🚧 Website Maintenance in Progress: Thank you for visiting! We are currently in the process of enhancing our website to serve you better. Please check back soon for our new and improved website.

In my research and clinical practice with PTSD patients I have been particularly fascinated by the phenomenon of traumatic flashbacks, which appear to be one of the few symptoms that are unique to the disorder.

In the early days of PTSD most traumatologists considered intrusive memories of specific stressful events to be a distinctive symptom but we now know they are found in most forms of psychopathology (Brewin, Gregory, Lipton, & Burgess, 2010). Instead, evidence is gradually accumulating that it is the reliving in the present that distinguishes intrusive memories in PTSD from those in other disorders (Brewin, 2013; Brewin et al., 2010). DSM-5 has already taken the important step of distinguishing intrusive memories more clearly from intrusive rumination which, although a common feature of PTSD, does not now contribute to its diagnosis.

An even more specific focus on intrusive memories that are relived in the present is part of the proposed changes to the PTSD diagnosis in ICD-11, which focus on identifying the six symptoms that it is thought are the most reliable way of discriminating PTSD from other disorders (Maercker et al., 2013). Preliminary evidence suggests that if the ICD-11 proposals were implemented PTSD prevalence would remain largely unchanged but levels of comorbidity with depression would be lower than they were under DSM-IV (Morina et al., in press).

Curiously, given their status as one of the few distinctive markers of traumatization, there has been relatively little attention paid to defining flashbacks. There has been uncertainty about whether the term should be reserved for extreme dissociative episodes in which individuals completely lose contact with their surroundings for periods of minutes or more, or whether they should include all intrusive memories that are accompanied by a sense of reliving the event in the present, even if  only fleeting.

Both DSM-5 and the proposed ICD-11 have now opted for the more inclusive definition in which flashbacks are seen as existing on a continuum between these two extremes.This clarification is likely to be very helpful clinically. It should facilitate new questions, for example whether exposure methods are equally useful for intrusive memories that are and are not relived in the present, and to what extent those with intrusive memories that are not relived in the present may be better diagnosed as having other disorders such as depression.

Along with this should come more systematic assessment of flashbacks. As they are typically a response to an internal or external cue, questionnaire items about their frequency in the past week or month may not be the most accurate method of assessing them. We also know little about what patients understand by typical questions such as “Have you ever suddenly acted or felt as if the event(s) was happening again?" and whether different forms of wording would elicit more accurate answers.

One possibility is to ask a more specific question about whether individuals would experience flashbacks if they allowed themselves to fully remember the event or confront reminders of it. Alternatively, people could be asked within the context of a structured interview to recall the most difficult moments of their trauma and then describe their response. Although more time-consuming, responses are likely to be more ecologically valid and consistent with what would be seen during the course of psychological therapy.

The scientific investigation of flashbacks will benefit from theories that, like our revised dual representation model (Brewin et al., 2010), make specific predictions about the neural basis of this type of memory. The utility of such models is likely to depend on the success with which they can integrate evidence from conditioning and human autobiographical memory studies with clinical observations.

There is now a considerable quantity of evidence supporting the idea that flashbacks depend on an involuntary perceptual memory system that is distinct from ordinary episodic memory (Brewin, 2014), and this perceptual memory system may largely bypass medial temporal lobe structures that ordinarily provide memories with their spatial and temporal context. This suggests that imaging studies can benefit from investigating what characterizes PTSD, not just at the diagnostic level but at the level of individual symptoms.

For example, an initial study has found that patients reporting more flashbacks have reduced brain volume in a theoretically predicted area, inferior temporal cortex (Kroes et al., 2011). Imaging studies are also likely to benefit from designs that allow the experimenter to directly compare trauma memories that are and are not relived in the present. Using such a design we have provided preliminary evidence consistent with the dual representation model (Whalley et al., 2013).

That PTSD exists and deserves its place in the diagnostic manuals is now beyond question. But PTSD is still, compared to other disorders, excessively complex, and this very complexity is likely to be holding back clinical and scientific advances. Some positive developments include the definition of a dissociative subtype (in DSM-5) and complex PTSD (in the proposed ICD-11). I believe the field is now ready to ask more focused questions about the existence of specific phenotypes and how they can best be addressed in diagnosis and treatment.

The study of flashbacks and other symptoms that are distinct to trauma-related disorders seems like a good place to start.

About the Author

Chris R. Brewin, AcSS FBA, is a professor of clinical, educational & health psychology at University College London and is the recipient of the 2013 ISTSS Robert S. Laufer, PhD, Memorial Award for Outstanding Scientific Achievement. The award was established by Ellen Frey-Wouters, PhD, in memory of her husband, Robert S. Laufer, and acknowledges an individual or group for their outstanding contribution to research in the field of traumatic stress.


Brewin, C.R. (2013). “I wouldn’t start from here”:  An alternative perspective on PTSD from ICD-11. Journal of Traumatic Stress, 26, 557–559.

Brewin, C.R. (2014). Episodic memory, perceptual memory, and their interaction: Foundations for a theory of posttraumatic stress disorder. Psychological Bulletin, 140, 69-97.

Brewin, C.R., Gregory, J.D., Lipton, M. & Burgess, N. (2010). Intrusive images and memories in psychological disorders: Characteristics, neural basis, and treatment implications. Psychological Review, 117, 210-232.

Kroes, M.C.W., Whalley, M.G., Rugg, M.D., & Brewin, C.R. (2011). Association of flashbacks and structural brain abnormalities in posttraumatic stress disorder. European Psychiatry, 26, 525-531.

Maercker, A., Brewin, C.R., Bryant, R.A., Cloitre, M., Humayan, A., Jones, L., et al. (2013). Diagnosis and classification of disorders specifically associated with stress: New proposals for ICD-11. World Psychiatry, 12, 198–206.

Morina, N., van Emmerik, A., Andrews, B., & Brewin, C. R. (in press). Comparison of DSM-IV and proposed ICD-11 formulations of post-traumatic stress disorder. Journal of Traumatic Stress.

Whalley, M.G., Kroes, M.C.W., Huntley, Z., Rugg, M.D., Davis, S.W. & Brewin, C.R. (2013). An fMRI investigation of posttraumatic flashbacks. Brain and Cognition, 81, 151–159.