Imagine that you went on a fantastic holiday five years ago. Picture strolling along the golden beaches, the waves lapping at your feet, the feel of the warm sunshine and the taste of that delicious coconut drink. If you daydream for long enough you may well be able to convince yourself that you did have that holiday. And as imaginary holidays are much cheaper than real ones, this could be quite a bonus for your bank balance.
This may sound ludicrous, but recent experiments have shown that we are capable of creating false memories simply by imagining them. This is fine if you think you have been on a relaxing holiday, but not if you dredge up memories of traumatic and terrifying events that never really happened.
For some people, this nightmare becomes reality when they believe they have recovered a previously forgotten memory of being sexually abused as a child. Many of these "recovered" memories emerge during therapy or while reading a self-help book. But is it really possible to repress a traumatic memory, banishing it from awareness, only to see it resurface many years later?
This question is fiercely debated among psychologists. Recovered memories of childhood sexual abuse (CSA) have fractured families, at times resulting in criminal trials and civil suits. Their reliability or otherwise is, therefore, a hugely important issue, and psychologists are now devising new tests to measure the true worth of such recollections.
Professor Giuliana Mazzoni of Seton Hall University, New Jersey and Dr Amina Memon of Aberdeen University have recently shown that people can develop a memory of an event that didn't happen to them by simply imagining its occurrence. In a series of experiments, they asked 82 British students to rate the likelihood that they had experienced 20 given events before the age of six.
What the students didn't know was that one of the events could not have happened to them. This "non-occurring" event was: "Having a nurse remove a skin sample from my little finger." This medical procedure has never taken place in the UK. By contrast, another of the events was a relatively frequent procedure that would apply to many - "having a milk tooth extracted by the dentist before the age of six". Three control events were included that may or may not have happened. These were: "Finding a £10 note in the car park"; "Getting sick and going to casualty late at night"; and "Feeling an earthquake".
In the first experiment the participants filled out a questionnaire about the likelihood of these 20 events having happened to them before the age of six. A week later the participants returned and were split into two groups. One group was asked to imagine the milk-tooth event happening to them and read a passage about having a skin sample taken. The other group did the reverse, imagining having a skin sample taken and reading about having a milk-tooth removed. Then each group was asked to fill out a second version of the questionnaire about the likelihood of those 20 events happening to them before the age of six. Finally, the participants were recalled after a further week to fill out a third version of the questionnaire, asking them about the likelihood of childhood events. This time they were also asked to describe any memories they had of the skin-sample event, the milk-tooth event and the three control events.
Amazingly, over the three-week period there was a significant increase in the number of people who believed that the event they had been asked to imagine had actually happened to them. After imagining the skin-removal event, participants were four times more likely to believe that it had happened to them. Some even started to remember details, such as: "I remember the stairs, grey and shiny. The place smelled of disinfectant. There was a nurse, she took my finger. I felt no pain." Equally, memories of milk-tooth removal became more likely after having imagined it happening. By contrast, there was no obvious increase in belief about an event for those people who had simply read about it.
"This study demonstrates that memory is easily malleable," Mazzoni says. "Simply imagining an event made 25 per cent of the participants develop a memory for it and a belief that it had happened." For legal cases that depend upon the reliability of someone's memory, these results have worrying implications. "In the light of psychological research such as ours, the courts will have to consider how a memory was elicited when assessing its veracity," Dr Memon says.
Meanwhile, another group of psychologists have been looking at memory formation from a different angle. Professor Gail Goodman and Dr Jodi Quas of the University of California head a team who have been looking at how likely it is that people will forget a traumatic event that happened to them. They interviewed 175 young adults who, more than a decade earlier, were involved in criminal prosecutions as victims of CSA. A past study had indicated that nearly 40 per cent of adults fail to report their own documented CSA when asked about it, but in Goodman's study only 19 per cent of the participants failed to report their CSA. "This shows that the majority of people who have experienced CSA do remember the events," says Goodman. She doesn't rule out the possibility of repressed memories, but thinks that there are plenty of other reasons why people may not want to talk about their CSA. "Perhaps they were embarrassed, or possibly they were too young at the time of abuse to remember it," she says.
At Harvard University, Professor Richard McNally has been carrying out experiments to test mechanisms of how people either forget and then recover memories of traumatic events, or develop false memories of them. Specifically, he has been looking at whether some people are better at disengaging their attention during a traumatic event, thereby reducing their chances of remembering it. In one experiment, participants were shown a series of words on a computer screen. Some of the words were trauma-related (such as "molested"), some were positive ("charming"), and the rest were neutral ("mailbox"). After each word was shown, the person was told to either remember the word or forget it. At the end they were asked to write down all the words they could remember, regardless of whether they had been told to remember or forget them.
"If CSA survivors have a heightened ability to disengage attention from threat cues, we would expect them to recall few of the trauma words in this experiment, especially the ones they had been told to forget," says McNally. But exactly the opposite happened. Participants who had recovered memories of CSA tended to forget many of the positive and neutral words they had been told to remember, while demonstrating an excellent memory for the trauma words, including the ones they had been told to forget. By contrast, control subjects showed no enhanced memory for trauma words and were better at obeying the instructions to remember some words and forget others. "This shows that people with recovered memories of CSA are not characterised by a superior ability to forget trauma-related material," McNally says.
While not completely dismissing recovered memory, all these experiments are showing that in most cases, recovered memories are probably false memories, generated from our imagination.
Where does this leave people who claim to have recovered memories of CSA? Such memories are all too real and vivid to them, and the distress surrounding the victim is certainly not imagined. In 1991, the False Memory Syndrome Foundation (FMSF) was set up in the US to provide support and legal advice to families and friends of people who claim to have recovered memories of CSA. It was followed by the British False Memory Society (BFMS), which deals with hundreds of UK families whose lives are turned upside down when a member "recovers" a memory of CSA.
In the meantime, psychologists are learning more about the way we store and retrieve memories. Although memories feel very definite to all of us, it appears that they are not as perfect as we would like to believe. It is a little terrifying to think that our treasured memories may not be accurate, but perhaps there is a good reason for our daydreaming ability. Now where was I on my desert island holiday...?