like understanding. A retention test does not measure alertness. It measures what you can produce with the recording closed.

There is a second trap. Speeding up changes the ratio between hearing and thinking. You hear the whole lecture and think about perhaps a third of it. The lecture feels complete and the thinking never happened.

What a faster speed cannot fix

It cannot create understanding you never had. If a proof or a method was unclear at 1x, it will be unclear and faster at 2x.

It cannot fix an accent you cannot parse. Slow down instead. 0.75x exists for a reason, and the scoping review notes one study where a speed slower than normal improved learning.

It cannot replace a bad recording. If the microphone caught the room and not the lecturer, no speed saves the hour.

It cannot replace sleep, and it cannot replace the test you keep postponing. Speed is a decision about time. It is not a decision about memory.

When 1.5x is the wrong choice

• The vocabulary is new and every sentence carries a definition.
• You are listening in a second language, with an accent you cannot parse at normal speed.
• You are writing notes you will quote or build an argument from.
• The lecture teaches a procedure and you have not worked it once end to end.
• You cannot state the main point of the last ten minutes.
• You are using speed to avoid the pause where you would test yourself.
• You finished at 2x and cannot say what the lecture was about.