Why children’s accounts of the trauma they have experienced may vary

This was the conclusion reached by a team from King’s College London. The researchers analysed data from nearly 40,000 people across 49 studies. On average, around two and a half years elapsed between the two assessments.

The study has been published in the journal *Nature Mental Health*.

Details

The authors did not simply study ‘childhood trauma’ in the broad sense, but rather memories of childhood maltreatment — abuse and neglect during childhood. This includes emotional, physical or sexual abuse, as well as situations where a child does not receive the necessary care, protection or attention.

The researchers were interested in how consistently people recount these experiences over time. For example, does a person report the same details several months or years later as they did previously, or does their account change?

To investigate this, the researchers conducted a systematic review and meta-analysis. This is a type of research in which, rather than recruiting a new group of participants, the results of previously published studies are collected and statistically combined. The final analysis included 49 studies and 38,332 individuals who were reassessed. The average interval between assessments was 2.4 years, but across the different studies it ranged from 2 months to 12 years.

Overall, self-reports of past abuse proved to be fairly stable. The article notes a high overall stability — r = 0.79.

In simple terms, this means that if a person had previously reported such an experience, their subsequent responses often matched their earlier ones, although not always completely.

However, an important detail is that consistency was lower among children than among adults. Among adults, accounts of childhood experiences remained fairly consistent regardless of how much time had elapsed between the two assessments. Among children, however, memories changed more frequently, and the longer the interval between assessments, the less consistent the responses became.

Why does this happen?

The authors attribute this to the development of memory. A child’s memory, their ability to link events into a personal narrative, and their understanding of exactly what happened to them are still developing. What seemed ‘normal’ to a young child may later be recognised as neglect or abuse. Or, conversely, a child may not immediately find the words to describe what they have experienced.

The study also showed that roughly one in five participants changed their answers between assessments. The authors emphasise that this should not automatically be taken as a sign of dishonesty. Such changes may reflect the development of memory, a new perspective on what happened, or a shift in their willingness to talk about their experiences.

Another important finding concerns neglect. Accounts of neglect proved to be less consistent than those of abuse. This can be explained by the fact that neglect is more difficult to ‘pin down’ to a single specific event. Violence is often linked to a single incident, whereas neglect is a lack of care that a child may only become aware of later, once they understand what normal support and safety entail.

What this means in simple terms

Memory is not a video recording. This is particularly true for children. A child may initially recount only part of what happened, then add details, change their wording or explain their feelings differently. This does not automatically make their account untrue.

Context is key: who is asking, how safe does the child feel to speak, do they understand the adult’s words, are they afraid of the consequences, do they feel guilt or shame, and have they received support? All of this can influence exactly how they describe their traumatic experience.

That said, the research does not claim that any memories are absolutely accurate. It suggests something else: accounts of childhood abuse are generally quite consistent, but they cannot be assessed simplistically — ‘if it matches, it’s true; if it has changed, it’s a lie’. Such logic is particularly dangerous when it comes to children.

Why this is important

The findings are important for psychologists, doctors, social workers, investigators, the courts and anyone who works with children. If a child’s account changes, this should not automatically be taken as evidence of deception. A careful, professional assessment is required: the child’s age, the nature of their experience, the circumstances of the conversation, the level of safety, and the way the questions were asked.

This is particularly important in legal and social practice. Misinterpretation can be harmful to both sides: either the child will not be believed or protected, or adults will draw hasty conclusions without sufficient verification. The research in fact shows that accounts of past experiences require careful, non-accusatory and well-prepared work.

There is also a therapeutic implication. The authors suggest that there may be a period in childhood when traumatic memories and the associated evaluations are more malleable. This does not mean that therapy should ‘rewrite memory’. The point is rather this: early professional support can help a child process their experiences more safely and reduce long-term consequences.

This is not a medical recommendation, nor is it advice on how to deal with trauma on your own. If a child has experienced abuse, neglect or another traumatic experience, help must be provided by a specialist trained in working with childhood trauma.

Background

There has long been debate surrounding memories of childhood abuse and neglect. On the one hand, such memories are frequently used in research, clinical practice and the courts. On the other hand, they are often considered unreliable because they are based on personal accounts, memory and a subjective understanding of events.

New research clarifies this picture. It shows that self-reports of abuse are not ‘chaotic’ or inherently unreliable. On average, they are fairly stable. But this stability is not the same for everyone: it is lower in children than in adults; memories of neglect change more frequently than memories of abuse; and a long interval between assessments can particularly affect children’s responses.

This helps to avoid two extremes. The first extreme is to assume that any change in the account undermines trust in the child. The second is to assume that memories are always fixed and do not require verification. The research supports a more nuanced approach: children’s accounts should be taken seriously, but analysed professionally and with due regard for the child’s developmental stage.

Source

Study: Oonagh Coleman, Maya Al-Jaber, Geethma Aponsu, Daniel Stahl, Andrea Danese and colleagues, “Stability of childhood maltreatment self-reports: a systematic review and meta-analysis”, Nature Mental Health, 2026.