Recent elegant research has raised the salient issue of altering traumatic memories and its treatment implications. Kindt et al ^{Reference Kindt, Soeter and Vervliet1} suggest that ‘if emotional memory could be weakened or even erased, then we might be able to eliminate the root of many psychiatric disorders, such as post-traumatic stress disorder’. In a similar vein, Schiller et al ^{Reference Schiller, Monfils, Raio, Johnson, LeDoux and Phelps2} reported that ‘fearful memories can be wiped out for at least a year using a drug-free technique’. The prospect of erasing distressing memories is indeed compelling and has led to widespread media coverage.

However, this issue elicits important ethical and clinical considerations: first, would we want to erase trauma memories, and second, is it clinically helpful to erase such memories?

Loss of knowledge about the past or oneself may be ethically problematic, although reducing suffering clearly may take precedence. ^{Reference Liao and Sandberg3} Our sense of self is constructed from autobiographical memories, and the authenticity of how they link and our trust in this narrative is important for well-being. Furthermore, losing memory can compromise a victim's ability to provide legal evidence: autonomy and beneficence may trump justice, but it would be better if the evidence could be used and the victim did not suffer.

Paradoxically, erasing memories of trauma may not in itself reduce suffering and could even lead to the reverse. In clinical cases where explicit memory of an event has been lost, for example owing to a severe head injury or drug rape (e.g. via flunitrazepam), extreme distress can ensue. The clinical literature suggests that avoidance of trauma memories is associated with worse rather than improved outcome.

We note that the data in the above papers do not in fact indicate memory ‘erasure’. Rather, both studies found that fear responding (the emotional component of the memory) was reduced while declarative memory (knowledge about the event) was left intact. The data therefore point towards erasing the pain but not the knowledge of the trauma memory: an important ethical and clinical difference.

We need to challenge the erroneous public perception of a science seeking to ‘erase’ painful memories as such media headlines are obscuring the true interpretation of the data and what treatment development seeks. Such consideration may help prevent us from inadvertently misleading people (especially those who have suffered trauma) to believe that we are pursuing an ‘eternal sunshine of the spotless mind’.