Age and Memory Muddles

'I got up for something - but now I can't remember what it was.'

I cannot imagine a happier working lifetime than I have spent researching cognitive ageing but I do get uneasy when people ask me to explain, “Just what have you  found out?” (with the clear sub-text “after all these years”). The pawky “scientific” language in which my colleagues and I describe hypothetical mechanisms that might plausibly support bits of our mental lives does not map on to our marvellous everyday languages, created and enriched by writers poets and philosophers, that elegantly convey and interpret our  conscious experience of all the glitter and wonder of the world.

This gap becomes especially clear when non-psychologists (NPs) and   Psychologists (Ps) try to talk to each other about memory – an enthralling capability because the images, words, sounds and smells it displays are the cast, dialogue, stage, scenery and backdrop of the bright theatres of our consciousness (or, admittedly, for some of us, the gloomy flea-pits of our minds). For four centuries Marvell’s “The mind, that ocean where each kind does straight its own resemblance find” 1 has been the most vivid and concise statement that memory is both marvellously capacious and startlingly efficient because we need only a few hundredths of a second to recognise, name, eat, reject or use any of the millions of different objects in our perceived worlds.



Shared experiences of their memories should make it easy for Ps and NPs to have useful conversations. So we can, up to a point, so long as we recognise that we use very different kinds of metaphors to describe what memory does and how it works. Moreover, when NPs use metaphors they know just what they are doing, and do not mistake playful  illustrations for reality. Ps ,  always pathetically needy for professional dignity, call their metaphors  “models” or, when they really do not understand what they are talking about, “frameworks” and they actually do believe that these  are precise descriptions of function.

Sometimes Ps and NPs can share the same metaphors. We all agree that we have both short-term and long term memory, and that these are different. A majority of decent, diffident NPs think of Short Term Memories as lasting for hours or, at most, days and weeks 2.  More sophisticated NPs, often mischievous philosophers, explain to me that Short-term Memory (STM) is a different system than Long Term Memory and that it lasts only seconds. The capacity of this envisionment of STM is estimated by the number of words or digits that we can repeat back in the same order that we experienced them. Both diffident and sophisticated NPs are happy with similar metaphors to explain why STM can only briefly hold a little information: It is like a small warm cup into which we drop frozen images of words or numbers. These immediately start to melt, so the number we can recover is sharply limited by rapid rate of mind-melt. NPs and Ps agree that, by some unexplained trick, we can pick particular things out of our cups before they vanish and lodge them in the permafrost of long-term memory where records of words, deeds and images may possibly last as long as our brains survive.

When I was an undergraduate this was  the kind of metaphor used by mathematically-minded Ps like Wayne Wickelgren who worked to estimate the rates of rot of things briefly held in STM. He also pointed out that things that are  similar, like words that sound nearly alike, (e.g. cat, rat, mat) will earliest become confused with each other as rotting blurs them. (This was very long ago and WW is remembered for many other different and excellent things he accomplished before he, sadly, died of Lou Gehrig’s disease).

The problem is that after thinking a little more about our short-term memories both Ps and NPs realise that this is a threadbare description and misses the point of what memory is for.  Before Wickelgren’s work my kindest and most articulate mentor Alan Welford  pointed out that

“…… are somehow held in a form of short-term storage while other data are being gathered. Obviously unless data can be so held the amount of information that can simultaneously be applied to any problem is very small indeed … old people the amount that can be stored tends to diminish, and that which is stored is more liable than it is in young people to interference and disruption from other activity going on at the same time. Such a decline in short-term retention would be capable of accounting for a very wide range of observed age changes in learning and problem-solving” 3.

This is a language that Ps and NPs can share. Memory is not a passive receptacle of slowly degrading information but a busy, active system for understanding and keeping up with a fast changing world. Most crucially, it is not there only to display recorded fragments of the past but to actively predict what is going to happen next. Since, on this account, Working Memory is needed to solve intelligence test problems it also has to be taken seriously as a functional description of what intelligence is. A provocative finding is that age has little or no effect on how accurately we remember lists of words or numbers in the order we were given them (in our North English samples word and digit “spans” changed little or not at all between age 50 and age 90). In contrast, age sharply reduced the accuracy with which people could juggle the same number of items in our minds, as by repeating them backwards.

Fifteen years after Alan Welford’s summary, Alan Baddeley and Graham Hitch 4 captured these ideas in a neat phrase, “Working Memory”. I remember, with embarrassment, that when they first used the term at a London meeting of the UK Experimental Psychology Society it seemed so all-embracing that it was mocked as “whatever bit of memory you happen to be working on at present”.  This has  proved to have been fatuously irreverent because Working Memory  is now a caption for so much research in cognitive psychology that it is almost a “Grand Theory of Everything” in our small science. As Welford noticed, we can only cope with a rapidly changing world by continually rearranging data that we have just registered to interpret what has just happened; relate this to what has just happened and to what we have already known for ages and to work out what we expect to happen next; make plans to deal with this  prediction; choose how and where to find more information to do this and suppress irrelevancies; switch attention between different activities as becomes necessary and, while all this is going on, still keep a firm  grip on what it is that we need to look out for and to do (“Goal Maintenance”). These frantically active processes cannot be imagined as movements of  fading shadows on the screens of our minds. They imply active control.

What metaphors can we find for the controller, and how far can Ps and NPs share these? In everyday language we might call it a “Self”, or “Ego” or even “Soul” or “Spirit”. Descartes would have housed it in the pineal gland. Alan Welford was a clergyman, so may well have considered these alternatives, but he was also an Anglican and kept this, and all other hints to his  beliefs to himself. In 1974 Alan Baddeley and Graham Hitch  4 found a term in better tune with their  period:  the “Central Executive”. This  conveys harassed middle-management tasked with getting things done. When young this creature of the 1960’s and 70’s is ostentatiously competent and brash 5.


young executive


What happens as it ages? It may help to consider what we know about the brain-basis of working memory

Studies of patients with localised brain injuries suggested that damage to the frontal and pre-frontal brain cortex causes problems with just the tasks that Alan Baddeley set his executive. Suggestions that age depletes brain tissue earlier and faster in these than in other parts of the brain encouraged experiments to test whether elderly people also perform unusually poorly on tests used to diagnose frontal damage in neurological patients. Early results were unconvincing, probably because individuals experience different patterns of brain changes as they age. These vary so much between persons that we may not find consistent averages if we can only collect behavioural data on groups of 50 or fewer people. As brain scans have become better and more easily available they have confirmed that, in general, frontal, pre-frontal and medial temporal cortex do seem responsible for activities that fit many of the job-descriptions suggested for the central executive 6.


working memory frontal lobes

So far so good. We have a lively metaphor that both Ps and NPs can share; a job-description for the Central Executive and an address for his office. At this point NPs might nod politely and wander off to wherever their personal executives decide to take them.  Can we Ps usefully prolong the conversation? A recent collection of essays on “Working Memory and Ageing”7 seems promising.

I am a keen fan of this sort of stuff because it is just what I have tried to do for the last 50 years. One of these nice essays suggests that decrepitudes of the ageing executive are best accounted for by general changes in all aspects of brain efficiency, such as “global slowing”. Performance on all sub-tasks begins to fail as the entire system runs out of juice. Others think that the best way forward is to try to measure and compare different amounts of declines in each of the various executive abilities we can identify: For example they test whether “updating” is affected more than “goal-maintenance” or “switching” between different activities?   Whether records of visual information on the executive’s “scratch pad” fade faster than records of the sounds of words? This sometimes seem a questionable way to go about things because it seems to forget that “updating”, “switching”, “re-arranging information” etc. are only common language metaphors for the effects that working memory achieves (the things that it seems to do) and that it is rash  to suppose that each of these partial task-descriptions must be supported by a corresponding different, and probably independent process. This line of argument also seems to miss a difficult methodological problem that, before we can decide whether age affects some  tasks such as “switching” more than others such as “rearranging information” or “updating” we must find a way to equate them for difficulty on some absolute scale rather than begging a question by simply ranking  how hard most people find them.  I can’t think of satisfying ways to do this.

Rebecca Charlton and Bob Morris8 suggest a way forward by evicting the executive from any single office. They discuss interesting clues that its activities are distributed over much of the brain and so depend on richness and speed of connectivity between nerve cells that are impaired as many of them age and die leaving as markers their tiny sad tombstones “white matter lesions”. As connectivity suffers so must performance on all sub-tasks.

It is unnecessary to talk about a collection of quasi-independent devices:  one “updates”, another “switches” a third “holds goals” another holds “images” and another “noises”. Steve Jobs was far too smart to design i-pads this way. He recognised that we can run apps for each task on a common processer and associated circuitry. If i-pads gradually aged, rather than suddenly and totally blanking  with the blue screen of death, we might well find that long declines affect some apps earlier and more than others. Comparing computers in terms of bench-mark characteristics, such as processor-speed and RAM capacity also reveals which programs are most likely to be affected by changes in these system performance characteristics and why particular programs can run on some systems but not on others. As far as I know, most Ps still don’t think about working memory in this way.

Until we find better descriptions have P’s anything interesting to tell NPs than Alan Welford and Alan Baddeley said long ago? Their descriptions are certainly clever, and NPs can surely recognise this, just as they can recognise the skill of Sudoku experts, but without feeling that this in the least enriches our mental lives or tells us anything that we really care to know.

Non-scientists are not less smart than scientists. They are just smart about different things. They keenly appreciate science when it tells them something that illuminates our human experience. We need marvels to dazzle and nourish our minds, to enrich appreciation of the Entire Grand Shebang and to understand our place in it. We are excited and awed even by our vague understandings of difficult topics such as the shadowy existence of Schrodinger’s cat; different passage of time for long-distance cosmonauts  and stay-at homes; Black Holes; Big Bangs; Quantum Entanglement and the speculative Multiverse. We eagerly buy, and quite often read, expensive books about them. Even vague comprehensions of these wonders are satisfying, and provoke nourishing thoughts and ideas.

Cognitive Psychologists (CPs) have tried our best for more than a hundred and fifty years, but I find it hard to convince  NPs that our most recent work has more to tell them about their inner lives than manuals of how to do Sudoku or play Go. Are our metaphors, which we call models, intellectually un-nutritious because the more detailed they have become, the more dull and artificial they seem to anybody who has not worked to devise them?  Or are we just incompetent at translating them for others? When people ask me what I have been doing for the last 50 years must I continue to hedge and say that I do not think that they would really much like to hear about it but, whatever it has been, I   really have enjoyed it very much indeed.

1.Andrew Marvell. c 1664. The Garden.

  1. See use of Short-Term Memory in “Alzheimer’s  Reading Room”
  2. Welford, A.T. (1958) Age and Human Skill. Oxford, Oxford University Press.(p285).
  3. Baddeley, A.D. & Hitch, G. (1974). Working memory. In G.H. Bower (ed) the Psychology of Learning and motivation, (Vol 8, pp 47-89). New York
  4. crf John Betjeman, (about 1960) “I am a young executive. No cuffs than mine are cleaner; I have a slimline brief-case and I use the firm’s Cortina”.
  5. Stoltzfus, E. R., Hasher, L., & Zacks, R. T. (1996). Working memory and aging: Current status of the inhibitory view.Working memory and human cognition, 66-88.
  6. Logie, R.H. & Morris, R.G. (2014). Working Memory and Aging. London and New York, Psychology Press.
  7. Charlton, R. and Morris, R.G. (2014). Associations between working memory and white matter integrity in normal ageing. In . Logie, R.H. & Morris, R.G. Working Memory and Aging. London and New York, Psychology Press.


About Gray Rabbitt

Grumpy gerontologist
This entry was posted in Uncategorized. Bookmark the permalink.

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Google photo

You are commenting using your Google account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s