Why where you learn matters

In the run-up to the exam season, you may be drilling students in their memory of key facts. But, says Jared Cooney Horvath, you might want to consider the ‘where’ of this learning as much as the ‘what’. The neuroscientist explains that the context in which a student absorbs information can have a big influence on their ability to recall it the exam hall

It's not what you learn, but where you learn it

On a balmy afternoon more than 20 years ago, I watched my great-uncle Mooney perform one of the most incredible feats of skill I’ve ever seen.

On this particular day, Mooney was using an old shovel to dig out a dead tree stump from the oversized backyard of his home – at the time, our family was all living in the US. When the trunk was out, I watched as he walked over to a concrete incinerator (about the same size and shape as an oil drum), reached in, pulled out half-a-dozen golf balls, and tossed them haphazardly around his lawn.

I assumed my uncle was simply clearing the balls out to make room for the tree trunk – but I was wrong. Instead, Mooney walked up to the nearest golf ball, took aim and, using his shovel like a golf club, chipped the ball high into the air and dropped it perfectly into the incinerator. He then proceeded to do the same for each remaining ball: six shovel swings, six perfect shots.

When I asked how he’d learned to do that, he explained that, during the Great Depression, he had needed to come up with a creative way to make money. His idea: practise chipping golf balls into the incinerator using the only “club” available to him: his father’s work shovel.

After honing his skills, he began to invite still-wealthy golfers from the local course over for a little wager: they could pick any spot on the lawn and Mooney would bet 5 cents he could pitch the ball closer to the incinerator using his shovel than they could using their golf clubs.

Over time, this little wager became so popular that golfers would skip the course and spend the entire afternoon chipping around his backyard for cash.

Here’s the twist: one day, a group of golfers invited Mooney to the local course to participate in an official chipping competition. With his shovel in tow, he expectantly lined up his first shot … and completely missed the green.

Blaming it on nerves, he confidently lined up his second shot … and missed again. Out of 10 shots, he only hit the green twice.

What had happened? Why could my uncle hit a target the size of a manhole cover when in his backyard, but miss a target the size of a swimming pool when at the local course?

The answer is down to a crucial component of learning that is often ignored but is very important indeed for teachers and for the results of their students.

If you’ve ever had to study the brain, there’s a good chance you devoted ample time to learning about the hippocampus. Arguably one of the more important neural regions, the hippocampus has been dubbed our “gateway to memory”. More specifically, in order to create new memories for particular facts or events (declarative memory), that information must flow through this seahorse-shaped structure like water flows through a pipe.

Interestingly, tucked into the base of the hippocampus is a small structure called the parahippocampal place area, or PPA for short. Decades of research have revealed that the PPA is constantly and continuously feeding information through the hippocampus, suggesting that whatever is being processed by the PPA must form an integral aspect of every new memory we create. So, what exactly is it processing?

Total recall

Researchers have found that the PPA essentially codes environmental information. For example, in one study, researchers showed participants dozens of images of everyday objects, popped the individuals into a brain scanner, and asked them to remember as many as they could.

Here’s the rub: some objects were displayed against a blank white background, while others were displayed within a realistic setting. The PPA showed enhanced activation any time an object appeared in a realistic setting – even though none of the candidates was explicitly paying attention to or trying to memorise that setting.

What this and studies like it tell us is that whenever we create a new declarative memory, the PPA automatically (and largely subconsciously) tags it with the characteristics of the surrounding environment at the time of learning. So, you may think you are focusing – and storing in memory – only the words of the book you are reading, but you are likely storing everything else in your vision at that time, too.

Beyond this, all sights, smells, sounds and textures present at the time of learning also make it into all newly formed memories. Although the precise process by which these sensory signals make it into declarative memory remains uncertain, the answer almost certainly lies within the thalamus.

A deep brain region tucked beneath and in heavy contact with the hippocampus, the thalamus processes all incoming sensory signals and almost certainly embeds these within memory. In fact, several theories suggest that new memories are almost exclusively these sensory signals (combined with contextual and timing elements).

All this is good news: these features can be used to trigger the recall of specific memories in the future.

An experiment conducted in the late 1970s demonstrates this process perfectly. In this study, deep-sea divers were asked to memorise a word-list while 20ft underwater. The following day, some divers returned to 20ft depth, while some remained on dry land – all were asked to recall as many words as they could remember from the previous day.

Even though no diver was explicitly focusing on the sights of the surrounding reefs, the sound of their air recyclers or the taste of their breathing masks while learning the words, this information was nonetheless encoded and embedded within their memories. As such, those divers who recalled their words underwater were able to recall 35 per cent more than those divers who recalled their words on dry land.

So where we do our learning forms an integral aspect of what we ultimately learn.

Researchers call this context-dependent learning, and it is the reason why we don’t always recognise colleagues when we bump into them outside of the workplace, why certain smells can trigger vivid memories and why returning to places from our childhood can bring back long-forgotten episodes.

State of flux

There is another dimension that appears to be embedded within each new memory we form, too: our internal environment.

A true story from a centuries-old physiology textbook brilliantly illustrates this concept. In this tale, a delivery man decides to indulge in a liquid lunch and, in his drunken state, misplaces an expensive package. Sober the following morning, he can’t remember where he left the parcel. After a fruitless day of searching, he gives up and decides to have a drink. Surprisingly, once he again achieves drunkenness, he is able to walk straight to the house where he had mistakenly left the package the evening before.

This same process occurs with any and all internal states. Joy, anger, sadness, hunger, thirst, exhaustion, heat, cold, tension: all newly formed memories are infused with the emotions felt during learning and will be easier to access in the future when in that same emotional state.

In short, how we feel during learning forms an integral aspect of what we ultimately learn. Researchers call this state-dependent learning, and it is the reason why sometimes we can’t “click” into work until after the third latte, why nerves occasionally lead us to forget even the most basic of ideas and why militaries conduct training drills under extreme stress and volatility.

Mix it up

But if things were forever tied to the specific circumstances in which they were learned, then human beings would be frozen, needing to relearn the same concepts each time we left the house. This means there must be a process by which information can decouple from a specific context and become accessible across any scenario.

Indeed, there is. The key is variety.

Earlier, we learned that declarative memory is our ability to remember specific facts or events. Importantly, declarative memories come in two distinct flavours: episodic and semantic. Episodic memories are facts or events tied to a specific time and place. For instance, I can recall dropping my niece’s ice-cream cake on the kitchen floor on the afternoon of her 5th birthday. Semantic memories are facts or events independent of any particular time and place. For instance, I know that the term “birthday” signifies the anniversary of an individual’s birth.

As we now know, each newly learned piece of information is strongly tied to the context in which the learning occurred. In other words, all new memories begin as episodic.

However, as we encounter the same piece of information across many contexts, we can separate it from any particular context and create a standalone fact. In other words, with repeated exposure across diverse scenarios, semantic memories emerge.

Here’s a practical example. Imagine a child undertakes four maths practice sessions across four different environments – a classroom, at home, in the library and during gym class. Each of these sessions will create a unique episodic memory, which includes relevant context and state details.

Eventually, however, this child can contrast these different episodic memories, siphon out any features common to them all and use those similarities to construct a brand-new, standalone semantic memory. In this case, seeing as the only commonality is the maths itself, her semantic memory will likely read: “Maths is an isolated skill that can be freely accessed across any environment.” Context and state dependency have been weakened.

Imagine, instead, that this same child undertakes four maths practice sessions, each in the exact same environment – a classroom. As before, each session will form a unique episodic memory and, eventually, these will be contrasted to filter out any commonalities. However, seeing as nearly every aspect of these memories is similar, her semantic memory will likely read: “Maths is a skill that can be applied only within a particular classroom using a particular chalkboard.” In this case, context and state dependency have been enhanced.

This is why my great-uncle Mooney couldn’t perform on an actual golf course. Seeing as he had only ever practised in a single location, his constructed semantic memory likely read, “Chipping is a skill done in a backyard using a concrete incinerator and an old shovel.” Had he simply practised across different locations, utilised different targets and occasionally switched his shovel for a rake, he might have been able to dissociate the essential chipping skill from any particular context and he might have performed fine in the official competition.

So, what does this mean for us (as teachers) and our students?

1. For discrete applications, match the training context to performance context

A final exam in the school gymnasium, an end-of-year presentation in the cafeteria or a big audition in the school theatre? When the circumstances of a performance are well known, it is worth matching the training context (as closely as possible) to the performance context. In this way, only relevant contextual information becomes intertwined with new memories.

Later, when it’s time to perform, this contextual information can be used as a guiding cue to more easily access and apply skills. Study after study has revealed improved performance among students who study in the testing location, teachers who train on site and athletes who practise in the official venue.

2. For flexible applications, mix up learning contexts

Multiple quizzes across different settings, a presentation delivered within numerous classes, a stage show going on a national tour? When performance circumstances are many, varied or largely unknown, it is worth training in as many different and diverse contexts as possible. In this way, ideas will become largely decoupled and dissociated from any particular context.

Numerous studies have revealed improved performance within novel environments among students who study across varied locations, teachers who train across multiple settings and athletes who practise across diverse fields.

3. Use senses to reinstate memories

If everything we see, hear, taste, smell and feel while learning becomes part of new memories, then we can leverage our senses to boost future performance.

Chew a particular flavour of gum while learning, then use that same gum to ease memory access during performance. Wear a particular soft shirt while learning, then use that same shirt to ease memory access during performance. Use a particular pen, spray a particular cologne, hum a particular melody – the possibilities are endless.

4. Music is a special case

Listening to music during revision can create a context-dependent effect that makes it harder for students to access relevant memories come exam time (where music is largely absent).

That said, listening to music during revision (when done correctly) can actually assist some students in focusing their attention, studying longer and learning more. As this theme was explored in depth in a prior column, I will not go into detail here (“Does listening to music during study help or hinder learning?”, Tes, 1 November 2019). Suffice it to say, students and teachers should be aware of these competing influences and make decisions accordingly.

Personally, I am willing to take the context-dependency hit on future recall in order to gain the sustained learning and attention boost – but that is simply my opinion. This should be considered on a case-by-case basis.

5. Apply context-dependent effects to help others recognise information

If you want others to simply recognise a piece of information (rather than need to recall it), then build a number of clear and consistent contextual elements around it. As we learned, even if people never consciously attend to these elements, they will become intertwined with the material being learned and can be employed in the future as guiding cues to trigger recognition.

Consistent assignment layout, classroom organisation, presentation slide design, timing scheme, etc: these contextual aspects will not replace learning but they can be used to trigger rapid recognition, thereby freeing up attention and cognitive resources to focus on other, more relevant information.

6. Be aware of state dependency when learning

According to some surveys, 99 per cent of students admit to cramming revision into the night before an exam (and maybe 1 per cent of students are liars). These cram sessions are typically supplemented by caffeine, junk food and other stimulants.

As we learned, these chemicals will become a part of the memories being formed. Accordingly, when students return to a more normal state, without the chemicals present during revision, memory and performance can drop markedly.

I’m nobody’s mother, so I’m not going to delve into a discussion of what people should and should not be putting into their body. What I will say is that it’s worth being cognisant of state-dependent effects. If students prepare while in a unique state, it might be worth trying to mimic that state come performance time. Conversely, if students know they will be in a unique state come performance time (say, nervous or excited), then it might be worth attempting to mimic this state during preparation.

Furthermore, nobody should be surprised when students who undertake revision while lying in a comfortable bed, eating snacks and listening to relaxing music struggle to recall relevant information while sitting in an uncomfortably stiff chair in a silent, highly tense room. It’s not that these students didn’t learn anything while studying: it’s simply that they tied this learning to the wrong context and have lost a wealth of guiding cues that ease the accessing of this learning.

So there you have it. As you teach your students, remember my Uncle Mooney: the where matters if you want them to remember the what.

Jared Cooney Horvath is a neuroscientist, educator and author, and is director of the Science of Learning Group. He is an honorary research fellow at St Vincent’s Hospital and the Melbourne Graduate School of Education

For references, see tes.com

This article originally appeared in the 6 March 2020 issue under the headline “It ain’t just what you learn, it’s where you learn it”

Other articles in this issue