Memory

Humans have a long tradition of holding genes responsible for individual differences in behavior (of course, we called it "blood", then, or "family"). In the 20th century, a counter-belief arose: that it was all down to environment, to upbringing. In more recent decades, we have become increasingly aware of how tightly and complexly genes and environment are entwined.

It's not enough to say merely that environment tempers genes, or that genes affect how the environment works on an individual — genes and environment work on each other in an ongoing interaction, that continues throughout our lifetimes. This ongoing change even affects attributes most people deeply believe are, if not hard-wired in the womb, at least set in childhood: attributes such as intelligence, 'natural' talent, and gender differences.

Planning memory contains your plans and goals (such as, “I must pick up the dry-cleaning today”; “I intend to finish this project within three months”). Forgetting an appointment or a promise is one of the memory problems people get most upset about.

Why remembering to do things is so difficult

Remembering intentions is in fact much more difficult than remembering events that have happened, and the primary reason is the lack of retrieval cues. This is why, of all memory tasks, remembering to do things relies most heavily on external memory aids. Reminder notes, calendars, diaries, watch-alarms, oven-timers, leaving objects in conspicuous places — all these external aids are acting as cues to memory.

In partial compensation for the paucity of effective retrieval cues , planning memories are more easily triggered by quite marginal cues. Thus a friend of mine was reminded that her son’s friend would be spending Saturday night with them when she saw an advertisement for a movie about John F. Kennedy (the child’s father had the same initials: JFK).

Setting up effective cues

When we form an intention, we usually link it either to an event (“after we go to the swimming-pool, we’ll go to the supermarket”) or a time (“at 2pm I must ring Fred”). But these trigger events or times frequently fail to remind us of our intention.

This is often because the trigger is not in itself particularly distinctive. Your failure to remember to ring Fred at 2pm, for example, may be because you paid little attention to the clock reaching that time, or because there were other competing activities triggered by that same time signal.

Moreover, not all planning is linked to a trigger event or time. Quite a lot of planning simply waits upon an appropriate opportunity (“must buy some stamps sometime”). Such intentions usually need quite explicit cues. Thus, if I happened to see stamps on sale, I would probably remember my intention, but walking past, or even into, a shop that happens to sell stamps, may not be enough to trigger my memory.

On the other hand, I might keep being reminded of my intention when I am in the same context as when I originally formed the intention (when I am not in a position to carry it out!) — hence your increasing exasperation that you can never remember a particular intention when you can do anything about it.

Link the trigger and the intention

To deal with opportunistic planning, you should try to specify features of an appropriate opportunity. Thus, to remember to buy bread on the way home, you should think about what actions you need to take to buy the bread (for example, going a different route) and try to form a strong link between the trigger event and your action (“today when I get to the traffic lights I’ll turn left”).

A reminder of your intention is much less effective than being reminded of both the trigger event and the intended activity. Even being reminded of the trigger event is better than being reminded of the intention on its own.

To remind yourself to do something, focus on the trigger not the intent itself.

Don’t assume that because something is important to you, you will automatically remember it — somewhat to their surprise, researchers have found no evidence that personal importance has any effect on the likelihood of remembering to do something.

 

"I'm terrible at remembering names"

"I'm great with names, but I'm hopeless at remembering what I've read."

"I always remember what people tell me about themselves, but I'm always forgetting birthdays and anniversaries."

There is no such thing as a poor memory!

There will be memory domains that you are less skilled at dealing with.

Information comes in different packages

Think about the different types of information you have stored in your memory:

• the name of your dentist
• your PIN number
• the taste of chocolate
• the sound of train whistle
• the scent of a rose
• the feeling of fear
• the knowledge of how to drive a car
• your intention to pick up bread on the way home

etc etc.

Is it likely that these so-different types of information are stored in the same sort of codes in your brain? That they are processed in the same way? That the same types of cues will trigger recall?

No.

The concept of different memory domains is useful

The idea that there are separate memory systems for different types of information has been around a long while. While not all researchers agree on how many different domains there are, the idea of memory as a multimodal system has become respectable among cognitive psychologists. At a practical level, it isn't necessary for us to concern ourselves with the precise details of academic concerns regarding the nature of memory domains and how many there are. The idea of memory domains is a supremely useful one, for anyone wanting to improve their memory.

Different types of information require different memory strategies, and the idea of memory domains helps us match different strategies to different memory tasks. Knowing the principles by which we encode different types of information, we can understand which strategies will be useful.

Knowledge memory vs personal memory

One fundamental distinction that can be made is between your knowledge of "facts", and knowledge that is more personal.

Knowledge memory contains information about the world.

Personal memory contains information about you.

Within knowledge memory, separate domains may exist for numbers, for music, for language, and for stories. These are all types of information which appear to be dealt with in different ways.

Personal memory also contains a number of different domains:

Diagrams and table taken from The Memory Key

Autobiographical memory contains the information you have about yourself. It includes several domains:

• self-description (the source of a large part of your sense of identity), containing information such as:
  • whether or not you like ice-cream
  • what your favorite color is
  • what you think about a political party
• emotional memory, which not only contains our memories of emotional experiences, but also helps us control our moods. By dwelling on appropriate memories, we can sustain a mood. By recalling memories that involve a contrasting emotion, we can change a mood.
• event memory

Your memory for events

This is the largest component of autobiographical memory, containing three separate but related domains:

• memory for specific events that have happened to you
• memory for general events, which tells you the broad sequence of actions in events such as going to a restaurant or going to the dentist
• a potted summary of your life, which enables you to answer such questions as, “Where did you go to school?”, “Where were you working last year?”.

These may be thought of as being connected hierarchically:

Recalling specific events

Event memory is usually entered via the general-event level, although the information we are searching for is usually at the specific-event level. Thus, if you're trying to retrieve the memory of going to see the movie Titanic, you will probably start by accessing the general event "going to the pictures"

Specific events over time become merged into a general event - all the occasions you've been to the dentist, for example, have blurred into a generic "script", which encapsulates the key experiences and actions that are typical of the going-to-the-dentist event. After the specific event has become consolidated into the script, only distinctive events are likely to be specifically remembered. That is, events when something unusual/interesting/humorous happened.

The power of these scripts is such that people often "remember" details of a specific event that never happened, merely because they are typical of the script for that event.

Our memory for events reflects what we expect to happen.

It is perhaps because of this that unexpected events and new events (first-time experiences) are better remembered. If you don't have an existing script for the event, or if the event is atypical enough not to easily fit an existing script, then you can't mold the experience to your expectations.

The more distinctive an event - the more the event breaks with your script for that type of event - the better your memory for that particular event will be. (Failures to remember trivial events, such as where you’ve put something, or whether you’ve done something, are reflections of the fact that we pay little attention to routine actions that are, as it were, already scripted).

To remember an event therefore, you should look for distinctive details.

What makes a good cue for remembering events?

One of the most interesting areas of research in the study of event memory is a small set of diary studies. In one such study, a Dutch psychologist called Willem Wagenaar recorded his day's events every day for six years, noting down:

• who was involved
• what the event was
• where it occurred
• when it occurred

Wagenaar was hoping to discover which of these different bits of information were the best retrieval cues. At the conclusion of his study he reported that what was the best cue, followed by who and where. When was the least effective (have you ever tried to remember an event on the basis of its approximate date?).

There is nothing particularly special about these types of information however. Later, Wagenaar reanalyzed his data, and found that most of the difference in the memorability of these cues was due to their relative distinctiveness. Thus, the nature of the event is usually the most distinctive aspect of the event, and the people involved, and the location, are usually more distinctive bits of information than the date or time of occurrence.

To remember a specific event, we need a key - a unique feature that allows us to readily distinguish that event from similar events.

You may have heard of “g”. It’s the closest we’ve come to that elusive attribute known as “intelligence”, but it is in fact a psychometric construct, that is, we surmise its presence from the way in which scores on various cognitive tests positively correlate.

In other words, we don’t really know what it is (hence the fact it is called “g”, rather than something more intelligible), and in fact, it is wrong to think of it as a thing. What it is, is a manifestation of some property or properties of the brain — and we don’t know what these are.

Various properties have been suggested, of course. Speed of processing; synaptic plasticity; fluid cognition. These are all plausible, but experimental studies have failed to provide clear evidence for any of them. The closest has been fluid cognition, or fluid intelligence, which is paired with crystallized intelligence. These two terms point to a useful distinction.

Fluid intelligence refers to cognitive functions associated with general reasoning and problem-solving, and is often described as executive function, or working memory capacity.

Crystallized intelligence, on the other hand, refers to cognitive functions associated with previously acquired knowledge in long-term store.

There is of course some interplay between these functions, but for the most part they are experimentally separable.

There are a couple of points worth noting.

For a start, different IQ tests measure fluid intelligence and crystallized intelligence to varying extents – the Raven’s Progressive Matrices Test, for example, predominantly measures fluid intelligence, while the WAIS disproportionately measures crystallized intelligence. An analysis of the most widely used intelligence test batteries for children found that about 1/3 of the subtests measure crystallized intelligence, an additional ¼ measure knowledge and reading/writing skills, while only 7% directly measure fluid intelligence, with perhaps another 10% measuring skills that have a fluid intelligence component – and nearly all the fluid subtests were found in one particular test battery, the W-J-R.

The so-called Flynn effect – the rapid rise in IQ over the past century – is for the most part an increase in fluid intelligence, not crystallized intelligence. While it has been hypothesized that fluid intelligence paves the way for the development of crystallized intelligence, it should be noted that the distinction between fluid and crystallized intelligence is present from a very early age, and the two functions have quite different growth patterns over the life of an individual.

So, what we’re saying is that most IQ tests provide little measure of fluid intelligence, although fluid intelligence appears to reflect “g” more closely than any other attribute, and that although crystallized intelligence is assumed to reflect environment (e.g., education) far more than fluid intelligence, it is fluid intelligence that has been rising, not crystallized intelligence.

In fact, for this and other reasons, it seems that fluid intelligence is far more affected by environment than has been considered.

I’ll leave you to ponder on the implications of this. Let me make just one more point.

The brain areas known to be important for fluid cognition are part of an interconnected system associated with emotion and stress response, and it is hypothesized that functions heretofore considered distinct from emotional arousal, such as reasoning and planning, are in fact very much part of a system in which emotional response is involved.

We’re not saying here that emotions can disrupt your reasoning processes, we all know that. What is being suggested is more radical – that emotions are part and parcel of the reasoning process. Okay, I always knew this, but it’s nice to see science coming along and providing some evidence.

The point about the close interaction between emotional reactivity and fluid intelligence is that stress may have a significant effect on fluid intelligence.

And I’ll leave you to ponder the implications of that.

Widely cited gender differences in cognition

It is clear that there are differences between the genders in terms of cognitive function; it is much less clear that there are differences in terms of cognitive abilities. Let me explain what I mean by that.

It's commonly understood that males have superior spatial ability, while females have superior verbal ability. Males are better at math; females at reading. There is some truth in these generalizations, but it's certainly not as simple as it is portrayed.

First of all, as regards spatial cognition, while males typically outperform females on tasks dealing with mental rotation and spatial navigation, females tend to outperform males on tasks dealing with object location, relational object location memory, and spatial working memory.

While the two sexes score the same on broad measures of mathematical ability, girls tend to do better at arithmetic, while boys do better at spatial tests that involve mental rotation.

Having said that, it does depend where you're looking. The Programme for International Student Assessment (PISA) is an internationally standardised assessment that is given to 15-year-olds in schools. In 2003, 41 countries participated. Given the constancy of the gender difference in math performance observed in the U.S., it is interesting to note what happens in other countries. There was no significant difference between the sexes in Australia, Austria, Belgium, Japan, the Netherlands, Norway, Poland, Hong Kong, Indonesia, Latvia, Serbia, and Thailand. There was a clear male superiority for all 4 content areas in Canada, Denmark, Greece, Ireland, Korea, Luxembourg, New Zealand, Portugal, the Slovak Republic, Liechtenstein, Macao and Tunisia. In Austria, Belgium, the United States and Latvia, males outperformed females only on the space and shape scale; in Japan, the Netherlands and Norway only on the uncertainty scale. And in Iceland, females always consistently do better than males!

Noone knows why, but it is surely obvious that these differences must lie in cultural and educational factors.

Interestingly, the IEA Third International Mathematics and Science Study (TIMSS) shows this developing -- while significant gender differences in mathematics were found only in 3 of the 16 participating OECD countries for fourth-grade students, gender differences were found in 6 countries at the grade-eight level, and in 14 countries at the last year of upper secondary schooling.

This inconsistency is not, however, mirrored in verbal skills -- girls outperform boys in reading in all countries.

Gender differences in language have been consistently found, and hardly need reiteration. However, here's an interesting study: it found gender differences in the emerging connectivity of neural networks associated with skills needed for beginning reading in preschoolers. It seems that boys favor vocabulary sub-skills needed for comprehension while girls favor fluency and phonic sub-skills needed for the mechanics of reading.The study points to the different advantages each gender brings to learning to read.

There's a lesson there.

There are other less well-known differences between the sexes. Women tend to do better at recognizing faces. But a study has found that this superiority applies only to female faces. There was no difference between men and women in the recognition of male faces.

Moreover, pre-pubertal boys and girls have been found to be equally good at recognizing faces and identifying expressions. However, they do seem to do it in different ways. Boys showed significantly greater activity in the right hemisphere, while the girls' brains were more active in the left hemisphere. It is speculated that boys tend to process faces at a global level (right hemisphere), while girls process faces at a more local level (left hemisphere).

It's also long been recognized that women are better at remembering emotional memories. Interestingly, an imaging study has revealed that the sexes tend to encode emotional experiences in different parts of the brain. In women, it seems that evaluation of emotional experience and encoding of the memory is much more tightly integrated.

But of course, noone denies that there are differences between men and women. The big question (one of the big questions) is how much, if any, is innate.

Studies of differences, even at the neural level, don't demonstrate that. It's increasingly clear that environmental factors affect all manner of thing at the neural level. However, one study of 1-day-old infants did find that boys tended to gaze at three-dimensional mobiles longer than girls did, while girls looked at human faces longer than boys did.

Of course, even a 1-day-old infant isn't entirely free of environmental influence. In this case, the most important environmental influence is probably hormones.

Hormones and chemistry

A lot of studies in recent years have demonstrated that estrogen is an important player in women's cognition. Spatial ability in particular seems vulnerable to hormonal effects. Women do vary in their spatial abilities according to where they are in the menstrual cycle, and there is some evidence that spatial abilities (in both males and females) may be affected by how much testosterone is received in the womb.

Another study has found children exposed to higher levels of testosterone in the womb also develop language later and have smaller vocabularies at 2 years of age.

Hormones aren't the only chemical affecting male and female brains differently. Significant differences have been found in the brain activity of men and women when engaged in a broad range of activities and behaviors. These differences are more acute during impulsive or hostile acts. But — here's the truly fascinating thing — nicotine causes these brain activity differences to disappear. A study has found that among both smokers and non-smokers on nicotine, during aggressive moments, there are virtually no differences in brain activity between the sexes. A finding that supports other studies that indicate men's and women's brains respond differently to the same stimuli — for example, alcohol.

What does all this mean? Well, let's look at the question that's behind the whole issue: are men smarter than women? (or alternately, are women smarter than men?)

Is one sex smarter than the other?

Here's a few interesting studies that demonstrate some more differences between male and female brains.

A study of some 600 Dutch men and women aged 85 years found that the women tended to have better cognitive speed and a better memory than the men, despite the fact that significantly more of the women had limited formal education compared to the men. This may be due to better health. On the other hand, there do appear to be differences in the way male and female brains develop, and the way they decline.

For example, women have up to 15% more brain cell density in the frontal lobe, which controls so-called higher mental processes, such as judgement, personality, planning and working memory. However, as they get older, women appear to shed cells more rapidly from this area than men. By old age, the density is similar for both sexes.

A study of male and female students (aged 18-25) has found that men's brain cells can transmit nerve impulses 4% faster than women's, probably due to the faster increase of white matter in the male brain during adolescence.

An imaging study of 48 men and women between 18 and 84 years old found that, compared with women, men had more than six times the amount of intelligence-related gray matter. On the other hand, women had about nine times more white matter involved in intelligence than men did. Women also had a large proportion of their IQ-related brain matter (86% of white and 84% of gray) concentrated in the frontal lobes, while men had 90% of their IQ-related gray matter distributed equally between the frontal lobes and the parietal lobes, and 82% of their IQ-related white matter in the temporal lobes. Despite these differences, men and women performed equally on the IQ tests.

It has, of course, long been suggested that women are intellectually inferior because their brains are smaller. A study involving the intelligence testing of 100 neurologically normal, terminally ill volunteers found that a bigger brain size is indeed correlated with higher intelligence — but only in certain areas, and with odd differences between women and men. Verbal intelligence was clearly correlated with brain size for women and — get this — right-handed men! But not for left-handed men. Spatial intelligence was also correlated with brain size in women, but much less strongly, while it was not related at all to brain size in men.

Also, brain size decreased with age in men over the age span of 25 to 80 years, suggesting that the well-documented decline in visuospatial intelligence with age is related, at least in right-handed men, to the decrease in cerebral volume with age. However age hardly affected brain size in women.

What is all this telling us?

Male and female brains are different: they develop differently; they do things differently; they respond to different stimuli in different ways.

None of this speaks to how well information is processed.

None of these differences mean that individual brains, of either sex, can't be trained to perform well in specific areas.

Here’s an experiment and a case study which bear on this.

It's all about training

The experiment concerns rhesus monkeys. The superiority of males in spatial memory that we're familiar with among humans also occurs in this population. But here's the interesting thing — the gender gap only occurred between young adult males and young untrained females. In other words, there was no difference between older adults (because performance deteriorated with age more sharply for males), and did not occur between male and female younger adults if they were given simple training. Apparently the training had little effect on the males, but the females improved dramatically.

The “case study” concerns Susan Polgar, a chess master. The Polgar sisters are a well-known example of “hot-housing”. I cited them in my article on the question of whether there is in fact such a thing as innate talent. Susan Polgar and her sisters are examples of how you can train “talent”; indeed, whether there is in fact such a thing as “talent” is a debatable question. Certainly you can argue for a predisposition towards certain activities, but after that … Well, even geniuses have to work at it, and while you may not be able to make a genius, you can certainly create experts.

This article was provoked, by the way, by comments by the President of Harvard University, Lawrence Summers, who recently stirred the pot by giving a speech arguing that boys outperform girls on high school science and math scores because of genetic differences between the genders, and that discrimination is no longer a career barrier for female academics. Apparently, during Dr Summers' presidency, the number of tenured jobs offered to women has fallen from 36% to 13%. Last year, only four of 32 tenured job openings were offered to women.

You can read a little more about what Dr Summers said at http://education.guardian.co.uk/gendergap/story/0,7348,1393079,00.html, and there's a rather good response by Simon Baron-Cohen (professor in the departments of psychology and psychiatry, Cambridge University, and author of The Essential Difference) at: http://education.guardian.co.uk/higher/research/story/0,9865,1399109,00.html

What is neurogenesis?

Neurogenesis — the creation of new brain cells — occurs of course at a great rate in the very young. For a long time, it was not thought to occur in adult brains — once you were grown, it was thought, all you could do was watch your brain cells die!

Adult neurogenesis (the creation of new brain cells in adult brains) was first discovered in 1965, but only recently has it been accepted as a general phenomenon that occurs in many species, including humans (1998).

Where does adult neurogenesis occur?

It's now widely accepted that adult neurogenesis occurs in the subgranular zone of the dentate gyrus within the hippocampus and the subventricular zone (SVZ) lining the walls of the lateral ventricles within the forebrain. It occurs, indeed, at a quite frantic rate — some 9000 new cells are born in the dentate gyrus every day in young adult rat brains — but under normal circumstances, at least half of those new cells will die within one or two months.

The neurons produced in the SVZ are sent to the olfactory bulb, while those produced in the dentate gyrus are intended for the hippocampus.

Adult neurogenesis might occur in other regions, but this is not yet well-established. However, recent research has found that small, non-pyramidal, inhibitory interneurons are being created in the cortex and striatum. These new interneurons appear to arise from a previously unknown class of local precursor cells. These interneurons make and secrete GABA (see below for why GABA is important), and are thought to play a role in regulating larger types of neurons that make long-distance connections between brain regions.

How does neurogenesis occur?

New neurons are spawned from the division of neural precursor cells — cells that have the potential to become neurons or support cells. How do they decide whether to remain a stem cell, turn into a neuron, or a support cell (an astrocyte or oligodendrocyte)?

Observation that neuroblasts traveled to the olfactory bulb from the SVZ through tubes formed by astrocytes has led to an interest in the role of those support cells. It's now been found that astrocytes encourage both precursor cell proliferation and their maturation into neurons — precursor cells grown on glia divide about twice as fast as they do when grown on fibroblasts, and are about six times more likely to become neurons.

Adult astrocytes are only about half as effective as embryonic astrocytes in promoting neurogenesis.

It’s been suggested that the role of astrocytes may help explain why neurogenesis only occurs in certain parts of the brain — it may be that there’s something missing from the glial cells in those regions.

The latest research suggests that the astrocytes influence the decision through a protein that it secretes called Wnt3. When Wnt3 proteins were blocked in the brains of adult mice, neurogenesis decreased dramatically; when additional Wnt3 was introduced, neurogenesis increased.

How are these new neurons then integrated into existing networks? Mouse experiments have found that the brain chemical called GABA is critical. Normally, GABA inhibits neuronal signals, but it turns out that with new neurons, GABA has a different effect: it excites them, and prepares them for integration into the adult brain. Thus a constant flood of GABA is needed initially; the flood then shifts to a more targeted pulse that gives the new neuron specific connections that communicate using GABA; finally, the neuron receives connections that communicate via another chemical, glutamate. The neuron is now ready to function as an adult neuron, and will respond to glutamate and GABA as it should.

The creation and development of new neurons in the adult brain is very much a "hot" topic right now — it's still very much a work-in-progress. However, it is clear that other brain chemicals are also involved. An important one is BDNF (brain-derived neurotrophic factor), which seems to be needed during the proliferation of hippocampal precursor cells to trigger their transformation into neurons.

Other growth factors have been found to stimulate proliferation of hippocampal progenitor cells: FGF-2 (fibroblast growth factor-2) and EGF (epidermal growth factor).

Recently it has been discovered that the normal form of the prion protein which, when malformed, causes mad cow disease, is also involved in neurogenesis. These proteins, in their normal form, are found throughout our bodies, and particularly in our brains. Now it seems that the more of these prion proteins that are available, the faster neural precursor cells turn into neurons.

The immune system's T cells (which recognize brain proteins) are also critically involved in enabling neurogenesis to occur. Among mice given environmental enrichment, only those with healthy T-cells had their production of new neurons boosted.

Factors that influence neurogenesis

A number of factors have been found to affect the creation and survival of new neurons. For a start, damage to the brain (from a variety of causes) can provoke neurogenesis.

Moderate alcohol consumption over a relatively long period of time can also enhance the formation of new nerve cells in the adult brain (this may be related to alcohol's enhancement of GABA's function). Excess alcohol, however, has a detrimental effect on the formation of new neurons in the adult hippocampus. But although neurogenesis is inhibited during alcohol dependency, it does recover. A pronounced increase in new neuron formation in the hippocampus was found within four-to-five weeks of abstinence. This included a twofold burst in brain cell proliferation at day seven of abstinence.

Most drugs of abuse such as nicotine, heroine, and cocaine suppress neurogenesis, but a new study suggests that cannabinoids also promote neurogenesis. The study involved a synthetic cannabinoid, which increased the proliferation of progenitor cells in the hippocampal dentate gyrus of mice, in a similar manner as some antidepressants have been shown to do. The cannabinoid also produced similar antidepressant effects. Further research is needed to confirm this early finding.

If antidepressants promote neurogenesis, it won't be surprising to find that chronic stress, anxiety and depression are associated with losing hippocampal neurons. A rat study has also found that stress in early life can permanently impair neurogenesis in the hippocampus.

Showing the other side of this picture, perhaps, an intriguing rat study found that status affected neurogenesis in the hippocampus, with high-status animals having around 30% more neurons in their hippocampus after being placed in a naturalistic setting with other rats.

Also, a study into the brains of songbirds found that birds living in large groups have more new neurons and probably a better memory than those living alone.

Both physical activity and environmental enrichment (“mental stimulation”) have been shown to affect both how many cells are born in the dentate gyrus of rats and how many survive. Learning that uses the hippocampus has also been shown to have a positive effect, although results here have been inconsistent.

Inconsistent results from studies looking at neurogenesis are, it is suggested, largely because of a confusion between proliferation and survival. Neurogenesis is measured in terms of these two factors, which researchers often fail to distinguish between: the generation of new brain cells, and their survival. But these are separate factors, that are independently affected by various factors.

The inconsistency found in the effects of learning may also be partly explained by the complex nature of the effects. For example, during the later phase of learning, when performance is starting to plateau, neurons created during the late phase were more likely to survive, but neurons created during the early phase of more rapid learning disappeared. It’s speculated that that this may be a “pruning” process by which cells that haven’t made synaptic connections are removed from the network.

And finally, rodent studies suggest a calorie-restricted diet may also be of benefit.

It's not all about growing new neurons

A few years ago, we were surprised by news that new neurons could be created in the adult brain. However, it’s remained a tenet that adult neurons don’t grow — this because researchers have found no sign that any structural remodelling takes place in an adult brain. Now a mouse study using new techniques has revealed that dramatic restructuring occurs in the less-known, less-accessible inhibitory interneurons. Dendrites (the branched projections of a nerve cell that conducts electrical stimulation to the cell body) show sometimes dramatic growth, and this growth is tied to use, supporting the idea that the more we use our minds, the better they will be.

Short-term vs long-term memory

Working memory is a relatively recent term, a refinement of an older concept - that of short-term memory. Short-term memory was called thus to distinguish it from "long-term memory" - your memory store.

One important difference between the idea of short-term memory and working memory, is that short-term memory was conceived of as a thing. Different from long-term memory (variously analogized as a library, a filing system, a computer) chiefly in the duration of the records it held. But working memory, as its name suggests, is now conceived more as a process than a thing. A state of mind. A pattern of activation.

Working memory contains the information of which you are immediately aware.

To put information into our memory store, it must ... be worked on - i.e., be held in working memory. To get information out of the memory store - to “remember” something - it must again be in an active state - be in working memory. How can we know what we remember if we're not conscious of it?

However, you can only keep something "active" for a very short time without your conscious attention. It is this which so limits working memory capacity.

The magic number seven

Probably the most widely known fact about working memory is that it can only hold around seven chunks of information (between 5 and 9). However, this tells us little about the limits of working memory because the size of a chunk is indeterminate.

1 2 3 4 5 6 7 are seven different chunks - if you remember each digit separately (as you would, for example, if you were not familiar with the digits - as a young child isn't). But for those of us who are only too well-versed in our numbers, 1 through to 7 could be a single chunk.

Recent research suggests however, that it is not so much the number of chunks that is important. What may be important may be how long it takes you to say the words (information is usually held in working memory in the form of an acoustic - sound-based - code). It appears that you can only hold in working memory what you can say in 1.5 — 2 seconds. Slow speakers are therefore penalized.

Your working memory capacity

What we term "working memory" contains several functions, including the "central executive" which coordinates and manages the various tasks needed. The extent to which working memory is domain-specific (different "working memories", if you like, for different sensory and cognitive systems, such as language, spatial memory, number) is still very much debated. However, at a practical level, we may think of working memory as containing several different components, for which you have different "capacities". Thus, your capacity for numbers may well be quite different from your capacity for words, and both from your capacity for visual images.

Difficulty in remembering people’s names is one of the most common memory tasks that people wish to be better at. And the reason for this is not that their memory is poor, but because it is so embarrassing when their memory lets them down.

This isn’t just an issue at a personal level. It’s a particular issue for anyone who has to deal with a lot of people, many of whom they will see at infrequent intervals. Nothing makes a person — a client, a customer, a student — feel more valued than being remembered.

But we have, in fact, a remarkably good memory for other people’s faces. Think about the ease with which you distinguish between hundreds, even thousands, of human faces, and then think about how hard it is to distinguish between the faces of birds, or dogs, or monkeys. This is not because human faces are any more distinctive than the faces of other animals. Think about how much harder it is for you to distinguish between the faces of people of an unfamiliar racial type.

Contrary to what many European-descended people believe, Asian faces are no less distinctive than European faces, but the differences between any human face are sufficiently subtle that they take a great deal of experience to learn. The importance of learning these subtle differences is shown in the way new babies focus on faces, and prefer them to other objects.

Our memory for other people is of course more than a memory for faces, although that part probably has the most impressive capacity. We also remember people’s names and various biographical details. We can recognize people by hearing their voice, at a distance by seeing their shape or the way that they move, or even by their clothing.

But it’s faces that give certainty.

Many years ago, when I was in my second year at university, I left the student cafeteria and nearly bumped into a young woman in a white lab coat. I murmured some sort of apology and started to move on, and she said my name. I stared at her blankly. She said, ‘You don’t recognize me, do you?’ Even with this prompt, I didn’t immediately get it. I still remember staring at her unfamiliar face, and then … the features seemed to shift under my eyes. It was very weird. Suddenly I knew her. I was mortified, and stunned. I hadn’t seen her in a year, but we’d been best friends all through high school. How could I not immediately recognize her?

Identity information is complex

Identity information is encoded in memory in quite complex ways. To more effectively use those codes, to improve your memory for names, faces, and important personal details, it helps to understand how identity information is recorded in memory.

There are three ways we can “recognize” a person:

• we might recognize them as having been seen before, without recalling anything about them
• we might identify them as a particular person, without recalling their name (“that’s a friend of my son’s”)
• we might identify them by name

If you think about it you will realize that you never, ever, recall information about a person without recognizing them as familiar. While this sounds terribly obvious, there is actually a clinical condition (the Capgras delusion) whereby a person, while recognizing the people around them, believes they have been replaced by doubles (imposters, robots, aliens). This is simply because the normal accompanying feeling of familiarity is missing.

You also never remember a person’s name without knowing who she is. This is because names are held in a separate place to biographical details, and can only be accessed through those details.

Identity codes and how they are structured in memory

Why is there this hierarchy? Why can we only access names through biographical information? Because identity information is ordered. Your memory for a person is not like this:

But like this:

In other words, there are several different kinds of identity information, and they are clustered according to type, and can in fact only be accessed in a particular order.

Of the various identity codes (bits of encoded identity information), there are three kinds that are important for recognizing a person:

• structural codes (physical features)
• semantic codes (biographical details, e.g., occupation, marital status, address)
• name codes

There is a fourth type of code that is useful for remembering unfamiliar faces:

• visually-derived semantic codes (e.g., age, gender, attributions such as “he looks honest/intelligent/sly”)

Semantic codes that are visually derived have an advantage over biographical codes, because the link with the structural code is meaningful and thus strong, whereas the connection between the structural codes and biographical details is entirely arbitrary. To say someone looks like a fox connects meaningfully with the person’s facial features, whereas to say that someone is a lawyer has no particular connection with the person’s face (to say someone looks like a lawyer would of course be meaningfully connected).

Visually-derived semantic codes are useful for remembering new faces because the link with the physical features of the face is strong and meaningful.

However you cannot identify a person without reference to the biographical codes.

The interesting aspect of these different codes is that you can only access them in a particular order:

When you recognize a face as familiar but can’t recall anything about the person, the physical features have failed to trigger the biographical details. When you identify a person by recalling details about them, but can’t recall their name, the biographical information has failed to trigger the name.

Whether the name is recalled therefore depends on the strength of the connection between the biographical details and the name.

In other words, to improve your memory for a person’s identity, you must strengthen the link between the physical features and the biographical information. To improve your memory for the person’s name, you must strengthen the link between the biographical information and the name.

 

Note: A fascinating account of what it is like to be face-blind, from a person with the condition, can be found at: http://www.choisser.com/faceblind/

"Consolidation" is a term that is bandied about a lot in recent memory research. Here's my take on what it means.

Becoming a memory

Initially, information is thought to be encoded as patterns of neural activity — cells "talking" to each other. Later, the information is coded in more persistent molecular or structural formats (e.g., the formation of new synapses). It has been assumed that once this occurs, the memory is "fixed" — a permanent, unchanging, representation.

With new techniques, it has indeed become possible to observe these changes (you can see videos here). Researchers found that the changes to a cell that occurred in response to an initial stimulation lasted some three to five minutes and disappeared within five to 10 minutes. If the cell was stimulated four times over the course of an hour, however, the synapse would actually split and new synapses would form, producing a (presumably) permanent change.

Memory consolidation theory

The hypothesis that new memories consolidate slowly over time was proposed 100 years ago, and continues to guide memory research. In modern consolidation theory, it is assumed that new memories are initially 'labile' and sensitive to disruption before undergoing a series of processes (e.g., glutamate release, protein synthesis, neural growth and rearrangement) that render the memory representations progressively more stable. It is these processes that are generally referred to as “consolidation”.

Recently, however, the idea has been gaining support that stable representations can revert to a labile state on reactivation.

Memory as reconstruction

In a way, this is not surprising. We already have ample evidence that retrieval is a dynamic process during which new information merges with and modifies the existing representation — memory is now seen as reconstructive, rather than a simple replaying of stored information

Reconsolidation of memories

Researchers who have found evidence that supposedly stable representations have become labile again after reactivation, have called the process “reconsolidation”, and suggest that consolidation, rather than being a one-time event, occurs repeatedly every time the representation is activated.

This raises the question: does reconsolidation involve replacing the previously stable representation, or the establishment of a new representation, that coexists with the old?

Whether reconsolidation is the creating of a new representation, or the modifying of an old, is this something other than the reconstruction of memories as they are retrieved? In other words, is this recent research telling us something about consolidation (part of the encoding process), or something about reconstruction (part of the retrieval process)?

Hippocampus involved in memory consolidation

The principal player in memory consolidation research, in terms of brain regions, is the hippocampus. The hippocampus is involved in the recognition of place and the consolidation of contextual memories, and is part of a region called the medial temporal lobe (MTL), that also includes the perirhinal, parahippocampal,and entorhinal cortices. Lesions in the medial temporal lobe typically produce amnesia characterized by the disproportionate loss of recently acquired memories. This has been interpreted as evidence for a memory consolidation process.

Some research suggests that the hippocampus may participate only in consolidation processes lasting a few years. The entorhinal cortex, on the other hand, gives evidence of temporally graded changes extending up to 20 years, suggesting that it is this region that participates in memory consolidation over decades. The entorhinal cortex is damaged in the early stages of Alzheimer’s disease.

There is, however, some evidence that the hippocampus can be involved in older memories — perhaps when they are particularly vivid.

A recent idea that has been floated suggests that the entorhinal cortex, through which all information passes on its way to the hippocampus, handles “incremental learning” — learning that requires repeated experiences. “Episodic learning” — memories that are stored after only one occurrence — might be mainly stored in the hippocampus.

This may help explain the persistence of some vivid memories in the hippocampus. Memories of emotionally arousing events tend to be more vivid and to persist longer than do memories of neutral or trivial events, and are, moreover, more likely to require only a single experience.

Whether or not the hippocampus may retain some older memories, the evidence that some memories might be held in the hippocampus for several years, only to move on, as it were, to another region, is another challenge to a simple consolidation theory.

Memory more complex than we thought

So where does all this leave us? What is consolidation? Do memories reach a fixed state?

My own feeling is that, no, memories don't reach this fabled "cast in stone" state. Memories are subject to change every time they are activated (such activation doesn't have to bring the memory to your conscious awareness). But consolidation traditionally (and logically) refers to encoding processes. It is reasonable, and useful, to distinguish between:

• the initial encoding, the "working memory" state, when new information is held precariously in shifting patterns of neural activity,
• the later encoding processes, when the information is consolidated into a more permanent form with the growth of new connections between nerve cells,
• the (possibly much) later retrieval processes, when the information is retrieved in, most probably, a new context, and is activated anew

I think that "reconsolidation" is a retrieval process rather than part of the encoding processes, but of course, if you admit retrieval as involving a return to the active state and a modification of the original representation in line with new associations, then the differences between retrieval and encoding become less evident.

When you add to this the possibility that memories might "move" from one area of the brain to another after a certain period of time (although it is likely that the triggering factor is not time per se), then you cast into disarray the whole concept of memories becoming stable.

Perhaps our best approach is to see memory as a series of processes, and consolidation as an agreed-upon (and possibly arbitrary) subset of those processes.