Should someone who has lived through a traumatic event remember it accurately for the rest of their life? Memory research challenges that expectation. An intensely emotional experience may leave lasting memories without protecting a person against losing important details. This also matters when someone has to describe an event in court many years later.
What did Wagenaar actually study?
Willem A. Wagenaar’s work included autobiographical memory: memory for events in one’s own life. In a study published in 1986, he examined his recall of 2,400 events recorded over six years. He noted what had happened, who had been involved, and where and when the event had occurred. He then tested how different combinations of that information helped him retrieve the remaining elements of a memory.[1]
The study demonstrated the importance of retrieval cues. Information about time functioned differently from information about people, places and the event itself. However, observations of one person’s everyday life cannot be directly generalised to everyone who has experienced trauma.[1]
The term “Wagenaar curve” should therefore not be presented as a universal graph for calculating how many details of a traumatic event someone ought to remember after a given number of years. Nor are these studies a scale for measuring how long distress lasts.
Can traumatic events be partially forgotten?
A particularly relevant study by Wagenaar and Jop Groeneweg was published in 1990. The authors compared accounts from 78 witnesses concerning their experiences at Camp Erika in the Netherlands. The earlier testimony was collected in 1943–1947 and the later accounts in 1984–1987.[2]
The general picture of their camp experiences was well remembered. Yet, after approximately forty years, some witnesses no longer recalled important circumstances: the names and appearance of their tormentors, ill-treatment they had suffered, and even having witnessed a murder. The researchers’ conclusion was clear: the intensity of an experience alone does not provide sufficient protection against forgetting.[2]
This is the finding of a particular study, not evidence that everyone who experiences trauma forgets in the same way. It does, however, challenge the categorical assumption that a truly traumatic event cannot be forgotten even in part.
Stress can affect different stages of memory differently
Memory is often discussed as though it were a single ability: someone either remembers or does not. Research, however, distinguishes between encoding information, consolidating it and retrieving it later. Stress during an event and stress while describing it need not have the same effect.
A 2017 meta-analysis by Grant Shields and colleagues covered 113 independent studies involving 6,216 participants. Its findings indicated that the effects of acute stress depend, among other things, on when it occurs. Stress before retrieval generally impaired recall. Its effects during encoding were more complex and also depended on how the material related to the stressful situation.[5]
This does not justify concluding that a nervous witness must have impaired memory. It does support caution about expecting someone to retrieve the same information equally easily under all conditions. Findings about acute stress are also not automatically an account of how a particular person with PTSD functions.
Vivid recollections and memory gaps can coexist
Memory for the details of an event and an emotional response to remembering it are distinct matters. In its description of post-traumatic stress disorder, the US National Institute of Mental Health lists recurring memories, nightmares and flashbacks, as well as difficulty remembering key features of the traumatic event.[3]
There is therefore no necessary contradiction between reacting strongly to a particular reminder and being unable to reconstruct part of what happened. Someone may experience intrusive images or intense fear while being unable to describe everything in a clear sequence. However, this combination does not occur in everyone who has experienced trauma. Experiencing a traumatic event is not, in itself, equivalent to a diagnosis of PTSD.[3]
Confidence in a memory does not guarantee accuracy
A different issue arises from the belief: “I remember it so clearly that I cannot be mistaken.” Jennifer Talarico and David Rubin studied students’ memories of first hearing about the attacks of 11 September 2001, alongside memories of ordinary events. Consistency between later accounts and the initial record declined in both cases, but confidence in the memories associated with the attacks remained high.[4]
The study concerned learning about a catastrophic event, rather than directly experiencing violence. These situations should not be equated. Nevertheless, it illustrates why the subjective vividness of a memory must be distinguished from its consistency with an earlier account.[4]
Omissions, forgetting and contradictions are different situations
Comparing successive accounts requires precision. A witness who mentions a particular detail for the first time is not necessarily contradicting an earlier statement. Adding that the offender was carrying a bag is different from changing “he had no bag” to “he had a bag”. These examples illustrate the difference between an addition and a direct contradiction.
In a 2015 study by Alana Krix and colleagues, participants described a staged theft they had watched on film and gave another account a week later. Details reported only in the second account were almost as accurate as information repeated in both accounts. Experienced police detectives who estimated the expected memory results underestimated the accuracy of those later additions.[6]
A laboratory study cannot establish whether a late recollection of violence is true. It does show why new information should not automatically be equated with fabrication. At the same time, it does not prove that every later addition is correct.
When comparing interview records, it is therefore useful to establish whether a question about the detail was asked earlier and whether the witness ruled it out or simply did not mention it. Only then can the difference be described precisely. The statement “they did not say that before” leaves the issue unresolved.
How can a question influence an answer?
The way an account is obtained also matters. In classic experiments published in 1974, Elizabeth Loftus and John Palmer showed participants films of car accidents. Changing the verb used in a question about the collision influenced estimates of speed. A week later, participants who had previously been asked the question using the stronger term were more likely to report seeing broken glass, although none had appeared in the film.[7]
This does not mean that every question changes a memory. It does show that the wording of a question can influence a later description. How the account was obtained therefore matters when analysing testimony.
A practical illustration is the difference between “What was the person holding?” and “Was the knife in their right hand?” The second question already assumes that a knife was present. The answer may therefore not independently establish that the witness spontaneously remembered that particular object. This is an illustration of the implications of research on suggestion, not a question taken directly from Loftus and Palmer’s experiment.
What can a memory gap establish on its own?
An inability to remember a particular element does not, by itself, explain why it is missing. The studies discussed here do not provide a method for identifying trauma, deliberate concealment or a false account from a single memory gap.
Likewise, subsequently remembering a detail does not independently confirm that the event happened in that way. The source of the information and the possibility of checking it still need to be considered. Wagenaar’s research does not provide a test for identifying a genuine “recovered memory” in an individual case.
When analysing evidence, it is therefore helpful to separate two questions: whether the described pattern of remembering is possible, and whether a particular piece of information corresponds to what actually happened. An affirmative answer to the first does not settle the second.
What does this mean for assessing witness testimony?
The practical implication of the research discussed here is that neither memory gaps nor a witness’s confidence should be treated as a standalone test of whether an account is true. A missing detail may require explanation, but does not in itself establish deliberate dishonesty. Equally, an emphatic account does not replace verification of the circumstances described.
When analysing an account, it is useful to consider:
- how much time passed between the event and the first account;
- exactly what differs between successive statements;
- whether the witness is describing a personal recollection or information learned later;
- what independent material can help verify a particular element of the account.
This is a proposed approach to analysis derived from the findings discussed above, not a credibility test developed by Wagenaar. Nor should every contradiction automatically be attributed to trauma. That interpretation would require a separate basis.
An example: describing an assault several years later
Imagine someone describing an assault several years after it happened. They remember particular words and the attacker’s behaviour, but are unsure of the time, the sequence of two actions and the colour of clothing. In a later statement they add that they heard a phone before the incident. This is a hypothetical example, not the account of a participant in any of the studies.
Those features alone do not justify concluding that the person is lying. Nor is there a basis for treating every detail as true merely because the event may have been traumatic. Each piece of information raises a different question: can the time be established from a document; did the words appear in an earlier message; was the witness previously asked about the phone; and is there independent support for the added detail?
This approach directs attention to specific assertions and their sources. It also helps distinguish a mistake about one detail from the assessment of the remaining elements of an account.
Assessing an individual account requires both an understanding of the limits of memory and careful verification of the facts. Wagenaar’s research helps challenge simplistic expectations of witnesses; it does not replace establishing what happened in a particular case.
Sources
- Wagenaar, W.A. (1986), My memory: A study of autobiographical memory over six years, „Cognitive Psychology”, 18(2), 225–252. Open source
- Wagenaar, W.A., Groeneweg, J. (1990), The memory of concentration camp survivors, „Applied Cognitive Psychology”, 4(2), 77–87. Open source
- National Institute of Mental Health, Post-Traumatic Stress Disorder. Open source
- Talarico, J.M., Rubin, D.C. (2003), Confidence, Not Consistency, Characterizes Flashbulb Memories, „Psychological Science”, 14(5), 455–461. Open source
- Shields, G.S., Sazma, M.A., McCullough, A.M., Yonelinas, A.P. (2017), The Effects of Acute Stress on Episodic Memory: A Meta-Analysis and Integrative Review, „Psychological Bulletin”, 143(6), 636–675. Open source
- Krix, A.C., Sauerland, M., Lorei, C., Rispens, I. (2015), Consistency across Repeated Eyewitness Interviews: Contrasting Police Detectives’ Beliefs with Actual Eyewitness Performance, „PLOS ONE”, 10(2), e0118641. Open source
- Loftus, E.F., Palmer, J.C. (1974), Reconstruction of Automobile Destruction: An Example of the Interaction Between Language and Memory, „Journal of Verbal Learning and Verbal Behavior”, 13(5), 585–589. Open source