This article originally appeared in The Skeptic, Volume 17, Issue 1, from 2004.
On Saturday, 25 January 2003, a play was presented at the Gulbenkian Studio theatre in Newcastle. The play, written by Josephine Fagan and entitled Paper, Scissors, Stone, explored the relationship between events and memories, interpretation and belief, true and false realities. The production was funded by the SciArt Initiative that supports collaboration between artists and scientists, awarded to Professor Pam Briggs and director and designer Neil Murray. I was invited to lead an interactive discussion on the nature of memory following the performance. I reproduce here, more or less verbatim, my introductory talk.
My talk was preceded by a brief introduction by Neil Murray in which he provided another example of a false memory from his own experience. He recalled how many years ago he and his wife had been on a train when they realised they were sharing a carriage with Paula Yates. Murray had tried to pretend that he was unimpressed at being in the presence of this celebrity, but his wife quickly struck up a conversation with Yates on the nature of motherhood. Murray distinctly recalled the fact that Yates had been embroidering a pair of jeans as she chatted.
Over subsequent years, Murray had recalled this mildly interesting anecdote on several occasions. One evening, he was at a dinner with his wife sitting a few places away from him when he overheard her retelling the tale of her encounter with Yates. “But you didn’t mention the jeans she was embroidering,” he commented. “What are you talking about?” his wife replied. “Don’t you remember?” he retorted. “She was embroidering those fantastic jeans? How could you forget them?” His wife’s response caused him considerable consternation: “She wasn’t embroidering anything – and, in any case, how would you know? You weren’t even there!” Clearly, either Murray or his wife has a false memory for an event that they both insist they can clearly remember. My own comments follow.
I’d like to thank Neil for his introduction and I intend to follow his lead by illustrating the way memory works – and doesn’t work – by mainly relying upon anecdotes. Neil wanted this brief talk to be (and I quote) “engaging and not at all like a science lecture … not too stuffed with facts”. As a scientist, I obviously took very slight umbrage at the notion that a science lecture couldn’t be engaging – but I guess I know what he meant. I will attempt to keep this talk at least 95% “fact free”. But please take it as read that what I will say about memory is supported by a wealth of experimental evidence.
Textbooks often begin their discussion of memory with the rather obvious observation that memory does not work like a video camera. At first glance, it might strike you that it would be better if it did. Imagine if you could just press the right mental button and be treated to an exact mental replay of events you had witnessed in the past – wouldn’t that be fantastic? Well, yes and no. One can think of several situations where it would be advantageous – settling arguments between Neil and his wife, for example. But I only have to think about the disorganised state of our video collection at home to realise some obvious problems. How on earth would you ever find the memory that you wanted?
But the problem isn’t just one of locating the right bit of mental footage. Unlike material recorded on videotape, memories can and do change over time. A memory is more like a dynamic mosaic than a series of static unchanging frames recorded by a video camera. The very first time that you try to recall an event, even a recent one like getting to the theatre tonight, you will engage in a constructive process – you will literally build that memory from different types of bits and pieces. Some of those bits and pieces will correspond to more or less accurate memory traces laid down at the time you witnessed the event, but even then you may not put them together properly.
Furthermore, the mosaic that you build will be influenced by things that happened to you before the event in question, that led you to have particular expectations and a particular way of viewing the world. It will also be influenced by your current view of the world and of yourself. There will also be a general tendency to fill in any gaps in such a way that the whole ‘makes sense’.
In general, we do not remember the surface details of events at all well; we remember the gist. We automatically extract the important essence of events and forget the superficial and transitory packaging. This is both the great strength and the great weakness of the way our memories work. It is a strength because we don’t process and store all of the minutiae of life. We pay attention to the important stuff and forget the rest. Why lay down in memory a complete verbatim record of every conversation you ever had, every song you’ve ever heard? But the weakness is that sometimes we do need that level of detail and it will probably elude us. Or, worse still, we may confidently believe that a memory we hold is a true reflection of an event when in fact it may be distorted beyond recognition.
There are only a few areas where the accuracy of our memories is so important that we make any attempt to assess it. Examples include forensic psychology – it appears that we are probably far more impressed by eyewitness testimony to crimes than we should be. Another area where it becomes important is my own speciality, which I call ‘anomalistic psychology’ for want of a better description – the psychology of unusual experiences and beliefs. Just how accurate are eyewitness reports of UFOs, ghosts and the Loch Ness monster? Or even alien abductions? And of course there’s always the settling of marital arguments, as we’ve seen.
Most of the research on eyewitness testimony has been driven by the need to understand factors affecting the reliability of reports from witnesses to crimes. It is now generally accepted that such reports can often be wildly inaccurate, leading to gross miscarriages of justice. The circumstances surrounding crimes are often precisely those that will lead to poor recall. The event is unexpected, often over in seconds, and sometimes extremely frightening. Or it may be that the police need details of events that preceded a crime – events which no-one at the time realised would be that important. One obvious reason that we may fail to remember things accurately is simply that we failed to pay attention to the right details at the time. Typically, we pay attention to the information that is relevant to our goals at the time – and it can sometimes be amazing what we miss!
I can illustrate this with a true story of a visit to an estate agent that I made with my wife, when we were looking to move house about ten years ago. As we left the estate agent’s office, having looked at details of various houses, I said to my wife, who is also a psychologist, “That was very strange. It reminded me of a psychology experiment.” She was rather confused by this and asked me what I meant. I said “You mean you didn’t see it?” “See what?” she said. I told her to look through the estate agent’s window to see if she could see anything a little bit unusual. She did – and could not believe that she had failed to spot a full-size stuffed bison that for some unknown reason was on display in the office! This nicely illustrates the fact that people may vary in terms of what information they encode at any particular moment. My wife was very focused on house buying – what price, where, how many bedrooms – my mind was perhaps not so fully focused on the important task in hand.
Recent research into what is known as ‘change blindness’ provides another illustration of our inattention to aspects of our surroundings. In a typical study, people queuing at a library issue desk are handed a form to fill in by the librarian. At one point in the interaction, the librarian disappears from view, as though retrieving a dropped piece of paper, but another completely different person emerges in their place. Around half of the participants simply do not notice the change.
A huge amount of experimental evidence, in addition to our everyday experience, shows us that our memories are poor if we haven’t paid attention in the first place. No big surprise there. But what about those situations where it’s really important that we pay attention and get things right? Even here, our memory can play cruel tricks. Donald Thomson, a psychologist in Australia, was arrested by the police and forced to take part in a police line-up. He assumed he was being harassed in response to his strong views on the unreliability of such line-ups. Things got very serious, however, when he was identified by a very distraught woman and told that he was being charged with sexual assault.
It transpired that the assault had taken place with the television on in the room at the time of the attack. The program being shown was a live debate on the reliability of identity parades, featuring both Thomson and the Assistant Commissioner of Police. The victim had unintentionally based her description of the assailant on Thomson who was on television at the time. Fortunately, he had a large number of viewers to provide him with a watertight alibi.
There are, of course, those vivid memories that we just know are right. One example is so-called flashbulb memories. We can all remember with perfect accuracy where we were, what we were doing, who we were with, when we heard about the attack of September 11th – can’t we? There seems to be something about such moments that burns the details into our very brains cells.
I can still remember, for example, hearing the news of John F Kennedy’s assassination. I was only 7 years old at the time, so when I heard the newsflash on TV, it didn’t mean that much to me. But I remember ambling into the kitchen to tell my mum and dad about it anyway. It was from their reaction that I realised that this was news of stupendous importance.
I used this example of a flashbulb memory in my lectures for years in my adult life and on one occasion happened to mention this to my mother. She told me that it just didn’t happen that way at all. We were not at home and it was not me that broke the news. Interestingly, I had been the victim of a false memory that put me right at the heart of the action!
Again, experimental evidence has shown that flashbulb memories – frequently held with great conviction – are often just plain wrong. American students recorded details of how they heard the news of the Challenger disaster the morning after it happened – who they were with, what they were doing, and so on. A couple of years later, many of them had completely different recollections of that event.
It appears that when we try to recall something, the mosaic memory we bring to consciousness consists of memory traces of the original event plus other memory traces, perhaps relating to other similar events or even to daydreams or fantasies. Gaps in memory will sometimes be effortlessly and automatically filled in to produce our recollection – and we will often have no way of knowing which bits we can trust and which we cannot. Sometimes we may be fooled into thinking that
something really happened when in fact we only imagined or even dreamt it. Our ability to distinguish between memories for events that really happened and those memories that are internally generated is known as reality monitoring. An everyday example is trying to decide whether you really did lock the backdoor or just thought about it. At the other extreme is psychotic breakdown in which the sufferer is totally unable to distinguish between mental events and events out there in the real world.
Although psychologists have long recognised that firsthand accounts of witnessed events were unreliable, it is only within the last decade or so that research has been directed at the possibility that people may sometimes have rich and detailed memories for events that they have never actually witnessed at all. The main reason for this explosion of research into false memories was the sudden increase in cases of alleged recovered memories of childhood sexual abuse, especially in the USA.
Typically in such cases, adults would enter psychotherapy suffering from a variety of common psychological problems such as depression, low self-esteem, or insomnia. As part of their psychotherapy they would engage in mental exercises, such as hypnotic regression and guided visualisation, intended to unlock any repressed memories of traumatic childhood events thought to be causing their problems.
Many thousands of people who had entered therapy with no conscious memories of being abused as children became convinced that their now-aged parents had indeed inflicted terrible suffering upon them decades earlier. In some cases, these allegations included claims of satanic ritualised abuse, involving human sacrifice, cannibalism, sexual torture, and forced abortions. Many of these cases went to court and led to convictions even though the cases rested entirely upon verbal testimony. Families were torn apart in the most brutal way imaginable.
Experimental psychologists tended to doubt the accuracy of the memories recovered via hypnosis and related techniques. A huge amount of experimental evidence shows quite convincingly that hypnotic regression does not provide a magic key to unlock the unconscious mind, forcing it to reveal its hidden memories. Instead, the hypnotic regression procedure is such that it provides a context in which individuals often produce an account mixing fantasy with pre-existing knowledge and expectations – and then believe with total conviction that the account reflects events that really took place.
Indeed, experimental psychologists have expressed doubts about the very concept of repression itself. The idea that the unconscious mind can somehow automatically take over and hide away memories for traumatic events is not supported by any convincing experimental evidence, although there are many accounts of what appears to be repression occurring outside the laboratory.
In the early days of the controversy, those who believed that recovered memories were largely accurate would sometimes object that, although memory for peripheral details of a witnessed event might be distorted, there was little evidence that people were prone to false memories for episodes that had never actually occurred at all.
Things have moved on since then, thanks to the pioneering work of Elizabeth Loftus, among others. We now know that it is alarmingly easy to plant false memories in a sizeable minority of the population using well-established experimental techniques. It has been shown, for example, that hypnotic regression is not the only way to induce false memories. Simply getting people to imagine events that did not actually take place is often sufficient to lead people to believe that they did witness or take part in the events in question.
The difficulties of deciding whether a memory reflects a real event or not is illustrated by something that happened to Elizabeth Loftus herself. Loftus’s mother had died by suicide. She had drowned herself in a swimming pool, but Loftus had never actually seen her mother’s body – or so she thought.
Many years later, and long after Loftus had established her reputation as the leading psychologist in false memory research and one of the main critics of the concept of repression, she was attending a family get-together when one of her uncles insisted that she had in fact seen her mother’s body. He said that Loftus was the person who had found the body floating face down in the pool. An image of her mother’s lifeless body immediately filled Loftus’s mind. She was flabbergasted – for years, she had questioned the notion that the mind just locks away memories that are too ghastly to face. But it appeared she had done just that.
Over the next few days, her memory of that terrible sequence of events become clearer and more detailed as she dwelt upon this horrible revelation. And then she called her brother to tell him – and he said the uncle was wrong, Loftus had not found the body! This was confirmed by other family members. Far from experiencing a recovered memory, Loftus had been the victim of a false memory. I should, however, point out that I only think I’ve recounted that story accurately, as I couldn’t find the book where I read it when I wrote this talk – so maybe the whole thing is a false memory on my part?
The events portrayed so well in the play show all too realistically how our views of things can change over time. Initially, the brother and sister are viewed benignly – although one or two aspects of their lives are generally thought to be a little odd. But over time this view gradually changes, and with it the very memories that provide the evidence for those views. Each of the villagers make small contributions to the changing story, but the cumulative effect is like the sea eroding a rock. The final ‘truth’ explains everything – all the pieces of the jigsaw fit. Except, of course, that from our privileged viewpoint, we can see that this accepted ‘truth’ is far from historically accurate. There can be no doubt that the stories we tell ourselves and believe to be true, on both the individual level and on the societal level, are often just as fictitious.
For now, I’d like to finish with a somewhat more light-hearted practical demonstration of how our memory can play tricks. No doubt you would expect that you would remember a simple stimulus accurately if you had been exposed to that stimulus literally thousands of times during your life, especially if it was something a little bit odd and unexpected? If you’re wearing a watch, have a look at it to see if the numbers are represented by Roman numerals. If it is, without looking at your watch again, I want you to ask yourself this simple question: How is the number ‘4’ represented on your watch?
Now all look at your watch again. I suspect many of you will be surprised to see that it is represented by ‘IIII’. If it is, how many hundreds of times have you looked at your watch and never noticed that the four is represented in this unusual way? Almost everywhere else, four is represented as ‘IV’ in Roman numerals, but on the vast majority of clocks and watches it is represented as ‘IIII’. But we often see what we expect to see and remember what we thought we saw.
My wife, Anne Richards, and I carried out a little study based upon this. We showed people a clock with Roman numerals on it – our kitchen clock as a matter of fact! Some people were told to draw the clock from memory, others to copy it while in full view. Those who copied it tended to represent correctly the four as ‘IIII’; those drawing from memory misremembered the four as ‘IV’ in line with their general expectations for Roman numerals. We argued that people rarely notice this ‘oddity’ in everyday life because you don’t need to pay attention to the numerals themselves to tell the time – only their relative position is important. As I said earlier, sometimes our memories will be inaccurate simply because we didn’t pay attention to certain details at the time.
But even here there’s an interesting postscript to the story. When we wrote up this paper for the British Journal of Psychology over 10 years ago, we included an account of how this odd aspect of clocks first came to our attention. Please indulge me while I briefly quote from the paper:
The inspiration for this study was supplied by an incident involving the first author (CCF) and the second author’s daughter, Lucy Richards, a couple of years ago, when the child was aged about eight. In the course of visiting her grandparents, Lucy’s attention was caught by the Roman numerals upon a clock face in the room. The conversation then proceeded something like this:
Lucy: On the clock, why does ‘V’ come after ‘IIII’?
CCF (without looking up): It doesn’t say ‘IIII’. It says ‘IV’ for four.
Lucy: It doesn’t. Look.
CCF (looking at clock): Incredible! You’d think clock-makers of all people would know Roman numerals! But this is how it should be. (Shows his wristwatch.) Would you believe it, they’ve got it wrong here as well!
My wife agrees that this is more or less how it happened – except she insists that the conversation was between her and Lucy, not me and Lucy!
So we’re back, more or less, to where we started …
