Jared Cooney Horvath

How can we better structure the school day?


As the government considers extending the school day, we put this article by Jared Cooney Horvath in front of the paywall to demonstrate it is a lot more complex than ministers may believe...

Length of lessons

This article is usually only accessible to subscribers. To get exclusive weekly content from Tes, subscribe here

In most schools in most countries, a single lesson will, on average, last for around an hour (give or take 15 minutes). And it’s not just schools: whether you are learning to dance or drive, to play tennis or the trumpet, to speak Mandarin or make cakes, around 60 minutes is deemed optimal.

So there must be a really good scientific reason why sessions of around 60 minutes work, right? Otherwise, we would not all be doing lessons of that length. And what about the way the school day is structured beyond that?

There are three principles that may inform lesson duration and how many lessons you may want to schedule in each day. Let’s look at each in turn.

Principle 1: Processing power

The brain has a unique dual-system learning apparatus that swings between active engagement and passive automaticity. Put simply, whenever we first learn new material, we must actively and cognitively engage with that material (think about the first time you learned to drive a car). However, once we learn something, we can begin to automate it and run on passive autopilot (think of your last drive to the supermarket).

The reason why the brain contains this dual system concerns energy. When we are in passive autopilot mode, the brain uses approximately 25 per cent of all the energy within the body. However, when we are in active learning mode, the brain requires even more energy. Learning is costly.

Seeing as there’s a finite amount of energy in our body, this means there’s a finite amount of time that we can spend in active learning mode before burning out. The official term used to describe this phenomenon is cognitive fatigue, and if you’ve ever attended an all-day professional learning conference, you’ve likely experienced this process. After a couple of exciting sessions, attention starts to wane. After another couple of sessions, it becomes difficult to follow complex arguments. Finally, by the end of the day, you’re remembering almost nothing – you’ve reached cognitive fatigue.

So, what is this finite energy resource that the brain is burning through?

Although debate remains, our best bet is glucose – a simple sugar distributed throughout our body via the blood. Barring hypoglycaemia (where blood sugar runs low and cognitive functions decline), the blood typically transports only enough glucose to the brain to maintain passive autopilot mode. This means that in order to access active learning mode, the brain must tap into a secondary energy source.

It turns out that glucose can be stored in non-neuronal regions of the brain in the form of astrocytic glycogen – and herein lies our cognitive limit. Whenever we jump into active learning mode, neurons begin to tap into this secondary energy store. Unfortunately, this secondary energy store tends to replenish only when we are asleep. This means that once we run out of astrocytic glycogen (and hit cognitive fatigue), learning is all but done for the day and can’t be meaningfully resumed until after a sleep.

So, how long do we have? As you can guess, astrocytic glycogen stores will change depending upon a number of factors (sleep, diet, exercise, etc), but our best guess is that we have anywhere between 30 minutes’ and several hours’ worth of active, cognitive engagement.

Being generous, we can assume that students will be able to maintain focused, vigilant learning for three hours each day.

But a secondary consideration here concerns diminishing returns. Once you begin accessing astrocytic glycogen, performance doesn’t remain strong until you completely tap out. Performance decreases with sustained effort.

For instance, during a vigilance task (where people must watch rapidly switching images and respond whenever a specific target appears), performance commonly declines by around 30 per cent over the course of 40 minutes. Additionally, individuals who study for 60 minutes remember, on average, only 9 per cent more than people who study the same material for 30 minutes – nowhere near the double one would intuitively expect.

Verdict: Put together, these concepts certainly suggest that six hour-long lessons per day might not be ideal for learning. Rather, shorter periods lasting only 30 minutes might be ideal.

However, it’s important to remember that our energy-sapping learning system activates only when we undertake new learning or highly difficult tasks.

When we are simply practising or revising skills, we typically engage autopilot and need not tap into limited reserves. Accordingly, the common partitioning of 50-minute lessons into 15 minutes of learning, 15 minutes of practice and 15 minutes of discussion seems fairly sensible and aligns well with considerations around energy demand.

However, there’s one added thing worth mentioning: with limited astrocytic glycogen, afternoon classes are at a distinct disadvantage. If the morning is crammed with deep, involved learning, there is a very real chance that students will run out of cognitive reserves by the time periods four or five come around.

Principle 2: Attention thresholds

Put simply, in order to pay attention to and learn from something, relevant stimuli must be stronger than your attention threshold. For instance, a very loud alarm would likely be above your threshold (meaning you would notice it) while an ant sneezing would likely be below your threshold (meaning you wouldn’t notice it). In a way, our threshold acts as a filter, allowing us to more easily make sense of this chaotic world.

Here’s the problem: human thresholds are constantly changing. More specifically, whenever a person spends an extended period of time in a particular environment, her or his threshold will start to raise up, making it harder to maintain attention.

To consider this in action, think about driving. When you first get in the car, you’ll typically set the stereo to a nice level. However, after a while, you might notice the music isn’t as powerful as before, so you crank the volume up a bit. After a little while longer, the music once again starts to feel weak, so you crank the volume up again. In this instance, as your brain gets used to the environment of the car, the hum of the engine, the vibration of the road, your threshold raises up making it harder and harder to attend to the music. (Interestingly, after returning to your car after a day of work, you’ll likely find the stereo is way too loud – this is because you’ve spent enough time outside that environment that your threshold has reset to normal.)

This process is called stimulus-specific adaptation, and it makes sense. After spending extended time in a predictable environment, the brain will come to recognise that there is no threat to safety. It will raise thresholds and kick on autopilot to conserve astrocytic glycogen in case a threat appears in the future.

Verdict: Although it will differ across circumstances, environmental adaptation of this sort will typically occur between five and 30 minutes. If we split the difference, we can assume students will last about 15 minutes before their threshold begins to raise and their attention starts to wane. In fact, that’s what the research tends to show. This suggests that lessons should be even shorter than 30 minutes – perhaps only 10 minutes or shorter (the YouTube approach to education).

Not so fast. Luckily, thresholds will reset whenever predictability disappears or an unexpected stimulus arises. This means we needn’t replace lessons with soundbites; we need only consider ways to surprise or shock the system every 15 minutes. Asking students to engage in small group discussion, showing a video clip, telling a story, even getting up to stretch – anything that breaks predictability should be enough to reset thresholds and buy another 15 minutes of attention. Importantly, these prediction breaks needn’t be silly nor irrelevant; they can (and should) be topically relevant, so as to maintain the learning momentum.

This means that, so long as we are resetting thresholds, in attention terms, lesson durations can be as long or short as desired. Again, the common allocation of 50 minutes (15 minutes of learning, 15 minutes of practice, 15 minutes of discussion) does a decent job of playing this game.

Principle 3: Flow state

The final consideration is flow (also called “being in the zone” ). This is a very interesting psychological phenomenon whereby an individual becomes so deeply absorbed by a task that they lose the senses of time, self-conscious awareness and emotional valence (the ability to engage emotionally with external stimuli).

If you’ve ever become so focused while painting, playing sport, surfing the web or any other endeavour that several hours flew by without you noticing, then you’ve experienced flow.

During a flow state, people typically attain a deep sense of agency, ability and value. Productivity soars, confidence strengthens, distractibility disappears and wellbeing increases. Although debate remains, it’s believed that flow does not tap into astrocytic glycogen stores, meaning that it could conceivably last all day with no fatigue.

However, in order to reach flow, there are several requisites. First, this state occurs only when undertaking a task that is deeply familiar but challenging. This means flow will arise during periods of practice or revision, not during periods of learning.

Second, this state occurs only when undertaking a task that is intrinsically motivating. This means that flow will arise when undertaking an enjoyable or passion project; it cannot be cajoled using grades, rewards or praise.

Third, this state occurs only when undertaking a task for an extended period of time. This means that flow will arise after a fair amount of concerted effort (believed to be about approximately 30 minutes, but there’s no research to support this) – not immediately.

Verdict: Considering flow, we see inklings that perhaps lessons can (and should) be longer. By restricting lessons to 50 minutes, there simply is not enough time for any student to ever truly immerse themselves or get lost within a task, activity or assignment.

I hope you recognise that there are several important cognitive considerations that go into effective learning. Unfortunately, many of these considerations overlap, contradict and don’t always work in tandem to create a clear, unambiguous answer with regards to lesson duration. This is the translation problem in action – a topic explored in one of my previous articles. Accordingly, although optimising learning should be a consideration when scheduling, it can never be the only or ultimate consideration.

However, in the interest of fun (depending on your definition of the word …), how might we put these ideas together to build a schedule?

I suppose the day would really be cut in two: learning in the morning, work in the afternoon.

The morning would consist of five 30-minute class periods. These short lesson durations would play into the consideration of diminishing returns, while allowing a solid two hours and 30 minutes of active, cognitively engaged learning each morning.

I imagine each lesson would begin with a five-minute review of previous learning, followed by 15 minutes of explicit instruction, and conclude with 10 minutes of discussion about this new information. Realistically, this would allow for only one or two new ideas per class per day, which favours deep consideration of fewer items over surface consideration of many.

Seeing as the morning will have likely sapped astrocytic glycogen stores, there would be a nice hour-long lunch break, ensuring that we replenish blood glucose – the primary energy source for the rest of the day.

The afternoon would then be dedicated to two 90-minute sessions or one 180-minute session. The goal of these longer periods would be to allow for a state of flow to set in. Students could spend the afternoon creating and completing deeply considered work meaningful to them.

This would be one way of doing it, but there will likely be many others. In the end, the optimal lesson duration depends on what your ultimate goal is. Once you are clear on what you hope to achieve, then we can combine cognitive and organisational principles to schedule effectively.

Jared Cooney Horvath is a neuroscientist, educator and author

This article originally appeared in the 29 May 2020 issue under the headline “How long should lessons really be?”

Other articles in this issue


Audio Article

Articles with this icon can be read to you
Special report
3 min read 1 Mar 2021