A new study argues that people can remember previously forgotten experiences, including abuse by their parents

Can people remember previously forgotten traumatic incidents? The debate about so-called "recovered" memories has so far been very one-sided. It has focused overwhelmingly on notions that such memories are inaccurate, (witness the media label: "false memory syndrome") and on the perverse antics of therapists driven to extract memories of childhood sexual abuse from their clients at any cost. Bewildered parents accused by their children of bizarre sexual acts make compelling journalism. But why is their version of events unquestioningly accepted?

Media interest in "false memory" was fuelled by organisations founded by parents accused of sexual abuse by their children. Before the advent of these vocal pressure groups, the view that individuals could remember forgotten traumatic experience attracted little media opprobrium. Since the first world war, there has been clinical discussion of traumatic amnesia and memory recovery, particularly among war veterans, a process thought to involve some sort of unconscious "repression". This ill-defined term is now under attack from false memory proponents who say that there is no scientific evidence for repression, therefore recovered memories must be false.

Taking such an argument at face value is dangerous. Integral to the empirical tradition in the social sciences is the need to keep an open mind. The argument as it stands dismisses the phenomenon of memory recovery on the grounds of lack of evidence. But we do not have a complete understanding of how memories might come to be forgotten. Recovering previously forgotten memories may be feasible and the phenomenon may have an explanation other than unconscious repression. Cognitive science has developed since Freud first introduced the notion of repression. There are alternative mechanisms, at least one based on experimental evidence, that could explain why people might inhibit the remembrance of certain experiences until the right trigger occurs.

³ÉÈËVRÊÓƵ

As members of the British Psychological Society's Working Group on Recovered Memories, we felt the need to investigate some of these assumptions. The Economic and Social Research Council funded interviews with those therapists who took part in our earlier survey on recovered memories.

We targeted therapists who had recently seen a client who had recovered a traumatic memory. Our aim was to determine the characteristics of memories recovered in therapy, and their plausibility. The 108 therapists interviewed had seen nearly 700 recovered memory clients between them in the previous three years. Over a third had recovered memories of traumatic experiences that did not involve child sexual abuse. Another third had started recovering memories before entering any kind of therapy. Therapists gave detailed information on 236 cases. Many involved experiences of child physical abuse and other maltreatment. Others involved traumatic medical procedures, such as operations with insufficient anaesthesia or abortions, or witnessing a death of someone close. A few involved bomb explosions and car accidents. It was notable that very few of the reported memories involved implausibly bizarre elements of the type reported to be common by false memory proponents. For example, only 2 per cent involved experiences beginning and ending before the age of 3, and only 5 per cent contained elements of abuse of a satanic or ritual nature.

³ÉÈËVRÊÓƵ

What seems remarkable is the corroborative evidence the therapists told us their clients had put forward. In over 40 per cent of the detailed cases some kind of corroboration was available, and in over a third of these it came from more than one source. Sources included outsiders who had knowledge of the event confirming it had happened, and independent victims reporting abuse by the same supposed perpetrator. Other types of corroboration included confessions by abusers and confirmation of trauma from official records.

How seriously can we take this evidence? It could be argued that these reports are not reliable because they were not observed first-hand, but filtered through a therapist, (although in 11 cases the therapist had actually seen the evidence). Ultimately, whatever criteria are used to judge the validity of this evidence, the same criteria should also be applied to any evidence put forward to justify the supposed perpetrators' claims of innocence.

On the other hand, there is experimental evidence that, through suggestion, some people may come to believe certain (non-traumatic) events have happened to them, when they probably have not. This evidence has been used to defend the argument that if a therapist has a client who recovers traumatic memories this must be because the therapist has used suggestive "memory recovery techniques". But this was not usually the case in our study. Techniques to aid recall were used in 42 per cent of the cases, but they seemed to be used more to help the clients elaborate their memories rather than to recall them in the first place. In only a small minority of cases did the therapist strongly suspect child sexual abuse at the outset of therapy. As one therapist said when details of abuse emerged: "I was as surprised as she was."

Our study includes the largest number of detailed descriptions of therapy clients with recovered memories that has so far been reported. It underlines the importance of conducting systematic enquiries to test popular assumptions.

³ÉÈËVRÊÓƵ

It is apparent that recovered memories do not conform to a single recognisable script: rather there is considerable variation in what is remembered and when. The research supports our original conclusions in the BPS report on recovered memories; that both false and genuine recovered memories are possible.

Bernice Andrews, psychology department, Royal Holloway University of London.