Memory, p.34
Memory, page 34
She has adapted the same method of ‘in-scanner’ testing for tasks involving episodic memory. Unlike recall of facts or object recognition, memories of personal experiences are coloured with emotions and played out in a rich spatial, temporal and social context. To understand how the brain stores and recalls this form of memory, it is important to evoke the ‘whole’ memory during MRI scanning. One way of doing this is to project a photo of a party or wedding from a family album on to the screen, prompting the subject to recall and re-experience this particular event in their past.
As well as understanding how these memories are structured in ‘normal’ healthy subjects, Dr Maguire also hopes to find out more about what happens in the brains of people who lose these abilities.
‘The ability to find our way around an environment and to remember the events that occur within it – both thought to be mediated by the hippocampus – are fundamental to normal functioning in daily life,’ she says. ‘Unfortunately, the hippocampus is vulnerable to brain damage by epilepsy, dementia, and anoxia (when the brain is deprived of oxygen), which impacts on both these capacities, leaving patients severely debilitated and dependent on others for day-to-day living.’
The symptoms can be very problematic. ‘If spatial cognition is affected, people literally don’t know where they are, and if episodic memory is damaged, it can lead to amnesia,’ says Dr Maguire. People with amnesia live permanently in the present. Their speech and general intellect tends to remain intact, because remembering facts and general knowledge is not dependent on the hippocampus, but everything is frozen in time: they cannot remember anything that occurs after the damage took place. ‘If they do a couple of hours of tests with me, for example, and I leave the room for ten minutes and come back, they can’t remember anything about me or what they had been doing. They can’t live alone because they can’t remember if they turned the gas off or paid their bills. Sometimes, which is very sad, if a spouse dies, they can’t remember their loved one is now gone.’
For people suffering from hippocampal damage and associated difficulties with spatial and episodic memory, the question of whether the brain can mend itself and memory be recovered is a pressing one. ‘It has long been thought that the adult brain only has a limited amount of plasticity,’ says Dr Maguire. ‘But findings like those from our study of London cab drivers show that structural changes can occur in healthy human brains. Perhaps in the future we could use that kind of understanding to help people with hippocampal damage.’
Dr Maguire is therefore attempting to characterise the brain’s plasticity in more detail by taking a unique lifespan approach to her study of memory. ‘We’ll be measuring the effects of disease or injury on the brain and memory at all stages of life – from developmental disorders in children to dementia in the elderly – and comparing these with healthy subjects.’fn4 As well as examining memories at single points in time, she will also be tracking how people’s brains change over short and longer timescales, again comparing diseased or injured brains with healthy brains of the same age. ‘If you see a measurable difference in the brain of a particular individual over time, such as growth of their hippocampus, then you know that structural changes have definitely occurred in that particular person’s brain. This is different from simply comparing groups of people at one point in time.’
Alongside these projects, Dr Maguire will be conducting further work on black-cab drivers. This time she will be testing retired cabbies to see if the plasticity goes both ways – whether the hippocampus shrinks again when they stop full-time navigation around London. She also wants to establish whether this plasticity is limited to navigation, or whether it is generalisable to other areas of the brain.
Findings from her research will be used to provide benchmarks for assessing the effects of disease or injury on memory, for people of different ages, and could aid development of clinical memory tests for early diagnosis of pathology. ‘In the long term, we hope we’ll be able to use this information to develop new kinds of rehabilitation programmes for people with hippocampal damage – but we still have a very long way to go.’
Dr Maguire’s study on London black-cab drivers showed distinct structural changes to the brain linked to the memorising of large interconnected spatial environments. She was interested to find out if similar changes accompanied another feat of exceptional memory, those on show at the World Memory Championships, which take place every year in London.
‘People entering the World Memory Championships can do amazing things,’ she says. ‘They can memorise the order of cards in deck after deck of cards, for example. One memory champion passed time waiting in reception prior to his scan by memorising pages from the phone book – pretty well too; I tested him on it.’
Despite their high performance on memory tasks, however, Dr Maguire could find no structural changes. ‘I then asked them what strategies they used. Nine out of ten of them used the same strategy: an ancient Greek method, called the method of loci. It’s based on navigation: they imagine going down a street they know well, place items at certain positions along the street, then mentally retrace their route to find the items.’
Although this strategy uses spatial memory to boost performance, the amount of large-scale space memorised is small, possibly accounting for the lack of structural changes in the right hippocampus. ‘Their brain doesn’t have to change to accommodate a large map of London in their heads as it does for the cab drivers; the memory champions just need to memorise a couple of routes in detail.’
From the Economist, ‘Sleeping on it’ (2000)
When an Arctic ground squirrel hibernates, its body temperature drops below the freezing point of water and the blood-flow through its brain slows to a trickle. Though the squirrel’s brain survives, it loses many of the nerve-cell connections that govern how it operates. The brain regenerates itself soon after the animal emerges from its long sleep. Exactly how is a matter of intense research but one group of scientists thinks that part of the explanation lies with a protein associated with Alzheimer’s disease. And that raises the hope that the ravages wrought on the human mind by Alzheimer’s disease could be as reversible as the winter freeze.
Arctic ground squirrels hibernate for up to seven months of the year, sinking into a torpor from which they periodically rewarm their bodies to 37ºC before re-entering the supercooled state. Research has shown that during hibernation these animals lose memories they laid down beforehand and also their ability to form new ones. This loss must be temporary, however, or the animal would become more amnesic with each hibernation.
The brain stores information in neuronal networks. The chemical connections between neurons, called synapses, are thought to be critical to the formation of those networks and hence the laying down of memories. In 2003 a group led by Thomas Arendt of the University of Leipzig in Germany showed that the number of synapses in the hippocampus, a brain structure crucial for learning and memory, falls during hibernation. This is partly because hippocampal neurons lose many of their branching projections, or dendrites, and so provide less opportunity for forming synapses with neighbouring neurons.
All that changes within two or three hours of an animal emerging from hibernation, when a wave of new growth ensures that the number of synapses in the hippocampus soars beyond even prehibernation levels. The next 20 hours see a pruning back of those connections, rather as in the very young human brain. Just as in that developmental process, the new synapses seem to enhance memory only once the pruning has taken place.
Nobody knows what triggers these dramatic morphological changes in the hippocampus during and after hibernation. But Dr Arendt’s group has made the startling discovery that hibernating brains accumulate a protein called hyperphosphorylated tau. This protein is known also to accumulate in the neurons that degenerate in the brains of people with Alzheimer’s disease. Notably, though by no means exclusively, it accumulates in the hippocampal neurons, where it is associated with the formation of lesions.
There are several competing theories about what causes Alzheimer’s disease. One possibility is that the tau protein causes the lesions in the brain. Another is that something else causes the lesions, and the tau protein is the brain’s defence against that attack. Thus, it is possible that the tau protein might not be the problem, but rather a symptom of the problem.
During hibernation, the levels of tau protein in a squirrel’s hippocampal cells are directly correlated with the loss of synapses – but not with the appearance of lesions. On emerging from hibernation, the squirrel eliminates the tau protein from its brain. This has led Dr Arendt to suggest that rather than being a part of a disease process, the formation of the tau protein could be a mechanism by which the brain protects itself. He argues that the brain is armed with mechanisms for clearing the tau protein and that the reason it doesn’t in people with Alzheimer’s disease is because the protein is protecting the neurons.
His stance is contentious. ‘As the field stands, viewing pathology as anything other than pathogenic is controversial. Saying it is protective is heretical,’ says Mark Smith of Case Western Reserve University in Cleveland, Ohio. He has conducted studies on living neurons which suggest that the tau protein is produced in response to oxidative stress, thus lending support to the protective hypothesis.
Dr Arendt’s group is now engaged in discovering exactly how the tau protein can be cleared from the brain. Help for Alzheimer’s patients remains uncertain and a very long way off, but spring seems to have come a bit closer.
The Economist (4 February 2000)
ANTONIO DAMASIO, ‘The Hidden Gifts of Memory’ (updated in 2006 from The Feeling of What Happens, 2000)
One of the gifts memory bestows on the human mind is related to an apparent paradox identified by William James in his discussions of consciousness. Here is the paradox: the self in our stream of consciousness appears to change continuously as it moves forward in time; at the same time, we retain a sense that the self remains the same throughout our existence. The solution of the paradox comes from the fact that the seemingly changing self and the seemingly permanent self, although closely related, are not one entity with two irreconcilable aspects but rather two entities. The ever-changing self identified by James is the core self, the sense that is born out of the process of core consciousness. It is not so much that the core self mutates but rather that its existence is transient, ephemeral, that it needs to be remade and reborn continuously. The self that appears to remain the same is the autobiographical self, the sense that is born out of the process of extended consciousness. The autobiographical self appears intransient because it is based on a repository of memories for stable facts in an individual biography. Those facts can be reliably reactivated and thus provide continuity and seeming permanence in our lives.
This arrangement requires the availability of memory. Core consciousness provides us with a core self, but we also need conventional memory to construct an autobiographical self, and we need both core consciousness and working memory to make the autobiographical self explicit, that is, to display the contents of the autobiographical self in extended consciousness. Creatures with limited memory do not face James’s paradox. They inhabit a world one step up from innocence. They probably have the seemingly continuous experience of moments of conscious individuality, but they are neither burdened nor enriched by the memories of a personal past, let alone by memories of an anticipated future.
In my proposal, core consciousness is a central resource produced by a circumscribed mental and neural system. The fact that core consciousness is central does not mean that it depends on one structure. Indeed, a large number of neural structures is necessary for core consciousness to occur. But the complexity of the system, the multiplicity of its components, and the required concertedness of its operation should not make us overlook the following fact: when we consider the anatomical scale of the whole brain, the basic system underlying core consciousness is relatively confined to one set of anatomical sites rather than being widespread throughout the brain. In brief, there are plenty of brain sites not concerned with the making of core consciousness.
The robustness of core consciousness comes from this anatomical and functional centrality, and from the fact that any content of mind, whether actively processed in a live interaction or recalled from memory, can coax the core consciousness system into action, provoke it, so to speak, and in so doing generate a pulse of transient core consciousness. Core consciousness is not organised by sensory modality. In other words, there is no ‘visual’ core consciousness or ‘auditory’ core consciousness. Rather, core consciousness can be used by any sensory system and by the motor system to generate knowledge about any object or movement.
The contents of the autobiographical self are the organised, reactivated memories of fundamental facts from an individual’s biography. They are the prime beneficiaries of core consciousness. Whenever an object X provokes a pulse of core consciousness and the core self emerges relative to object X, selected sets of facts from the implicit autobiographical self are also consistently activated as explicit memories and provoke pulses of core consciousness of their own.
At any given moment of our sentient lives, then, we generate pulses of core consciousness for one or a few target objects. But that is not all. We also generate pulses of core consciousness for a set of accompanying, reactivated autobiographical memories. Without such autobiographical memories we would have no sense of past or future, there would be no historical continuity to our persons. However, without the narrative of core consciousness, and without the transient core self that is born within it, we would have no knowledge whatsoever of the moment, or of the memo-rised past, or of the anticipated future plans that we also have committed to memory. Core consciousness is a foundational must. It takes precedence, evolutionarily and individually, over extended consciousness. And yet, without extended consciousness, core consciousness would not have the resonance of past and future. The interlocking of core and extended consciousnesses, of core and autobiographical selves, is complete.
To understand the neuroanatomical foundation of the autobiographical self that I envision we need to outline, however briefly, the theoretical framework with which mental images and brain structures can be brought together. The framework posits an image space, the space in which images of all sensory types explicitly occur in mind, and a dispositional space, a space filled with records of implicit knowledge on the basis of which images can be constructed in recall, movements can be generated, and the processing of images can be facilitated. Dispositions can hold the memory of an image perceived on some previous occasion and can help reconstruct a similar image from that memory; dispositions can also assist the processing of a currently perceived image – for instance, in terms of the degree of attention accorded to the image and the degree of its subsequent enhancement.
There is a brain counterpart for the image space and a brain counterpart for the dispositional space. Brain structures such as the so-called early cortices of the varied sensory modalities (e.g. vision, audition, touch) support neural patterns that are likely to be the basis for mental images. On the other hand, higher-order cortices and varied neural nuclei beneath the cortical level hold dispositions with which both images and actions can be generated, rather than holding or displaying the explicit patterns manifest in images or actions themselves.
I have proposed that dispositions are held in neuron ensembles which I call convergence zones. To the partition of cognition between an image space and a dispositional space, then, corresponds a partition of the brain into (1) neural-pattern maps, activated in the early sensory cortices, in the so-called limbic cortices, and in some subcortical nuclei, and (2) convergence zones, located in the higher-order cortices and in some subcortical nuclei.
The brain forms memories in a highly distributed manner. Take, for instance, the memory of a hammer. There is no single place in our brain where you can hold the record for hammer, no single place where you would find an entry with the word hammer followed by a neat dictionary definition of what a hammer is. Instead, as current evidence suggests, there are several records in our brain that correspond to different aspects of our past interaction with hammers. Those aspects include the shape of hammers, the typical movement with which we use them, the hand shape and the hand motion required to manipulate a hammer, the result of the action, and the word that designates hammer in whatever many languages we know. These records are dormant, dispositional, and implicit. Dormant, dispositional and implicit are key traits in the nature of memory storage. Memory records exist in dormant and implicit form until they are activated and become explicit; and memory records are dispositional in the sense that to become explicit a certain number of procedures or dispositions must be executed. Dispositional records of memory unfold, as it were. And all of the records related to the hammer of the example are based on separate neural sites located in separate brain regions. The separation is imposed by the design of the brain and by the physical nature of our environment. Appreciating the shape of a hammer visually is different from appreciating its shape by touch; the mental pattern corresponding to our manipulation of the hammer cannot be stored in the same brain region that ‘stores’ the pattern of its movement as we see it; the former is stored in somatosensory and motor regions, while the latter is stored in visual regions; the phonemes with which we make the word hammer cannot be stored in the same place, either, and are housed in auditory and somatomotor regions. The spatial separation of the records poses no problem, as it turns out, because when all the records are made explicit in image form they are exhibited in only a few brain sites and are coordinated in time in such a fashion that all the recorded components appear seamlessly integrated.
