Although we have successful treatments for full-blown PTSD, crisis interventions to reduce the build up of symptoms in the early aftermath of trauma are lacking. Current interest in the manipulation of memories post-trauma is particularly focused on pharmacological means, for example, propranolol administration [4]. In addition to the potential for side-effects with pharmacological approaches, there are potential ethical concerns if voluntary memories for human experience are suppressed [5]. For example, removing flashbacks at the expense of being able to deliberately remember what happened during a trauma could compromise a trauma victim's ability to testify in court. We have also raised clinical concern over treatment innovations stemming from exciting theoretical developments [6], [7] but which advocate psychological approaches which promote the suppression of memory for traumatic experiences as way of dealing with negative sequelae, since suppression is clinically contra-indicated [8], [9].

The psychological intervention with the strongest evidence-base for full-blown PTSD is trauma-focussed Cognitive Behaviour Therapy–a treatment which is only indicated when delivered weeks or months after the trauma [10]. However, what can be given to trauma victims suffering flashbacks in the first few weeks? Unfortunately, using talking therapy as a crisis intervention in the immediate aftermath of trauma has caused international clinical concern [11]: interventions such as critical incidence stress debriefing can worsen rather than ameliorate later trauma symptoms [12]. Given the scale of traumatic events globally–war, terrorism, natural disasters, interpersonal violence–there is a huge unmet need for widely-available and easily accessible interventions. We need to develop fresh theory-driven interventions to reduce the build up of flashbacks in the early post-trauma period.

Can cognitive science suggest a way to reduce the build-up of flashbacks to trauma? We suggest that certain cognitive tasks, informed by the neuropsychological domain of working memory, may indeed be used tap into processes underlying flashback memory consolidation. This is based on two key findings: first, cognitive science has shown the brain has selective resources with limited capacity [13]; second, the neurobiology of memory suggests there is a 6-hr window to disrupt memory consolidation [14], [15]. In particular, we suggest that visuospatial tasks will be useful in this regard according to the following rationale: (1) trauma flashbacks are sensory-perceptual images with visuospatial components [2], [16], [17]. (2) visuospatial cognitive tasks compete for resources with visuospatial images [18]–[21]. (3) the neurobiology of memory consolidation suggests a 6-hr time frame post-event within which memories are malleable [14], [15]. Thus (4) we predict visuospatial cognitive tasks given within 6-hr post-trauma will reduce flashbacks.

Following a long tradition in experimental psychopathology, we use the trauma film paradigm [22] in our laboratory as an experimental analog of viewing real trauma and the subsequent flashbacks [23], [24] suffered in PTSD. We have previously demonstrated that the frequency of flashbacks for analog traumatic content can be manipulated experimentally by completing standardized cognitive tasks during trauma film viewing (i.e. peri-traumatically): In healthy participants, completing a complex visuospatial working memory task [19], such as finger pattern tapping, during exposure to a trauma film subsequently reduced flashbacks of that film over 1-week compared to no task, or non-visuospatial tapping [24], [25]. Interestingly, and in line with cognitive theories of PTSD [2], verbal cognitive tasks during the film, such as counting backwards in threes, actually increased the number of flashbacks, confirming that the effects were not simply due to distraction for general working memory resources but to the specific nature of the task [24].

In the current study it was critical to identify a visuospatial working memory task that would be widely-available in the real-world. Clearly a standard neuropsychological test battery visuospatial task such as the WAIS block design [26] would be impractical to deliver en mass. The popular computer game “Tetris” has been demonstrated to be a visuospatial task [27]–[30], drawing on mental rotation and the type of processing we recruit when forming mental images. Playing “Tetris” can even result in participants experiencing subsequent visual images of the game itself at a later time [30], implicating its involvement with intrusive, image-based memory. The capacity of visual memory is both limited and vulnerable to proactive interference, i.e. interruption of memory for presented stimuli by the presentation of similar, but different stimuli after a time delay [31]. Thus “Tetris” provides a promising candidate. Interestingly, watching violent films and playing prolonged computer games are typically associated with negative impact on psychological well being and behavior in both children and adults [32]. This has led to public concern over their ready availability. However, clearly not all computer games are bad for you [27].

We now report the impact of a cognitive visuospatial task, a computerized mental rotation game, on the modulation of analog flashbacks to trauma. Given our earlier experimental demonstration that engaging in visuospatial tasks peri-traumatically (i.e. during trauma film viewing) reduces subsequent intrusive imagery (analog flashbacks), we tested here whether the window for intervention could be extended into the post-trauma period (i.e. after trauma film viewing). This is a critical question since in real-world trauma successful manipulation of flashbacks would need to be conducted post-event rather than peri-traumatically. Further, investigating a time-frame for intervention in the near aftermath of trauma, yet within a window of memory malleability has clear clinical implications. Recent statistics indicate that the average waiting time in an emergency department in the United States is 30-mins [33]. To test the real-world application of our paradigm, we explored the manipulation of flashbacks in the laboratory 30-min after watching a trauma film.

Our suggestion to manipulate flashbacks in the aftermath of a traumatic experience is supported by various experimental demonstrations that newly formed memories are initially labile for a short time and thus subject to interference. In rats, the consolidation of fear memories can be inhibited in a dose dependent fashion with administration of either anisomycin or Rp-cAMPS in the immediate post-training phase [34]. Furthermore, even established memories may be subject to manipulation upon reactivation–the triggering of previously consolidated memories may return them to a labile state and in need of reconsolidation if they are to persist [14], [35], [36]. In humans, the consolidation or stabilization of motor memory (in this case, a finger pattern tapping task) has been demonstrated to occur in the first 6-hours following initial learning: learning a new variation on the motor task after reactivation of the previously learned motor task can block the reconsolidation of memory for the original task [15]. Although there are doubts as to whether such reconsolidation may occur for all types of memory [37], of particular interest to PTSD research is that flashbacks for trauma may be pharmacologically modulated [4], [38], though this has not been without controversy [5]. Given the potential advantages of a non-invasive procedure, we were interested in testing whether a cognitive intervention-a visuospatial task-could modulate later flashbacks to traumatic material.

Following the 30-min interval period in which participants completed standardized filler tasks, a brief reminder task for the trauma film was administered to both groups. Participants then either completed the visuospatial condition or sat quietly (no-task control condition) for 10-min. During these 10-min, all participants recorded the frequency of trauma film flashbacks they experienced. Significantly fewer flashbacks were experienced in this initial period (i.e. while playing the game or not) during the visuospatial condition than the no-task condition (Table 3).

After leaving the laboratory, participants then kept a daily diary in which they recorded their flashbacks to the trauma film over a period of 1-week. Crucially, we found that participants in the visuospatial condition experienced significantly fewer flashbacks over the week (Figure 2) than those in the control condition. Furthermore, at 1-week, participants returned to the laboratory–participants in the game condition had significantly lower scores on the measure of clinical symptomatology of trauma–the Impact of Events Scale [40] (Figure 3).

On a recognition memory test for the trauma film given after 1-week, performance was comparable in the visuospatial and no-task conditions (Table 4), indicating that voluntary memory for the film appeared intact. That is, completing the visuospatial condition had served to reduce only the trauma flashbacks typically associated with psychological distress and dysfunction, but not the actual memories for the events which could still be deliberately recalled.

When viewing traumatic film stimuli, the type of sensory-perceptual focus engaged in is predominantly visual (rather than say olfactory or auditory). To interfere with this type of visual processing we need to target what are known cognitively as ‘visuospatial’ resources [19]. That is, visuospatial tasks that use the same type of processing as do visual flashbacks will interrupt memory consolidation of those flashbacks by competition for the same limited cognitive resources [24]. Thus, by selectively interfering with visual sensory-perceptual processing of the traumatic film via visuospatially demanding cognitive tasks, subsequent analog flashbacks are reduced. Note, this is not the same as simple distraction, since other types of tasks such as verbal tasks during traumatic films are predicted (and have been shown) to lead to increased flashbacks [23], [24]. Our data is the first indication that the manipulation of visuospatial processing in the consolidation phase of recently activated trauma memories can serve to modulate future intrusive, involuntary flashbacks (despite leaving voluntary memory intact). “Tetris” participants experience fewer intrusions even while playing the game, supporting the competition for resources rationale. Significantly, we demonstrate that the visuospatial task conducted 30-min post-exposure to traumatic stimuli is effective in reducing flashbacks of that trauma as well as associated clinical symptomatology over 1-week.

Interestingly, the clinical literature offers potentially converging support for these findings. Eye Movement Densensitization and Reprocessing (EMDR) is an empirically-supported treatment for established PTSD [10]. During this therapy, the patient undergoes a series of eye movements whilst holding an intrusive traumatic memory in mind, leading to a reduction in the emotionality and vividness of the unpleasant mental image. One of several possible accounts of how EMDR might work is that the eye movements draw on visuospatial processing [18], [46], [47] and thus provide a dual task competing specifically for resources with the trauma imagery, reducing its impact. Our proposal similarly draws on a working memory rationale for how “Tetris” may affect flashback formation in a modality specific manner [48], though emphasizes all the senses involved in imagery (not just the visual modality). However, a critical difference between the current experiment and EMDR is that EMDR is used for treating existing flashbacks in PTSD (at least one month post-trauma), but is not intended to be used during the memory consolidation phase targeted in the current study. Our interest in the immediate aftermath of trauma is to understand preventative (rather than just curative) measures to the development of PTSD flashbacks.

Speculatively, the effects of “Tetris” may not be limited to the immediate post-trauma period during which it is played but may even continue to compete for visuospatial resources later. For example, it has already been demonstrated [30] that images of “Tetris” can intrude during sleep-a period during which memory consolidation is known to occur. Future research is required to examine the precise mechanisms of action by which “Tetris” reduces flashbacks to trauma. We predict that a verbal task would not have comparable effects and may even worsen flashbacks. Thus future studies should compare both a visuospatial task (e.g. “Tetris”) with a verbal task (e.g. a verbal computer game) against a no-task control group.

According to randomization to condition (simple, computer-generated random allocation), participants either completed the visuospatial condition or were in a no-task control condition for 10-min. Participants in the visuospatial condition played the game (“Tetris”) on a computer and used the cursor keys to move and rotate falling blocks to complete the largest number of complete rows across the screen. In the no-task control condition, participants were asked to sit quietly for 10-min. During both conditions participants recorded initial flashbacks of the trauma over the 10-min.