Should You Watch Math Videos at Double Speed?

A review that combined 12 papers found a clear overall downside to accelerated instructional videos. Faster playback increased mental load, meaning how much information the learner had to handle at once. It also lowered scores on later memory questions and similar new tasks. Learners reported enjoying the experience less as well^[1].

That review does not mean every increase in speed is harmful. One experiment placed 76 undergraduates into groups watching at 1×, 1.25×, 1.5×, or 2×. The strongest learning happened at 1.25× and 1.5×. Students placed in the lower-ability group did best at 1.25×, while the higher-ability group did best at 1.5×^[2].

Other experiments were more encouraging about double speed. Across four experiments, faster playback up to 2.5× did not reduce test performance. Visual material was especially helpful when videos moved faster, although the researchers still advised against going beyond 2×^[3].

Two more experiments involving 320 learners found no clear changes in memory or mind-wandering through 2×. Enjoyment of the chosen speed fell as playback became faster. Above 2×, memory and liking both declined^[4].

These findings do not give every student one ideal number. They support testing moderate increases before jumping straight to 2×.

A math video often combines speech, symbols, and written changes. Consider this equation:

$x^2+6x-7=0$

A teacher completing the square might write:

$x^2+6x=7$

$x^2+6x+9=16$

$(x+3)^2=16$

$x=1 \text{ or } x=-7$

At double speed, those lines can look like a smooth chain. Yet each line answers a different question.

First, why move $-7$? Adding 7 to both sides isolates the terms containing $x$.

Next, why add 9? Half of 6 is 3, and $3^2=9$. That creates the square $(x+3)^2$.

Why does the right side become 16? The same 9 must be added to both sides, so $7+9=16$.

Finally, why are there two answers? If $(x+3)^2=16$, then $x+3$ can equal 4 or $-4$.

Playback speed matters when those reasons are new. Hearing every sentence is not the same as being able to explain each operation. As a practical judgment, pause whenever the written expression changes. Say why the change is allowed before continuing.

Then test the method on a fresh equation:

$x^2-8x+5=0$

Move the constant:

$x^2-8x=-5$

Half of $-8$ is $-4$, and $(-4)^2=16$. Add 16 to both sides:

$(x-4)^2=11$

Therefore:

$x=4\pm\sqrt{11}$

If you can produce those steps and explain them, the video pace probably worked. If you only recognize the teacher’s completed steps, slow down.

Across two studies, learners who could control playback reported less mind-wandering than learners given a fixed speed. Their learning performance remained comparable^[5]. That supports changing speed within a lesson instead of choosing one setting for the whole video.

Use faster playback for material you already know. Return to normal speed when a new rule appears. Pause before the teacher reveals the next algebra step. Rewind when you cannot explain why two expressions are equal.

Notes can also help. In two experiments, taking notes supported memory at both normal and double speed. Test performance tended to fall as speed increased, with a clear drop in the laptop experiment. The researchers concluded that notes may help make up for some disadvantages of faster playback^[6].

For math, useful notes are not a transcript. Copy the problem, the key transformation, and its reason. For example, write “add 9 to form a perfect square,” not every sentence the teacher says.

Learn4Less sessions often use this pause-predict-explain check with a worked example.

Here is one workable routine for the night before your test:

1. Start the first video at 1.25× or 1.5×. Those speeds produced the strongest learning in one experiment with undergraduates, although your best pace may differ^[2].
2. Pause before each important algebra step. Predict what the teacher will write next. This is a practical check that you are following the reasoning.
3. Write short reason notes. Note-taking supported memory at both normal and faster speeds in two experiments^[6].
4. Try one related problem without the video. If you cannot begin, replay the example more slowly. Do not merely watch it again at the same speed.
5. Use 2× only for familiar sections. Experiments found that memory often held up through 2×, but enjoyment fell and speeds above 2× hurt memory^[4].

After the first video, close your notes. Write the main rule and solve one short problem. If that works, use the same pace for the next video. If it does not, lower the speed.

Finishing two videos with working understanding may prepare you better than finishing all three without being able to reproduce a method. That is practical judgment, not a result directly tested by these studies.

The abstracts in this source pool do not report a direct experiment with high-school students learning algebra. One recent experiment studied people ages 19 to 26 watching information-based videos rather than high-school math lessons^[7].

The results also differ by task. A review of 12 papers found lower retention and weaker performance on similar new tasks with acceleration^[1]. Yet several individual experiments found little or no loss through 2×^[3][4]. In the soil science course, 2× caused small quiz losses in some cases but not others, while speeds through 1.5× caused no losses^[8].

These studies therefore cannot name the perfect speed for your algebra video. They also do not tell us whether every kind of math works equally well at the same pace. A proof, a graph explanation, and a routine calculation may place different demands on the viewer.

Do not choose one speed and leave it untouched. Begin at 1.25× or 1.5×. Pause at every unfamiliar symbolic change. Predict the next step, then explain why it works.

Move to 2× only when the section is review and your predictions remain accurate. If you stop being able to explain the algebra, lower the speed immediately. The goal tonight is not three completed progress bars. It is being able to solve the next problem when no video is playing.