Holly Shale prided herself on her memory: when she was fifteen, she won the silver medal in her junior memory championship for recalling a random sequence of thirty-five words. She had memorised all the capitals and flags of Africa. She remembered the atomic number and symbol of every element on the periodic table. She was a wonder child.
But when she was twenty-four, a car crash damaged her brain’s hippocampus (a structure crucial for memory formation), producing anterograde amnesia. Unlike retrograde amnesia (the inability to recall past memories), anterograde amnesia is the inability to create new memories. Holly is thus perpetually locked in the present, powerless to continue the story of her life. For Holly, there are no tomorrows; there is only now and the feeling that something terrible has happened.
In March 2021 I video-called Holly at her home in east London. Her mother, Elizabeth, took the call in the kitchen. Holly was hunched over the counter in a blue apron, whisking eggs and staring at a cookery book. She was tall and wiry, with almond eyes and dark frizzy hair. Elizabeth introduced us and then said to Holly, ‘I think this will be the one, darling. What number is this?’
‘I’m not sure,’ she said. ‘I keep getting the mixture wrong.’
‘Ah, right. Well come sit down and we’ll make the cake later.’
There were four cakes on the kitchen table; according to Elizabeth, she will bake a cake, serve it up and head straight back to the counter to make another one. Elizabeth accompanies her when she leaves the house to stop her buying the same thing again and again. She is forty-two, but still thinks she is twenty-four.
‘I’m a bit like Dori from Finding Nemo now,’ she laughs. ‘It’s not too bad, as long as I’ve got someone with me. In the beginning it was awful. Every day was a blank slate. I would read a newspaper article and instantly forget what it was about. I would wash my hair over and over again. I just couldn’t function. The only reason I’m able to tell people any of this is because of the care and support of my mum.’
As I listened to Holly’s story, noting how her mother often guided her towards what she was trying to say, I was struck by how resilient and courageous she had become. Here was a woman whose mind was caught in an abbreviated version of life, a woman who now relied on others to be her memory. I was reminded of my grandfather, whose battle with Alzheimer’s disease obliterated his memory in the reverse direction, and how, for a time, he too was forced to remember life vicariously through his family. But what struck me most about Holly’s plight was that it shattered all our preconceptions about memory.
The idea that memory is about the past – a giant filing cabinet in which the files are past experiences, ready to be accessed with the right cues – stretches all the way back to ancient Greece. Plato viewed memory as a wax tablet: our memories and thoughts could be stamped, stored and retrieved for later use. Socrates likened memories to birds in an aviary, ‘some in exclusive flocks, some in small groups, and some flying alone, here, there and everywhere among all the rest.’1
Surprisingly, it turns out that memory has almost nothing to do with the past. Memory is about the present and the future. From the moment you are born you are constantly encountering new phenomena in the world, which your brain interprets and organises into experiences. They include our physical surroundings, our bodily sensations, the norms and values of social life, and the ever-changing stream of subjective feelings that only ceases at death. When we reflect on past experiences, we are not merely taking a trip down memory lane, we are also accessing an earlier brain state in order to influence our current or future behaviour. The act of remembering something is a type of learning; it’s a constructive process, allowing us to predict the future based on memories of the past. This is why we are such poor witnesses in court: our future-orientated memory is better at imagining possible scenarios than remembering real ones.
It’s often said that memory makes us who we are; but evolution doesn’t care about the personal history of each human brain. It cares only about the changes that have occurred throughout the brain’s 7-million-year journey, and whether those changes give us a better chance to survive and reproduce. Consequently, the story of human memory contains a number of important subplots – each concerned with a monumental shift in memory.
Every animal has some kind of memory. Elephants, for example, can identify at least thirty of their relatives and are known to remember the Maasai tribe members who abuse them in ritual displays of masculinity. Crows including ravens and magpies recognise individual human faces, ignoring those considered friends and scolding those deemed foes. Single-celled organisms such as the paramecium, a water-loving animal covered in microscopic hairs, will learn to avoid being electrically shocked in lab experiments, encoding the memory in a web of organic chemicals.
These observations tell us that memory is old. Very old. Most researchers believe that human memory was born in the ancient oceans. Free-floating neurons coalesced to form so-called nerve nets, seen today in jellyfish, sea anemones and corals. Scientists are very interested in these creatures. If we can decipher the code by which their neurons generate memory, we may one day be able to place that capacity onto a silicon chip and implant it into patients with memory disorders. Memory ‘neuroprosthetics’ might sound far-fetched, but governments in the Western world are investing tens of millions of dollars into it (of which more anon).
It used to be thought that all kinds of memory – recalling your name, where you live, what you ate for dinner last night, how to swing a tennis racket – were all the same mechanism in the brain. Memory was regarded as a simple phenomenon concerned only with the automatic and unconscious storage of information. Now, however, we are learning that memory has experienced an evolutionary path as long and winding as every other faculty of the mind. Scientists believe that seven different memory systems evolved in order to help the brain solve problems and exploit opportunities encountered by early humans in the distant past. These memory systems are thought to have started their journey in early animals, but then evolved independently in humans some 3 million years ago, when four species of human ancestor – Australopithecus africanus, Australopithecus bahrelghazali, Australopithecus deyiremeda and Kenyanthropus platyops – roamed the earth.
The first system of memory to evolve was reinforcement memory: remembering predators and the location of food sources.2 In the deadly proving grounds of the savannah, learning what – and who – our enemies were and how to feed our children was critical. Modern examples can be seen in the behaviour of any prey–predator relationship you care to name: zebras must remember to avoid lions, deer to avoid wolves, rabbits to avoid foxes, and so on. A long history of bloodshed slowly taught animals which actions paid off and which came at a price. The biology behind this type of learning was a mystery until Ivan Pavlov (1849–1936) famously showed that dogs could learn to associate a particular stimulus (a bell) with another event that followed (being fed), causing them to salivate. Pavlov called this response psychic secretion, but psychologists now know it as reinforcement memory. The neurobiology behind it has always been a puzzle. We know that the cerebellum (the ‘little brain’ located at the back of the brain, just behind the brain stem) must be involved, given that remembering movements – for example, the movements needed for escape – is clearly important. And we know that other brain regions such as the amygdala are involved, given that fear and other emotions also play a role. But that’s it. Like much in neuroscience, we know a lot about what and where but much less about why and how.
Closely related is the second system – that of navigation memory: our ability to build a cognitive map of the world. While Google Earth is certainly impressive, its processing power pales in comparison to the millions of place cells in the hippocampus, each firing in response to the smallest change in our environment – a pointillism of sights, smells, objects and places. They belong to a class of neurons called pyramidal neurons, which become active when we enter a specific location, collectively mapping the location in a pattern of neuronal activity. We take this memory for granted – remembering the journey to work or where we parked the car rarely feels impressive – but life has gone to extraordinary measures to invent it. Migratory birds for instance use a magnetic compass in their eye for navigation memory. Honeybees use a map of polarised light, visible even on the cloudiest of days. Bats and many other mammals use echolocation. In humans, place cells allowed our ancestors to carve out new territory and become world travellers, journeying all the way from Africa to Asia.
Next came something called biased-competition memory: the ability of various memories to compete for our attention. The brain pulls off this trick by encoding memories in neurons that suppress the activity of neighbouring neurons. In fact, memories formed within six hours of each other activate the same group of neurons, which then try to suppress neurons encoding memories formed in the more distant past. In this way, our brains can select which memories from our past are worth keeping. When early humans had to decide between various hunting strategies, weighing up the pros and cons of each based on previous experience served them well. More than 2 million years ago, Homo habilis used this faculty to ambush herds of antelope and gazelles. Experts now believe that they hid in trees and leapt on their prey from above, spearing the animals at point-blank range.3 For this method to prevail, memories of successful hunts would have had to outcompete memories of failed hunts.
Competition memory doesn’t mean that we lose out-of-date memories, only that we can build on them. By learning which situations required an up-to-date approach and then creating a bias towards that approach, human memory developed a level of ingenuity not seen in other primates. So can we choose what we forget? Yes, according to Jeremy Manning at Dartmouth College, New Hampshire. He’s shown that simply telling people to ‘push thoughts out of their head’ is enough to make them forget a list of words, and he is now helping patients with PTSD (post-traumatic stress disorder) to forget painful memories that keep resurfacing. The neuroscience underlying this process is unclear, but it probably takes place at the synapse. When memories are formed, synapses strengthen their connections by increasing the number of a particular type of receptor called AMPA. These receptors are notoriously unstable, constantly shuttling in and out of the synapse. To forget something, our brains must actively remove these receptors, weakening and eventually destroying the connections at the synapse. Manning observes: ‘Forgetting is typically viewed as a failure in some sense, but sometimes forgetting can be beneficial, too… we might want to get old information out of our head, so we can focus on learning new material.’4
Fourth to evolve was a system for remembering how to grasp certain objects in the thin branches of the trees, called manual foraging memory, or muscle memory. This would later become an important way of creating objects such as stone tools and fishhooks. While reaching for your morning coffee certainly doesn’t feel dependent on memory, experiments on a patient named Henry Molaison (known to psychologists as HM) have proved otherwise. In 1953 HM had two-thirds of his hippocampus surgically removed to treat his epilepsy. The procedure worked but left him in the same predicament as Holly Shale: unable to form new memories. In 1998 the neuroscientist Reza Shadmehr and his colleagues trained HM to grasp a robotic handle that was designed to resist being handled. After the experiment, HM quickly forgot who the experimenters were and claimed never to have seen the robotic device before. And yet, when the team asked him to try grasping the robotic handle again, he knew exactly how to perform the task without any help.5 In the brain, this memory relies heavily on the cerebellum; specifically, a special class of neurons within the cerebellum called the Purkinje cell, named after the Czech anatomist Jan Evangelista Purkynĕ who discovered them in 1839. Purkinje cells are instantly recognisable due to their vast, intricately branched network of dendrites, which creates muscle memory by constantly remodelling itself in response to neurons projecting from the spinal cord. Some scientists believe this memory evolved to let us quickly rebuild muscle mass during times of strife, freeing us from having to maintain a constant warrior physique.
Once we had evolved memory based on touch, evolution turned to sight and hearing. For this we evolved feature memory, a sensory memory for important visual and auditory cues. For example, the sight of a cave usually signified sanctuary, the growl of the sabre-toothed tiger peril. For this system to work, human brains had to interpret sight and sound at the same time, even though they are very different senses. The brain does this by merging electrical activity from the visual and auditory cortex, combining it in a special population of neurons called polychronous neuronal groups (PNG). These neurons ‘polychronise’, that is, produce a regular and repeating firing pattern, which we think lets the brain combine different sensory inputs into a single experience. Sometimes this merging goes too far and generates synesthetes: people who taste words, hear colours and see sounds. Synesthetes have an enhanced memory compared with the rest of the population. Some believe their richer perceptual experience improves their feature memory.
As human species became more diverse, between 500,000 to 2 million years ago, their territory expanded to occupy almost all of the planet’s habitable regions. Waves of human migrants, including Homo erectus, Homo heidelbergensis, Neanderthals and Homo sapiens, left Africa, swept across the Red Sea to the Middle East and Australia, up into Europe and across the Bering Strait into the Americas. Along the way Homo sapiens developed large brains, we lived longer and we evolved social thinking to form tribes. And for all this we evolved goal memory: the power to remember certain objectives and targets of action, a memory to allow our thoughts to create the future. Research points to the lateral prefrontal cortex for this type of memory: neurons in this region fire in response to reward and decision-making, allowing our brains to code the value of different goals and decisions for the future. Its applications were boundless. If a fellow Homo sapiens remembered to plant seeds instead of eating them, the whole tribe would benefit. If a fellow Homo sapiens remembered which rivers had the most fish, the knowledge could be passed on. And if a fellow Homo sapiens laid claim to more than their fair share of resources, that was worth remembering in order to prevent it occurring again.
Which brings us to the last system to evolve: the memory of one’s behaviour and its consequences in social systems, known as social subjective memory – another ingredient vital to the evolution of social minds, partly forming the basis of human morality. This kind of memory is thought to have evolved only a few hundred thousand years ago, as a means of adapting in early Homo sapiens societies. We all have to curb and sometimes alter our behaviour to make group living more harmonious. And those constraints depend on memories of good or bad outcomes in social situations. Let’s say you are hosting dinner for new neighbours and make a joke about accountants. They laugh politely, but you later discover during the course of the dinner that one of your guests is, in fact, an accountant and was offended by the joke. If you are a good neighbour, chances are you won’t make a similar joke in the accountant’s company again, not out of shame but because you remember the negative consequences of your behaviour. This is what psychologist Daniel Kahneman calls the ‘remembering self’. It’s how our brains plan for the future. The next dinner won’t be awkward. You know because you can already see it. As Kahneman notes, ‘We think of our future as anticipated memories.’6 The brain does this using a region of the parietal lobe called the precuneus, though the underlying neurobiology of this memory remains unclear.
To see how all these systems work together today, let’s imagine a typical day in the life of a modern human. Every day you wake up in the morning, get ready for work and walk out of the front door. What confronts you is not a savannah or wilderness; there are no wild dogs or big cats waiting to have you for breakfast, but typically a road with some cars gliding past. Here is where your reinforcement memory kicks in. Car accident deaths are common and deadly, making vehicles the most immediate ‘predators’ preventing you from reaching your ‘food source’ or job income.
After safely crossing or avoiding the road, you now need your navigation memory to direct you to work. Previous journeys have of course hardwired the route in your brain, but you still need your hippocampus’s place cells to light up in just the right sequence to get you there. If you walk or take public transport, you might decide to use the time on your way to work browsing on your phone or reading a book. The various options competing for your attention are your biased-competition memory speaking up. Perhaps you have an email that is overdue for a reply, in which case you’d better put the book away.
Everything you’ve been doing thus far on your journey depends on muscle memory, or manual foraging memory, without which you wouldn’t even have remembered how to brush your teeth or put on clothes. And if you’re planning a trip to the gym after work, your capacity for this type of memory will determine how successful your session is.
By now, you’ve arrived at work, an office filled with the sound of people talking, telephones ringing, pages turning, computer keys tapping and electric fans whirring. Add to this the sight of screens flickering, colleagues mingling, deliveries arriving and meetings in full flow, and were it not for your feature memory, meticulously categorising every auditory and visual detail, you might almost succumb to sensory overload.
Settled in at the office, you can now get on with the business of actually doing some work. No doubt you have a long list of objectives and targets to meet, all of which will demand your goal memory: your ability to use your thoughts to create the future. The key difference between you and one of your ancestors in this scenario is that your ancestors were hunting game, not clients.
All that’s left for your memory to worry about now is how you’re coming across to your colleagues and clients. Like a referee at a football match, your social subjective memory is keeping a close eye on your behaviour and its likely consequences. A flood of anticipated memories fills your mind, showing you all the ways in which your day could be a triumph or a shambles.
Memory gets us through each day in more ways than we know, and these systems are probably only the tip of the iceberg. Overall, the history of memory in humans is closely tied to the history that made us the complex organisms we are. There are 37 trillion cells in the human body, controlling everything from moving to thinking. Memory developed through stages of evolution to complement these functions and help humans react appropriately to every situation they encountered. A brain that can learn and memorise and make decisions based on prior experience is a quantum leap in biology. It’s what makes our species responsive to the environment rather than reflexive, purposeful rather than aimless. Memory is a story that has been endlessly edited, updated and retold for a new audience.
Throughout history, three important breakthroughs shaped the course of memory research. In 1749 David Hartley proposed that memory was a product of the nervous system and other biological processes. In 1904 Richard Semon proposed that experience leaves a physical trace in the brain, but that these memory traces were imperfect copies of experience, prone to distortion. And in 1932 Frederick Bartlett proposed that what people remember is based on our cultural values and the way we perceive the world around us: white Americans, for example, tend to stress the role of individuals in their descriptions of memories, while East Asians focus more on social interactions. There were more advances in the years following – in 1949 Donald Hebb showed that durable connections between neurons could encode memory and, in 1974, Eric Kandel showed that the substance of memory is protein molecules and neurotransmitters obeying the principles of biochemistry. But despite all this progress, the biology of memory remained unclear.
For simplicity, we now categorise memory into several different types. Working (short-term) memory is fairly self-explanatory. It refers to all the information you temporarily hold in your head, for instance remembering numbers when you do a sum or holding a person’s address in your mind. It lasts around ten to fifteen seconds and can only store a limited amount of information: most people, for example, can only accurately hold seven numbers in working memory at any given time. It relies on the frontal lobes in particular but also appears to involve most of the cortex; the neurophysiology is poorly understood. A useful analogy is to think of working memory as RAM in a computer, whereas long-term memory would be the hard drive. Not everyone’s working memory is the same. One in ten children has working memory deficits, meaning they perform poorly in school despite being intelligent otherwise.7 And though not proven, many experts believe working memory deficits are linked to ADHD (attention deficit hyperactivity disorder).8
Then there’s long-term memory, which can be divided broadly into explicit (or declarative) and implicit (or procedural). Explicit memory refers to factual information about your life – your parents’ names, where you grew up, what you do for a living – and can be consciously recalled (or ‘declared’). It can be further divided into episodic memory, the memory of past experiences and events, and semantic memory, the memory of knowledge gathered over a lifetime. Implicit memory is the memory of how to perform certain tasks, such as walk, talk, ride a bike or play the piano. Starting from very early in life, implicit memory can become so ingrained it is almost automatic. No one knows how the brain stores long-term memory, but it is almost certainly not in one place. Some of the most common brain diseases such as Alzheimer’s and stroke obliterate long-term memory by destroying vast swathes of brain tissue. It seems that long-term memory behaves more like a Wi-Fi network than a hard drive.
Many scientists argue that memories are physically encoded in the brain by a network of neurons in the hippocampus and cortex. The fundamental idea is that you have an experience – say, you lose your virginity – and a memory of the experience is sent to the hippocampus in the form of electrical signals travelling across synapses. The memory may then reside in the cortex as long-term memory, or it may not. That depends on complex molecular processes involving neurotransmitter receptors, enzymes, genes, epigenetics, and so on. Until recently this idea was mainstream neuroscience. Pursuing it was thought to be our best hope of finally understanding how memory works.
But I and many other neuroscientists are now convinced that it is wrong, thanks to a brilliant investigation by the neuroscientists Nikolay Kukushkin and Thomas Carew in the journal Neuron.9 Kukushkin and Carew were eager to challenge the mainstream theory when they noticed a fallacy in the way we talk about memory. The language that scientists use to describe memory – ‘memory retrieval’, ‘memory acquisition’, ‘memory trace’, ‘memory consolidation’ and so on – presupposes the notion of ‘a memory’, something that is separate from the person doing the remembering. Do you see the problem? The language defines memory as separate from the mind instead of an integral part of it; when the truth is, your brain doesn’t store or retrieve memories. It is memories.
If this has left you confused, or so bewildered that you feel we are tumbling down the rabbit hole, you are not alone. The world’s leading scientists don’t understand it either. The best explanation is that memory must be a change from one brain state to another. And so the act of remembering, Kukushkin notes, is ‘just a reactivation of connections between different parts of your brain that were active at some previous time.’ Neuroscience now accepts this idea, but it is astonishing to most people when they first ponder it. One almost feels like Proust as he bites into the famous madeleine, reviving his childhood memories and distorting the boundaries of time:
But at the very instant when the mouthful of tea mixed with cake-crumbs touched my palate, I quivered, attentive to the extraordinary thing that was happening in me. A delicious pleasure had invaded me, isolated me, without my having any notion as to its cause… Where could it have come to me from – this powerful joy? I sensed that it was connected to the taste of the tea and the cake, but that it went infinitely far beyond it… And suddenly the memory appeared. That taste was the taste of the little piece of madeleine which on Sunday mornings at Combray… my aunt Léonie would give me after dipping it in her infusion of tea or lime-blossom.10
Proust’s bewitching epiphany reveals another puzzling feature of memory. It suggests that neurons, synapses and molecules can sense the passing of time. Though we might recognise this phenomenon from our body’s circadian rhythm that is its own kind of timekeeper – an automatic process designed to regulate sleep and wakefulness – our new understanding of memory, on the other hand, reveals that the brain resurrects our conscious experience of past realities whenever it wants.
This suggests that the brain also invented memory to measure time. Indeed, we now know that the hippocampus contains what researchers call ‘time cells’: neurons that represent particular moments in time as well as the location of specific experiences. If the hippocampus is damaged, as it is in patients with Alzheimer’s and other memory disorders, people can only accurately recall the passage of time for short intervals. This is why Alzheimer’s destroys short-term memory first. Yet the question remains: how does the brain measure long time periods?
This is where things get a little complicated. It turns out that the brain contains multiple inner clocks, some for tracking very different time periods (1 or 10 minutes), some for tracking very similar time periods (10 or 12 minutes), some for tracking short time periods (5–10 seconds), and some for tracking longer time periods (20–60 minutes). And all of these clocks compete for our attention.
To see how the brain discriminates between them, a University of California neuroscientist named Norbert Fortin and his colleagues trained rats to signal time.11 To do this, they taught the rats to choose between different odours: odour A for a one-minute interval, odour C for a twelve-minute interval and odour B for the intermediate interval. As an incentive, tasty treats were given to the rats that got it right.
Before the experiment, the scientists administered a drug that briefly shuts down the hippocampus. This kind of interference tells them whether the hippocampus is required to perform the task accurately. They found that the hippocampus is necessary for discriminating between similar longer periods of time (say, 20–24 minutes) but not for discriminating between events occurring second by second. This is evidence, according to the authors of the study, that the hippocampus contains an inner clock for separating individual experiences during a sequence of events. So when, for example, you finish cooking supper and then sit with the family at the dinner table, your hippocampus is in charge. But when sensing the difference between cooking a steak that’s rare versus medium-rare, something else in the brain is active.
That something is thought to be the striatum, a region found deep within the front part of the brain. Most scientists believe that its neurons evolved to code time with extreme precision. Without it, we would be lumbering through everyday tasks like a paralytic drunk. Just imagine trying to write an email without remembering the first half of it. Imagine tying your shoelaces without remembering why you put your shoes on. Life would be almost impossible. For all its stages of evolution, memory is useless without a strong grip on time. This fact was vividly portrayed in the 2000 Christopher Nolan thriller Memento. The protagonist, an insurance broker named Leonard Shelby, is hit on the head by an intruder and soon develops anterograde amnesia. He then becomes obsessed with catching the attacker, who he believes raped and murdered his wife, but because his memory is wiped clean every fifteen minutes, he must have every clue to the murder painstakingly tattooed to his body. At one point in the film, he perfectly captures the essence of his distress: ‘How am I supposed to feel, when I can’t feel time?’
One of the most astonishing things about memory is how much it is shaped by culture. One study found that white Americans are more prone to false memories than people from Eastern cultures.12 The pioneer of this field, Frederick Bartlett, demonstrated the importance of culture by asking English subjects to memorise and then recount a North American folktale called ‘The War of the Ghosts’:
One night two young men from Egulac went down to the river to hunt seals, and while they were there it became foggy and calm. Then they heard war cries, and they thought: ‘Maybe this is a war-party.’ They escaped to the shore, and hid behind a log. Now canoes came up, and they heard the noise of paddles, and saw one canoe coming up to them. There were five men in the canoe, and they said:
‘What do you think? We wish to take you along. We are going up the river to make war on the people.’ One of the young men said: ‘I have no arrows.’ ‘Arrows are in the canoe,’ they said. ‘I will not go along. I might be killed. My relatives do not know where I have gone. But you,’ he said, turning to the other, ‘may go with them.’ So one of the young men went, but the other returned home.
And the warriors went on up the river to a town on the other side of Kalama. The people came down to the water, and they began to fight, and many were killed. But presently the young man heard one of the warriors say: ‘Quick, let us go home: that Indian has been hit.’ Now he thought: ‘Oh, they are ghosts.’ He did not feel sick, but they said he had been shot.
So the canoes went back to Egulac, and the young man went ashore to his house, and made a fire. And he told everybody and said: ‘Behold I accompanied the ghosts, and we went to fight. Many of our fellows were killed, and many of those who attacked us were killed. They said I was hit, and I did not feel sick.’
He told it all, and then he became quiet. When the sun rose he fell down. Something black came out of his mouth. His face became contorted. The people jumped up and cried. He was dead.13
Bartlett knew that this story would appear disordered and somewhat inscrutable to Anglo-Saxon minds. Every culture, he argued, has what he called a schema: a framework of thought that helps organise and interpret knowledge. If we are unsure about what has happened, we use our own schema to fill in the gaps and rationalise what we are trying to remember. And indeed when Bartlett asked his subjects to reproduce the story, he found that many features had been added, subtracted, simplified and transformed. Here is one attempt:
Two youths went down to the river to hunt for seals. They were hiding behind a rock when a boat with some warriors in it came up to them. The warriors, however, said they were friends, and invited them to help them to fight an enemy over the river. The elder one said he could not go because his relations would be so anxious if he did not return home. So the younger one went with the warriors in the boat. In the evening he returned and told his friends that he had been fighting in a great battle, and that many were slain on both sides. After lighting a fire he retired to sleep. In the morning, when the sun rose, he fell ill, and his neighbours came to see him. He had told them that he had been wounded in the battle but had felt no pain then. But soon he became worse. He writhed and shrieked and fell to the ground dead. Something black came out of his mouth. The neighbours said he must have been at war with the ghosts.
This may not strike you as very surprising, because we all know that memories can be reshaped like a game of whispers. Memories change each time we remember, often more than we realise. In fact, research shows that when we tell stories we actually change little details depending on the listener’s personality and political outlook, a phenomenon known as the audience-tuning effect. This then later changes how we ourselves remember the story, further spinning the wheels of disinformation. To counter this, most people try to rehearse a memory immediately after an event. But even this act makes us more susceptible to misinformation later, a phenomenon known as retrieval-enhanced suggestibility, which is a real nuisance to police officers who later discover that an eyewitness generated details that were completely false. The disappointing truth is that human memory is only ever as reliable as the most recent story we tell ourselves.
Bartlett’s discovery that human recollections are coloured by social constructs including race, language, education, and the lived experience of the individual is one of the most exciting ideas in neuropsychology. It proves that none of us is in a privileged position to define what is true. Our memories, like our emotional and social minds, are bound together as yet another adaptation of our culturally evolving minds. This is not to say that ‘what really happened’ does not exist; it does. The bedrock of good science is, after all, objectivity. The fact that different cultures remember the world around them differently says more about our social behaviour than our memory. It also says that for Homo sapiens the purpose of memory is partly to serve the collective.
There is a kind of memory that has been with us since the early days of our species; the kind stored in leather scrolls and wax tablets, books and art, computers and smart phones. We call it collective, or social, memory. It arises when individual memories are shared by members of a community, and when a community then enshrines those memories in public symbols and institutions.
The first to study it seriously was the French philosopher Maurice Halbwachs (1877–1945), who argued that memory only works within a collectivist context – that memory paints images in the mind only ‘in accord with the predominant thoughts of society.’ Unlike the types of memory that evolution took eons to devise, collective memory is constantly changing depending on the historical narrative of our species. As a group, we remember all kinds of events: the Holocaust, the fall of the Berlin Wall, the moon landing, the end of apartheid, 9/11. This is because members of a group – from small groups such as families to large groups such as nations – usually share similar memories. Americans, for example, remember their country’s dark history of slavery; Shia Muslims remember the death of Ali (regarded as the first imam after the prophet Mohammed); and my family remember my grandfather’s long struggle with Alzheimer’s disease.
As social animals, we are continually adjusting our memory of the past to conform to social precepts. Memory is a social tool. When humans share memories with one another they are creating a shared reality, one that is concerned less with the fidelity of the memory and more with whether or not the speaker and listener belong to the same social group. Like individual memory, collective memory is represented by neurons and synapses in the brain. In an attempt to discover the underlying neurobiology of this process, Micah Edelson, at the Weizmann Institute of Science in Rehovot, Israel, and his colleagues asked thirty people to watch an eyewitness-style documentary about a police arrest in groups of five.14 A few days later, the subjects returned to the lab and completed a memory test about the documentary while being scanned with an fMRI machine. At this point, though, the researchers got creative. The subjects were offered a ‘lifeline’: the answers given by their fellow observers. That’s what they thought at least, because the answers were actually composed of false memories.
Astonishingly, nearly 70 per cent of the subjects changed their answers to fit in with the group. The question is, did they knowingly change their answer to conform? Earlier studies of this kind have shown that people do indeed bow to social pressure and give a false answer, even though they privately hold a different belief.15 So the researchers asked the subjects to take the memory test again, only this time they told them that the previous answers given by their co-observers were in fact randomly generated by a computer. No social pressure, so no need to lie. The result: some of the subjects admitted what they really remembered, but 40 per cent remained convinced by the false memories. They still felt completely confident that their memory was accurate.
When Edelson and his team looked at which brain areas might be active during such puzzling behaviour, they found a strong co-activation between the hippocampus and the amygdala. This suggests that social pressure can actually change your memories. The endless desire to fit in, these scientists think, may even be the brain’s stamp of approval for a memory to be formed at all. As to why our brains evolved this way, they conclude:
Memory conformity may also serve an adaptive purpose, because social learning is often more efficient and accurate than individual learning. For this reason, humans may be predisposed to trust the judgement of the group, even when it stands in opposition to their own original beliefs. Such influences and their long-term effects… may contribute to the extraordinary levels of persistent conformity seen in authoritarian cults and societies.
Perhaps unsurprisingly, collective memory is coloured more by a person’s lived experience of an event than by later accounts in the history books. Lived memories describe events in personal terms and are often imbued with strong emotion and meaning. Non-lived memories, on the other hand, are usually described in abstract, matter-of-fact terms. Psychologists interested in this phenomenon study what they call temporal construal theory, which argues that the more psychologically distant one is from a historical event – that is, the more non-lived their memories are – the less likely they are to appreciate fully what really happened. This matters because it affects how we remember history. Someone may know a great deal about the communist horrors of the Soviet Union, for example, but this is a poor substitute for someone who actually lived through them. Equally, some young Germans have been found to excuse their grandfather’s Nazi affiliations because the fascist horrors of Nazi Germany and the murder of 6 million Jews is – for them – an abstraction rather than a lived experience.16
Unsettling though it may be, the truth is that memory is not what we think it is. It twists and distorts our perception of reality in countless ways. And if our brain’s tribal nature is not ultimately overcome, it may continue to deceive us for a very long time.
As we have seen, memory was born in stages of evolutionary development to help our ancestors respond to an ever-changing material and social world. Along the way, our brains understood that the memory of past experiences could be put into the service of the present and the future, ensuring reproductive success by anticipating the sequence of events that a human life was likely to contain. Our minds have trouble understanding memory, so most people falsely identify it as something living within us – something contained, retrieved and then filed away in the brain. But in reality, we are our memories. The brain is as much memory as the skeleton is bone. As the novelist John Irving has put it, ‘You think you have memory; but it has you!’
Our understanding of memory is also enriched by the discovery that culture shapes how we remember our past. If your culture values social interactions over individualism, for example, your memories will be influenced accordingly. In the same way, the study of memory has been enriched by the discovery that brains generate collective memories to create a shared reality for the group, which has important consequences for how we remember our past and shape our future.
What followed memory in the brain’s evolution was an event that set us apart from all other primates. For reasons that we are only just discovering, the modern human brain underwent a staggering and rapid transformation in cognitive and behavioural abilities some 200,000 years ago. Described by scientists as the Human Revolution, it would prove to be the wellspring of our species’ intellect, language and creativity.