Sue Miller's protagonist in the novel For Love returns home, only to feel "mugged by memory." Clinicians working with traumatic memories may feel mugged nowadays. After reading hundreds of studies on emotional memory, I find a perspective emerging. Those who seek to find in the literature evidence of memory's photographic perfection conclude that the glass is empty. Those who have stopped magnifying every memory flaw to disprove this extreme position have found substantial evidence of the tenacity of memory.

Prior to the 1980s, 90 percent of the literature on eyewitness testimony was laboratory simulation, which has limited application to real life. It is impossible to recreate in a lab, within ethical guidelines, an experience that is as rich, anxiety-provoking, personally affecting and consequential as a real-life trauma. While field studies overcome many of these disadvantages, little of what has been learned from them has surfaced in the memory debate.

The following principles are demonstrated in field studies: First, most of the details that crime witnesses tell police are correct and remain correct for several months. Second, investigators' attempts to suggest inaccurate details do not intrude into recall, nor does wrong information published in the media appear in witness statements. Third, among respondents recalling traumas suffered in concentration camps 40 years earlier, the accuracy of memory for conditions, mistreatment, daily routine, housing, food and main guards was remarkable.

Recently, I was the clinical discussant of a presentation by Elizabeth Loftus, Ph.D., in Washington, D.C. I found myself in an odd position. Her slick address was peppered with legal anecdotes and case material from newspapers and magazines. Anyone who works clinically with trauma survivors could go head-to-head at that level of evidence, but even if we wanted to, our professional audiences would not let us get away with it. And so I responded with conclusions from cognitive literature and empirical data from trauma survivors, most of which appeared in the October 1995 special issue of Journal of Traumatic Stress. Some of the audience comments referenced the problem that clinicians have not emphasized third-party validation of memories.

Certainly, any applied profession has a practice forefront that is as yet unvalidated, but I think that the field of trauma studies has amassed a very impressive body of literature in its short history. And we can continue to do more. Intrusive memory is one of the hallmarks of PTSD; however, we have focused too exclusively on PTSD as an endpoint to be predicted by personal characteristics, or to be evaluated with respect to the outcome of interventions. Descriptive data on the phenomena of intrusion are lacking, and we have not paid sufficient attention to whether our memory-focused interventions actually result in measurable changes in intrusive memory, or whether they are effective by acting via reduction of other symptoms of PTSD.

But overall, we are participating in the dialogue about memory productively, marshalling data to rebut fundamental criticisms. Loftus recently resigned from the American Psychological Association, writing a letter which attempts to further polarize and politicize the issues by dismissing us as unscientific players with whom she can no longer associate. No. The future belongs to those clinicians and cognitive scientists who come together to share technical vocabulary, theoretical formulations, objective approaches to measuring significant memory phenomena and access to ethnically and developmentally diverse populations experiencing real trauma.
For a detailed review, see Koss, M.P., Tromp, S. and Tharan, M. (1995). Traumatic memories: empirical foundations, forensic and clinical implications. Clin. Psychol.: Sci. and Pract., 1:111-132.