UK United Kingdom

Explainer: what is memory?

Memory is difficult to define without being circular. People often define memory as “something you can remember”. But we cannot deny the existence of a memory when there is no recollection. Sigmund Freud…

Did you forget to lock the door, or just forget to pay proper attention? jef safi \ 'pictosophizing

Memory is difficult to define without being circular. People often define memory as “something you can remember”. But we cannot deny the existence of a memory when there is no recollection.

Sigmund Freud was first to theorise this notion: memories that are not consciously remembered can exert control through the subconscious. Although much of his research methodology is now criticised, Freud opened our eyes on how past events can influence us in ways regardless of our awareness.

So what is memory? I would define memory as “a past that becomes a part of me”. Interestingly, past events can become a part of us in three different ways - as procedural, semantic, and/or episodic memory.

These different ways often occur simultaneously. Learning to ride a bicycle involves procedural memory, in which our bodies remember how to coordinate different limbs and perceptions based on previous attempts at bike riding.

Remembering that gears are used to switch the wheel rotation circumference represents semantic memory.

I can also recall when, where, and with whom I learnt how to ride – an example of episodic memory.


Despite distinctions between the three, all the types of memory appear to undergo similar stages across time. First, attention is critical for the initiation of memory formation. I often lock the door as I am leaving my house without paying attention. Subsequently, at work, I cannot recall whether I have locked my door or not.

This is not due to forgetting, but is a failure in memory formation. It is probably what happened to the police who lost their keys to the Wembley stadium during the recent Olympics.

Attention then leads to working memory, which lasts seconds to minutes. Once an event forms a short- or long-term memory, we are no longer thinking about the event, but retrieval is possible hours to weeks, or months to years, following the event.

Biologically, short- and long-term memories are different from each other, as indicated by the Nobel laureate Eric Kandel’s work.

Long- but not short-term memory involves long-lasting physical changes in neurons that ultimately result in altered levels of electrical and/or chemical communication between different neurons.

Such a biological correlate of memory strongly suggests individual differences in memory abilities. That is, just as our physical abilities are tied to our bodies, our memory abilities are tied to our brains.

On the flip side, we can improve our memory with exercise, just as we can improve our physical abilities with exercise. This makes sense, as the brain is a part of our body.

jef safi \ 'pictosophizing

Nevertheless, the brain is a special and mysterious organ because there exist people with a special ability referred to as “eidetic” or “photographic” memory. Such people display the ability to recall images, words or sounds with extreme precision.

In 2001 the British architectural artist Stephen Wiltshire was filmed flying over London aboard a helicopter and subsequently completing a detailed and perfectly scaled aerial illustration of a four-square-mile area of the British capital within a few hours. It’s a talent he has repeated on several occasions.

In the first half of the 20th century, the Russian mnemonist Solomon Veniaminovich Shereshevsky showed himself capable of recalling strings of random words he had seen decades earlier.

But does eidetic memory exist? Definitely not in the form of the popular misinterpretation of the term that assumes total recall of all events occurring in one’s life.

The only report of such ability comes from Dr Sheldon Cooper from the TV comedy show The Big Bang Theory. In fact, much like any other memory, eidetic memory is affected by factors such as attention, significance, duration/frequency of the event and, most importantly, mnemonic tools.

Shereshevsky reported using a mnemonic technique in which he imagined laying out objects along a stretch of a familiar road during the learning phase.

During the retrieval of those words, he went back and picked them up one by one in his imagination. Such a report proves that his memory was actually highly trained normal memory. Many who claim to have eidetic memory use such mnemonic techniques.

In fact, the existence of eidetic memory as an independent, special memory ability compared to an average person’s memory is very controversial.

Prodigies such as Wiltshire can only retrieve their memory within days, leading some to argue that those cases reflect a form of very strong working or short-term memory.

But there is one extraordinary report of long-term eidetic memory. In 1970, a Harvard vision scientist called Charles Stromeyer published an account in Nature of his subject, Elizabeth.

jef safi \ 'pictosophizing

Elizabeth could recall random dot patterns shown independently to each of her eyes with such precision she could reproduce the 3D image of fused dot patterns even three months after initial exposure to the images.

I urge you not to undermine the credibility of this study due to the fact that Stromeyer married Elizabeth. My bigger concern is how Stromeyer did not attempt any further studies on this incredible result.

Other than this single study, there is no evidence eidetic memory is actually fundamentally different from how an average person acquires, stores and retrieves memory. The difference appears quantitative rather than qualitative.

It is interesting that compared to extreme physical abilities, mental abilities are seen as more extraordinary. People who display eidetic memories are extensively studied in order to understand their beautiful minds. But no-one cares how Usain Bolt runs: in essence, we don’t think he runs differently from us – it’s just we are not that fast.

Whether it is average or special, the subject of memory will continue to fascinate us.

Kuato from the original Total Recall movie stated: “A man is defined by his action not his memory”. But his action will be defined by what he remembers and believes in.

Maybe that’s why scientists are working very hard to understand how memory works. On the other hand, understanding how forgetting happens is rather neglected (I am working on forgetting).

Just as super memory can exist as an extreme form of normal memory, perhaps erasure is also possible as a form of super forgetting.

See more Explainer articles on The Conversation.

Join the conversation

20 Comments sorted by

  1. Robert Moore


    Side by side with remembering is mis-remembering. Is it the same process as remembering correctly? Do we assimilate what we think are the "facts" and exercise that particular neural pathway and cement that disordered memory, or is there some override by our own subconscious to get the result we would rather have?
    Also, Freud's concept that, "memories that are not consciously remembered can exert control through the subconscious" is no better illustrated than through dreaming, and this may explain why I find most dreamscapes in narratives transparent and clumsy. If we are processing our unbidden memories in dreams then these contingent images are conjured from our past. In narratives, often the dreamscape foreshadows important narrative plot points and is, therefore, predictive.
    There is a contest here between contingency and desire which Freud could have a field day with :)

  2. Mike Swinbourne

    logged in via Facebook

    I am disappointed in this article, because I had hoped that it would go some way towards explaining the title - 'What is Memory"? I don't mean in a philosophical sense, but in a biologicial, physical sense.

    Is memory a hard-wiring of neurons etc, and if so, how does it occur? How does it change? How does it go wrong?

    There is obviously some genetic or physical basis to memory, if you think about animals which rely entirely or largely on 'inherited' memory. Consider - for example - a spider…

    Read more
    1. Greg Boyles

      Lanscaper and former medical scientist

      In reply to Mike Swinbourne

      If you want an analogy to help you understand then you need only look at a computer.

      Its DDR400 RAM (or what ever it has) is equivalent to short term memory Like all computer data it is stored as 1s and 0s or more precisely transistors that are in an on state and transistors which are in an off state. As soon as the PC is powered down all the infor stored in the DDR400 RAM vanishes.

      So you might compare the transistors in the DDR RAM to junctions between individual neurones that are either…

      Read more
    2. Greg Boyles

      Lanscaper and former medical scientist

      In reply to Greg Boyles

      That is as I understand it anyway from uni and reading on the subject.

      But subsequent new research may have modified this picture slightly or added more complexity to it.

    3. Ben Brooker

      inquisitive go-getter

      In reply to Mike Swinbourne

      Totally agree Mike!

      There have been so many studies commissioned recently that are investigating whether parts of memory are stored in DNA.

      It's interesting to note that moths remember what they learn as caterpillars even though they go through metamorphosis (see link below). Surely these memoires are stored in DNA...

    4. Sean Manning


      In reply to Ben Brooker

      I think the idea that some of the caterpillars brain persists through the metamorphosis is more likely.

      For memories to be stored in DNA the sequences would need to be altered and then read. I'm not sure there is a known mechanism for that.

    5. Jee Hyun Kim

      Head of Developmental Psychobiology Lab at The Florey Institute of Neuroscience and Mental Health

      In reply to Mike Swinbourne

      My aim was to give a very brief overview on the general concept of memory, in a way that any non-expert can understand. Also 'inherited memory' is really not a memory, but rather an instinct. A spider does not create a web from a past event that has happened in its life. Just like how we do not need to have a past sexual experience to have a sexual experience.

      As for the biological basis of memory, the articles are coming in the future ;) I can't promise how quickly, though...

    6. Jee Hyun Kim

      Head of Developmental Psychobiology Lab at The Florey Institute of Neuroscience and Mental Health

      In reply to Sean Manning

      You're right. There is an emerging concept of 'epigenetics'. Simply put, it means that genes (factors that regulate gene expression) can be altered due to the environment. It explains why some identical twins can start to look very slightly different as they grow and age.

      There is some evidence that epigenetics can be inherited. Also, aversive learning can cause epigenetics. Whether such epigenetics caused by aversive learning can be inherited is an unknown matter.

    7. Greg Boyles

      Lanscaper and former medical scientist

      In reply to Ben Brooker

      Or the possibly that what you are calling 'memory' is in fact merely programmed behaviour as a result of neural pathways that always develop in the same way in all caterpillars.

    8. Greg Boyles

      Lanscaper and former medical scientist

      In reply to Jee Hyun Kim

      The point is that epigenetics only concerns the expression of genes and NOT the genes themselves which remain as immutable as ever.

    9. Sean Manning


      In reply to Greg Boyles

      I as going to say the same thing but the article he has linked us to does indicate that the catapillers did learn a new behaviour.

      Your example is similar to the web building of spiders and the child rearing of lower animals. They don't remember how to do it, they just come with the correct firmware.

    10. Greg Boyles

      Lanscaper and former medical scientist

      In reply to Ben Brooker

      Just read the article and the author makes no suggestion that memory might be stored in DNA anyway.

      As some one else has suggested, part of the caterpillars nervous system(along with learned behaviour) is carried over through the pupal stage into the adult moth.


      Blackiston was convinced that some aspect of the caterpillar’ nervous system was carried over into adulthood. However, he also found that this only happened if caterpillars are trained at the last possible stage before they…

      Read more
  3. Sean Manning


    References to The Big Bang Theory AND Total Recal (the good one) in one article. You just gained a fan =D

    Also, interesting subject matter.

  4. Martin Linder

    logged in via Facebook

    "Once an event forms a short- or long-term memory, we are no longer thinking about the event, but retrieval is possible hours to weeks, or months to years, following the event"

    I was under the impression that our memory is no longer considered an object stored and retrieved as this quote to me seems to imply, but is rather reconstructed continuosly. Which is why memories change dependant on time and circumstances.

  5. James Hill

    Industrial Designer

    In the popular psychology book from a few decades ago "I'M OK You're OK"
    the introduction describes a patient whose brain is being stimulated by a micro voltage probe.
    The patient "relives" past events under this stimulation, not remembers, while still remaining conscious on the operating table.
    If true, this implies that all experiences are recorded and only some are recalled.
    It also allows some Sheldon Cooper individual who can relive past events drawing upon a totality of recorded experiences.

  6. Anthony Muscio

    logged in via Facebook

    Now I am reading "Thinking Fast and Slow" and the idea of System 1 (associative) and System 2 (logical) brain systems, I feel I need to revisit models of recollection. The brain evolved and thus it needed to capture in time and within the Genes coded "knowledge" that bootstraps the brain so it can learn. Rapid recollection, heuristics or gut responses are necessary for survival. With additional memories laid down in the context of they way the brain works. The Rational brain seems to be an evolutionary…

    Read more
  7. Debra Joan Smith

    Account Executive

    It is so great to read such sensible and illuminating material. I am noticing differences in my memory after 2 induced brainstem clots and a short coma. My memory used to be like fly paper- most stuff I attended to stuck but now, even though my long term memory is not altered, I am more forgetful. The other day I started to make a bed, got distracted by putting something away in another room and later in that day, I was horrified to realize that I had walked away with the bed only half made and I had thought I completed the task. This seems like a small example but it is frustrating to me - mostly because of my assumptions- I can no longer assume that I completed a task I clearly remember starting.