In the letter by Harvey1 on page 562 of this issue, a personal instance of memory loss for the events of a life threatening accident is described. In the absence of any concussion or violent acceleration/deceleration forces, Harvey1 concludes that this was “post-traumatic amnesia in which the trauma was wholly psychological.” There are at least two interesting aspects to this account—one being what was remembered, and the other being what was not recalled.

In situations of extreme stress, some memories are enhanced, detailed, and may be recalled intrusively thereafter, whereas other items are forgotten. In the present account, Harvey notes that the front seat passenger was wearing a flowery hat, that the car was a small red Honda saloon, and that a black and white soft toy was dangling in the rear window, but he has forgotten what happened next. Fragments of vivid memories such as these seem to be common in life-threatening trauma, and they become intrusive in post-traumatic stress disorder. Similarly, in head injury, Russell and Nathan2 referred to memories with the quality of “visions” arising from a brief “lucid interval” before the onset of post-traumatic amnesia—for example, for the screech of brakes or being struck by a car. In head injury, amnesia predominates and such vivid memories are infrequent, whereas in post-traumatic stress disorder intrusive thoughts predominate and memory lapses are less common: nevertheless, the two seem to lie at the extremes of a continuum, and what is still poorly understood in both instances is why certain things are vividly remembered, whereas others are forgotten.

Memory loss after trauma is well described in the psychiatric and clinical psychology literature. It can be global, as in fugue states, or situation specific for a particular traumatic incident.3 If situation specific, the amnesia can be complete or partial (‘fragmentary’). Fugue states are always precipitated by situations of severe stress, such as occur during marital or relationship breakdown, severe financial troubles, wartime, or being charged with an offence. They are very commonly associated with severe depression, and there is commonly a history of a transient, organic amnesia, which may have acted as a kind of “learning experience”. Situation specific amnesia occurs in so called “crimes of passion”, where the offence takes place in a state of extreme emotional arousal, is unpremeditated, and where the victim is (almost invariably) a lover, wife, or partner. Although it can be argued that such amnesia may be legally motivated, memory lapses are also reported in the victims and eye witnesses of offences.4 5 Amnesic gaps have also been reported in traumatised soldiers in the two world wars6 and subsequently.7 A recent review8 documents evidence of amnesia in the victims of lightning flashes, flood disasters, pipeline explosions, earthquake, concentration camp and holocaust survivors, and Bosnian refugees: these authors report amnesia or “memory disturbance” in 16 studies. Others have also cited cases of kidnap and torture,7 and I would add victims of the Herald of Free Enterprise disaster. In addition, Brown et al 8 found some evidence of forgetting in all 68 studies that they reviewed on the fraught issue of memory for child sexual abuse. Although there are problems in evaluating self reports of amnesia for child abuse, some smaller scale studies have attempted to examine the evidence for both the trauma and the subsequent forgetting in some detail.9Curiously unmentioned in this list are road traffic accidents, presumably because any memory loss is generally assumed to be organic: nevertheless, attempts to examine the interaction of neurological and emotional effects on memory after head injury are beginning.10

Many of the situations cited involved fear or threat to life. It has been claimed that such situations involve a narrowing of consciousness with attention focused on central perceptual details, sometimes evolving into amnesia5 or that emotional or traumatic events are processed differently from “ordinary” memories.7 In particular, emotional memories may implicate amygdaloid circuits.11 12 It may also be the case that, when something extraordinary happens, we ask ourselves to recall far more detail than we would normally expect. The mechanisms involved are not well researched, but Harvey's1 account will be invaluable if it restores attention to the putative contribution of emotion in at least some cases of accident or trauma.