Changing Our Minds Chapter 2. Learning – Scientists, Processors and Rats

← Changing Our Minds How children can take control of their own learning

Changing Our Minds Chapter 2. Learning – Scientists, Processors and Rats

Author: Naomi Fisher Publisher: London, UK: Robinson Publishing. Publish Date: 2021-2-4 Review Date: 2023-6-28 Status:📚

Annotations

What magic needs to occur for someone to go from ignorance to knowledge? We often assume that this happens by a process of instruction. We assume that if a person listens attentively enough to someone who is teaching, they will learn. If that doesn’t work, then you repeat the information, perhaps in a different format. And so on, until they get it.

To complicate things further, sometimes people can retain information and repeat it back, but with no understanding of what they have learnt. Why? What is actually happening in this process called ‘learning’?

What if you can’t reliably make learning to happen in another person by telling them what they should learn? In that case, we might need to reconsider the whole way schools are organised. Perhaps, rather than assuming children are not trying hard enough, maybe schools aren’t designed well enough to enable young humans to learn.

the basis of behaviourism. The most fundamental principle is that you can change behaviour through the deliberate manipulation of external events.

the experimenters, rather than the animals, controlled what learning should take place.

Psychologists were quick to see how this could be applied to children. In 1920, John B. Watson demonstrated classical conditioning on Little Albert, a nine-month-old baby. He did this by clanging an iron pipe and showing him a rat at the same time. Little Albert learnt to cry when he saw rats.

It wouldn’t get through a modern ethics committee, particularly since they didn’t decondition him afterwards. No one knows what happened to Little Albert but, by 1928, Watson was writing manuals on using behavioural principles to bring up children. He considered that there were only three unconditioned emotions: fear, rage and love. Everything else was learnt by behaviourist principles and therefore parents needed to be careful to condition their child in the right way.

Watson’s advice hasn’t dated well; in particular, his recommendation to interact with your children in a detached, business-like fashion. He did, however, start off a fashion for routine and habit-forming parenting which still continues today. Get the parent’s behaviour right, was his message. The child will learn the correct behavioural associations, and all will be well.

All the schools I have visited have used behaviourist principles. They use rewards and punishments to control children’s behaviour. Typical school rewards are grades, teacher approval, school prizes and good school reports. School punishments include bad grades, disapproval, being put on report, detentions and suspensions.

This works, on its own terms, for many children. Success is when there is a change in a child’s behaviour: perhaps they remember to hand in their homework after being given a detention for forgetting. The result of this apparent success is that many schools and teachers forget that it misses something out.

That something is the experience of the child themselves.

It doesn’t matter, from a behaviourist perspective, what the child thinks. A child might be complying with school requirements and yet feeling furious and resentful. When they are younger, many of them put up with it; as they get older, more of them start to show us how they feel.

The other problem is the child who doesn’t respond to behavioural strategies. There’s an assumption that the child can change, and if we just apply enough pressure, they will. So, if a child doesn’t respond to lunchtime detention, they are given an after-school detention. If they don’t respond to that, they are suspended for a day. If a day doesn’t work, let’s try a week. The onus is on the child to change, and the punishment increases until they do, or until they are permanently excluded from school.

Behaviourism assumes that the child can do the task … if only they would try hard enough.

They can’t comply, but it’s not necessarily for lack of trying.

By the 1960s, many psychologists were frustrated with the limits of behaviourism. It seemed like an overly simplistic way to understand complex human beings. Along came people like Jean Piaget and Lev Vygotsky, who were interested in how people thought, not just how they behaved. Behaviourism argues that humans learn in response to environmental stimuli; cognitive psychologists acknowledged that, in between the environment and behaviour, there was a thinking human being.

The question was, how could you measure their learning?

Studying thoughts is tricky. Unlike with behavioural change, which can be seen, you can’t ask a rat how its thinking has changed. Measuring a change in someone’s thoughts usually involves having to ask them or test them.

Psychologists solved this problem in two ways. The first way was through close observations of children learning naturally. Piaget watched his nephew and daughter as they grew and developed an understanding of the world. The other approach was experimental, devising paradigms to look at how well people learnt in controlled situations, or tests to work out what children knew.

Both of these methods are still used today. Psychologists around the world design experiments in order to work out how people learn, with some methodologies not unlike those used for the rats and the pigeons. These types of experiments frequently involve people learning useless information. It’s common for studies on memory, for example, to study people memorising lists, or abstract patterns. This is to avoid the thorny problem of people already knowing the information they are meant to be learning. However, it means that learning is taken out of context and is often devoid of any meaning for the learner. The assumption is that the same factors will come into play when remembering meaningful information.

In order to do experiments on the processes of learning, psychologists had to ignore many of the things which made learning interesting, and also ignore a lot of what people were learning.

Cognitive experiments on learning and memory are designed to focus on a particular question and, in order to do that, they simplify the situation down to the absolute basics. They strip learning of context, in order to understand the underlying processes. Which is useful if that’s your aim. But just as with behaviourism, when these theories are applied in education, there’s a tendency to ignore just how simplified the experiments were. Children can’t be stripped down to underlying processes.

Recently, a particular type of cognitive theory of learning has had a resurgence in education. Educationalists and cognitive scientists such as E.D. Hirsch and Daniel Willingham argue that we should directly apply cognitive models of memory to education and schools. They have the ear of government, and so these theories have led to widespread curriculum change in the UK and the USA.

It’s easy to see why they have had such success. The model of learning they promote sounds so simple and adapts so well to the school model. Willingham sees learning as information committed to long-term memory. This model suggests that we have two forms of memory. Working memory is short-term, is limited in capacity and cannot be greatly expanded by training. We use our working memory when we repeat a phone number to ourselves, punch it into the phone and then immediately forget it.

Our long-term memories, by contrast, can hold vast stores of information, but we can only bring small amounts up to our working memory at a time. Training does not significantly expand our working memory. What is possible is expanding what an ‘item’ in our working memory might be.

Here’s an example to illustrate what this means:

Here’s a set of letters; give yourself a few moments to look at them, then turn the page and see how many you can write down:

F I E N P D K M W P A Q B J O I

Typically, you’ll have remembered between five and eight. That’s a normal span for working memory.

Now try this one:

The enormous turnip jumped over the hedge.

How many letters did you remember here? If you got the whole sentence, it was 35.

And now try this one:

Jfd dscxdwers njeyy aqwqew ecxs ggnn okjko

Also 35 letters, but I’d be surprised if you got all of them. Why the difference?

In the ‘enormous turnip’ example, our brain combines the letters into words, and so we only need to remember the words rather than the individual letters. The meaning makes it easier to remember as well. Our expertise in reading changes how much information we can hold in our working memories. Words act as chunks of letters. In the final example, we don’t have any meaningful combinations and so we are back to the individual letters being the basic piece of information – and we’re stuck with our limited working memory.

There’s good evidence that people who have more information stored in their long-term memory are more expert than those who have less. Advocates of applying this model to education are fond of talking about experts – in particular, chess experts. Studies of chess players have found that expert chess players can remember the positions of pieces on a chessboard far better than novices — but only when the pieces are in meaningful positions (i.e., positions that might arise during a game). If the pieces are randomly arranged, the novices and experts both have similar trouble remembering the positions. In a real chess game, background knowledge and expertise mean that experts have an advantage over novices, because they can clump what they see into meaningful chunks – just like you did with those words on the previous page. This enables them to manipulate large quantities of information, unlike the unfortunate novice who has to remember the position of each individual chess piece.

The difference between experts and novices isn’t their working memory. In this model, the difference is simply the amount of information stored in their long-term memory.

In schools, those who advocate for this model suggest that the purpose of education should be to get as much information as possible into the long-term memory of children. The evidence, after all, shows that the difference between experts and novices is their long-term memory stores. They argue that, just like the chess players, having large stores of background information will enable children to manipulate more information in their working memory and to think like experts. Once the knowledge is there, so the theory goes, then creativity and higher-level thinking is possible.

These theories underpin the philosophy of several schools which have recently opened in the USA and UK. In the UK, Michaela Community School in west London is an example. Children at schools like Michaela are drilled in every lesson. They repeat material again and again, and are rigorously tested on it every day. For homework, they self-quiz. They follow along with what the teacher says in their books, are compelled to read over 10,000 words a day, and can be called on at any time to keep them focused on the lesson and to avoid the temptation to drift off into a daydream. Every moment of their day is controlled. It’s a lot like a memory laboratory, no distractions allowed.

It all makes perfect sense if you see education as an extended memory experiment. We know that information is forgotten over time and, in order to keep it in memory, it has to be repeated. At schools like these, that’s how the system works. It’s built on cognitive science, as they are fond of saying.

There are a few quibbles with this approach, which even those of us who lack the massive amount of background knowledge necessary to be designated an ‘expert’ might have noticed.

Remember the chess players? They are highly expert and have enormous amounts of background knowledge about chess. The theory says that this is why they are expert, and if we could teach children lots of background knowledge, they would become experts, too.

Except that the way to learn to play chess is to do it. It’s a process of playing, testing strategies, learning from others, perhaps reading books or websites. It’s never a question of sitting in a classroom learning lots of chess facts and strategies and waiting for the day in the future when you will be deemed expert enough to actually start to play chess. The endpoint for those expert chess players might be lots of chess configurations in their long-term memory, but most of those they will have worked out as they played. They will make sense to them because they deeply understand the structure of the game of chess. The information might not even be available to them verbally, but be coded in a different part of their memory (this is called implicit learning, and we use it when we learn how to do things like ride a bicycle, or swim, which we may not be able to explain verbally even when we can do it).

In addition, very few people are obliged to play chess. It’s not on the school curriculum. Those who become experts in playing chess are those who have chosen to put in the years of practice required. They are experts because they love playing chess and because people around them played chess with them. Their chess-playing has a purpose and context.

Like behaviourism, this sort of cognitivism is one level of explanation. It’s about how information storage might work in the brain – but that’s it. It doesn’t tell us anything about the context of learning, about how culture interacts with learning, or how people learn from each other.

In fact, cross-cultural studies have found that one of the things which schooled children do better than those who do not attend school is remembering lists of unrelated information. Schooled children learn how to remember things, even when they don’t make sense. It’s a skill that is only useful in a school context, where children are tested on information which they may not understand and didn’t choose to learn.

These theories of learning can’t tell us why one person is fascinated by algebra, while another loves history. They can’t even tell us why one person finds something easy while another works terribly hard and never gets above a ‘C’. They can’t tell us why Paul McCartney has music in his head.

To these cognitive scientists and educationalists, people are basically information processing units. Input goes in, they encode it, and then they can output it at a later date. Schools designed on this model focus on making the encoding as effective as possible, in the belief that that is what really matters. So, yes, approaches like this are based on cognitive science, but what their advocates don’t say is that the cognitive science they are referring to is based on experiments, which strip the context from real life. And what is learning, without context?

It’s memorising a list of random words.

There is something a bit odd about both behavioural and these particular cognitive theories, and it bothers me. They don’t actually seem to be about humans. It’s as if people’s learning is detached from their personalities and lives, and their memory store exists as a separate hard drive into which we can plug information through drills and repetition.

Luckily, not all cognitive scientists focus on pigeons and memorising lists. Some observe children learning with wonder at their capabilities and capacity for reflection. One of these is Alison Gopnik, an American developmental psychologist and philosopher. Rather than trying to teach children information and testing them on their retention, Gopnik designs experiments to show what young children already know, and how this knowledge interacts with their experiences.

For example, it turns out that, faced with two adults giving them contradictory information, children as young as three or four make logical choices about who to believe. They tend to believe their parent over a stranger, but they also take account of how confidently things are said. They’re more likely to believe someone who expresses their ideas with conviction than someone who sounds tentative.

And here’s where this research starts to get really interesting. Because it turns out that instructing young children can actually stop them learning. Elizabeth Bonawitz (a research psychologist who collaborated with Alison Gopnik) and her research group looked head on at the difference between exploratory learning and instruction. They used a toy which did several different things. Adults offered the toy to a child, and for half the children they told them explicitly how one part of the toy worked. For the other half, they ‘accidently on purpose’ showed the child how one part worked but gave no instruction. The children who weren’t instructed explored the toy and discovered all the other things it could do. The children who were instructed played with that toy in the way they were shown. The instruction seemed to stop them from looking for other possibilities. Other studies have found the same thing. When children are told how something works, they imitate. When they aren’t told, they explore. And in the second case, they learn more.

Of course, if you see the point of education as the acquisition of a particular body of knowledge, this doesn’t matter. Imitation is useful. Exploration and discovery aren’t on the cards until the children become ‘experts’. But if you’re concerned about children losing their joy in learning as they go through school, perhaps this offers one clue as to how that might be happening.

Gopnik’s writings are full of the children themselves. For her, children are never passive recipients; they bring their own prior knowledge and experience to every situation. From very early on, children actively try to understand what others are doing and why they are doing it, and adjust their behaviour accordingly. They are always active participants in their learning.

To Gopnik, schooling represents only one type of learning, and it’s not one that is superior to other forms. She suggests that other forms of social learning are both deeper evolutionarily and more sophisticated. From her perspective, Western middle-class parents are immersed in a parenting culture which focuses on moulding children to create a particular outcome (a mindset which Gopnik characterises as the ‘carpentry’ approach to parenting). This fits well with school culture, which has similar aims. This isn’t the only way to approach parenting, just as schooling isn’t the only way to approach learning. Young children are not schooled, and yet they learn. For most of human history, children were not schooled, and yet they became functioning adults and learnt how to live in their society.

Gopnik’s theory of learning and child development is sometimes called the ‘theory-theory’ because she argues that children construct their own theories about the world and use probabilistic reasoning to deduct likely answers. Fundamental to the approach is the idea that the child’s own perspective interacts with what they are experiencing, and that learning is always an active process. Studies show how young children learn through observation and listening, make predictions and test hypotheses. This science of learning bears a lot of resemblance to what we see in self-directed children. It’s not any less scientific because the children aren’t being asked to memorise lists.

When we see learning as an interaction between the child themselves, their pre-existing knowledge and their environment, then it becomes clear why each child’s learning trajectory can be so different, and how two children can learn such different things from the same experience.

In Western societies, this period of exploratory social learning is short-lived. Children are quickly channelled into school and formal learning, which is perceived to be more advanced and more important than informal learning. They are actively prevented from continuing to learn through exploratory play, as the focus in schools and parents shifts to literacy and numeracy. However, in some other countries, children do not all attend school and thus we can get some idea as to how children learn when they are not channelled in this way.

Studies in Guatemala by Barbara Rogoff and colleagues have shown that children who aren’t formally educated remain in a state of ‘alert awareness’ for longer than children who go to school. They learnt through observation and imitation more effectively than a control group of schooled American children, who waited to be shown how to do something before paying attention.

Many cross-cultural psychologists argue that we should view school as a cultural phenomenon. Schools teach culturally specific skills, which are then tested in culturally specific tests. People’s thinking and learning is always closely related to their cultural experience. We could see cognition and learning as something which develops as people learn how to live in their culture, rather than as something separate which can be abstracted and tested.

It’s about as far from the information processing model of cognitive science as it is possible to get.

Schools are carefully constructed learning environments which aim to deliver a particular type of learning. As such they take learning out of the context of life. Schooling involves an adult delivering specific actions towards a group of children with the aim of the children learning a particular set of knowledge or skills. This knowledge can in theory be used later but, right now, it’s being learnt because the school chooses to teach it, rather than because a child needs to know it now in order to live their lives. The school, not the child, decides what is important.

Take the example of reading. Schools decide that children need to learn to read around the age of five. They teach reading as a technical skill. The child learns to read words such as ‘cat’ and ‘hat’ so that, in the future, they will be able to read books. Most of them don’t learn to read through reading books of their choice, and they don’t learn to read because they want to understand the books. The skill of reading is separated from its purpose.

This separation of learning from purpose is not based on science. Nowhere do the studies show that people learn best when what they are learning is not meaningful for them.

Harriet Pattison, Senior Lecturer in Early Childhood Studies at Liverpool Hope University, has done extensive research on how children learn to read informally. She introduced me to a way of seeing learning to read as a cultural process rather than a cognitive one: ‘So you’d think of it more like learning to cook, because you’re helping your mum in the kitchen. You’re learning to read because you’re doing it as a family practice rather than because you’re making cognitive connections between sounds and symbols and so on, the way it’s thought about in mainstream phonics practice.’

This is the idea of communities of practice, first proposed by cognitive anthropologist Jean Lave and educational theorist Etienne Wenger in 1991. Communities of practice are where people come together with a collective purpose, or to do something together. Through their interactions with each other, they learn, and they share information and experiences. The learning is embedded in the practice, so literacy is embedded in the family reading together for various purposes.

Much of the learning that goes on outside school can be seen through the lens of a community of practice. Children learn through doing, and their learning is enmeshed with its purpose. The two are never separated.

In the institution of school, the child’s culture and internal world hardly matter unless they interfere with the instructional process. In fact, many early schools, particularly in America and Australia, were deliberately developed in order to wipe out indigenous cultures. School introduces a specific set of culturally specific standardised outcomes and prioritises them above all else.

Learning by immersion is messy. You can’t really tell how it’s going for quite a long time; no one brings out lists of vocabulary; there’s no logical order. You might learn how to say, ‘Where can I find the Wi-Fi password?’ before you can say, ‘What’s your name?’ You learn what is around you. If you go to a park, you learn how to say ‘slide’ and ‘playground’ right there. You use whatever you learn as soon as you can, because you need it. If you wait to be an expert before you start speaking the new language, the odds are that you will never speak.

Language learning is a fantastic way to gain insight into how we learn in the real world, because speaking French in France is a real-life assessment. It’s useful, flexible and meaningful. It’s not so easy to see if someone truly understands Maths, or History, or English Literature. Instead, we take test results to be a marker of someone’s ability. But passing standardised tests in French indicates very little about your actual ability to communicate in France.

Her French is suited to her environment and her needs. Even if she never becomes able to pass a French exam, her French is useful and meaningful.

Learning like this isn’t neat or predictable, just like young human beings. When French is learnt in order to pass an exam, then it is no longer primarily a way to communicate. Instead, it is reduced to a set of tasks necessary for the exam

As theories of learning become more sophisticated, the question of why keeps coming up. Why do we do what we do, and why do we choose to learn, or not? Traditionally, human cultures have answered this question through communities of practice. We do what we do, because it is how we live our lives. We learn through doing, because those are the skills we need in order to live well.

Schools answer this question by creating a set of circumstances which they hope will give children reasons to learn. They need to do this because the social learning environment – a community – is sapped by dissociating learning from context. In Chapter 3, I’ll look at the psychological theories behind why we do what we do.

The question of why we do things (or don’t do things) is fundamental to what it means to be human. That’s why education is not as simple as designing the right curriculum and watching the children learn. That might work for the rats running through their mazes, but human learning is far more complex. For humans, meaning and context are an integral part of why and what they learn. We ignore this at our peril.

Notes

Amount: 21

🔮 The Cosmos