Retraining Working Memory to Reduce PTSD Symptoms

Using working memory allows for the daily function of learning, reasoning, and comprehension, but its role in managing anxiety can pose difficulties in patients with PTSD.


Working memory (WM)—the function that allows us to temporarily store information and use it for cognitive tasks like learning, reasoning, and comprehension—is thought to play an important role in managing anxiety and intrusive symptoms of posttraumatic stress disorder (PTSD). Researchers have found inefficient filtering of threatening material from WM and increased storage of task-irrelevant threat distractors are associated with elevated anxiety. Thus, veterans with high levels of anxiety may disproportionately allocate cognitive resources toward threatening stimuli, and have trouble keeping threatening thoughts from entering WM and staying there. People with PTSD also have been shown to have problems using WM in emotional contexts, making it hard for them to deal with stressful situations in work and personal life. Some research has suggested that WM training that incorporates affective stimuli may increase the ability to use WM in emotional contexts. In a pilot study, researchers from Medical College of Wisconsin and University of Wisconsin-Milwaukee expanded on that possibility by testing 2 kinds of WM training in 21 veterans with elevated PTSD symptoms.


Participants completed a pretest, 15 sessions of at-home computerized training, a posttest, and a 1-month follow-up session. The computerized tests included tasks to measure WM capacity, such as solving math questions while trying to remember a set of unrelated numbers. The participants were then assigned to active emotional WM training ( n-back) or control emotional WM training (1-back). The n-back training involved constant updating of information stored in WM and shifting between visual and auditory stimuli using 8 different faces and 8 different negative words. The training sessions started at the 1-back level and increased in difficulty level by level with performance with > 95% accuracy, or were scaled back with performance with < 75% accuracy. The control training was the same but consisted only of 1 level.

Overall, the researchers say, contrary to their hypothesis, they did not find a significant difference in results from the 2 groups. But they did find that both trainings had an overall “significant and sizable” impact on PTSD symptoms in both groups: 73% of the n-back group and 60% of the 1-back group had a reduction of ≥ 10 points on the PTSD Checklist total scores, which the researchers say is considered a clinically meaningful change. Moreover, both groups showed a clinically meaningful reduction in symptoms at follow-up: 55% and 40%, respectively.

It was interesting, the researchers say, that both groups showed improvement, especially with the 1-back intervention, which they had used as a “minimally effective” control condition. They also note that, anecdotally, the participants found both interventions “quite challenging.” It may be that in this population even the 1-back intervention is enough to detectably reduce symptoms, the researchers say. They were also encouraged to see that the n-back condition (marginally) outperformed the 1-back condition in improving reexperiencing symptoms, theoretically the most relevant training target of WM-focused intervention. The researchers believe that their study yields useful pilot data, suggesting that n-back training can have a clinical impact primarily by reducing reexperiencing symptoms, which are highly likely to indicate the presence of impaired WM functioning.

Next Article: