Monitoring: attention in online lectures

The complaint about online lectures usually lands on the wrong object: the screen, the open tabs, the bed two metres away. All of that is real and none of it is the mechanism. Listening is a passive activity, and passive attention is the most demanding kind there is. Holding focus on something that is proceeding correctly without any help from you is harder than doing the work yourself, and an online lecture removes every prop that used to compensate.

Watching a system that is doing everything right

In a modern cockpit the hard part of the job is not flying the aircraft but monitoring an autopilot that is flying it correctly. Crews are trained for this because the failure mode is well documented: when every indication is normal and nothing needs correcting, vigilance decays, and it decays fastest precisely when the system is behaving. The task offers no handholds, and the mind goes looking for somewhere that something is happening.

Your seat in a lecture is the monitoring seat. Someone else is flying. Minute to minute nothing is required of you: no decision, no output, no visible result if you drift for four minutes and come back. Attention needs a task to hold, and watching supplies none. What feels like a personal failure of discipline is mostly the known cost of the position.

What the hall supplied and never charged you for

The lecture in a building was not better because your concentration improved inside that building. It arrived with scaffolding you never had to think about: a journey that marked a start, people whose behaviour you could see and who could see yours, no second screen, a door you could not leave through unnoticed. Mild social embarrassment did a surprising share of the work, and it did it for free.

At home each support is gone and nothing replaced it. Part of the scaffolding is cheap to rebuild: a fixed place that is not your bed, a start time you do not renegotiate every morning, the phone in another room, the lecture filling the whole screen rather than a quarter of it. None of that makes listening active. It only stops the room from working against you.

Listening you can be wrong about

Monitoring becomes bearable the moment it carries a prediction, and that single move changes the activity from the inside. Before the lecturer answers the question they have just posed, answer it yourself, in one sentence, in your head. Before the next slide appears, guess what step has to follow. Being right costs a second and confirms you are still inside the argument. Being wrong leaves you with a live question, and a live question holds attention in a way that receiving never does.

The second move is a summary at the seams. Lectures have joints: the end of a derivation, a change of topic, the moment somebody says now we come to. At each joint, say quietly what the last stretch claimed. Speaking forces you to finish the sentence, thinking lets you get away with half of one. If the sentence will not come, you have just located the exact place where you came off.

The recording is a different problem

A recording looks like the same lecture with the constraints removed and behaves like something else. Deferral comes first: a live session has a time and therefore happens, while a recording joins a pile that only grows, until watching it competes with material from three weeks later. Then the pause button, which removes a pressure that was doing useful work. Being unable to stop forced you past a point you half understood, which is often enough to recover two slides later.

Speed is the third effect. For someone who talks slowly, one and a quarter is genuinely better, because the gap between their pace and your processing is where attention leaks out. At double speed you keep up with the words and lose the argument, and nobody notices, because following words feels identical to following a point. Prediction is the test again. Give the recording a slot in the week the live session had, and watch it within a few days.

During the hour

What is worth doing while the lecture runs is all small. Watch in one block with no second device open. If the format allows questions in a chat, ask one. A typed question puts you back in as a participant, and you listen to an answer differently than to anything else that hour. If the group is small enough for cameras, turn yours on. Being visible is a cheap and slightly undignified trick, and it works.

Drift happens anyway, and the mistake is spending five minutes reconstructing what you missed while the lecture continues without you. In a recording, go back a minute. Live, let the gap be a gap, register the place as unclear and rejoin at the sentence being spoken. One gap is survivable. Chasing one produces four more.

The twenty minutes after it ends

What survives from a lecture is largely decided in the minutes after it stops, and those minutes usually go to closing a window. Before that, say three things out loud: what the lecture was about, which question it answered, and the one point you did not follow. The third item carries most of the value, because it is what will still be true in six weeks and what disappears quietly if you never name it.

Then, on the same day if you can, do one small thing with the material: one exercise, one step of a derivation reproduced without help, one question sent to somebody else on the course. The size does not matter and the delay does.

When the lecture itself is the weak part

Some lectures cannot be rescued by any of this. Slides read aloud with nothing added, audio that makes you guess every fourth word, an hour covering what the handout covers in eight pages. Attention technique applied to a bad lecture means spending your best hour of the day extracting very little, and the honest response is to stop attending. Test it properly first: watch two sessions with prediction and joint summaries, actually doing them. If the script then gives you the same content in half the time and you can say more about it afterwards, the script wins.

Two things complicate the call. Attendance sometimes counts, and lecturers say things in the room that exist nowhere in the written material, including what they intend to ask about. The more common error runs the other way, though. A subject that makes you feel slow gets filed as a bad lecture, and for the first fortnight the two are indistinguishable from the inside, while only one of them improves when you leave.