towards a sustainable infrastructure

Jump to Menu

Does combat stress have a long-term effect on attention and memory?

How grid computing is helping to understand the effects of combat on soldiers’ brains.

Veterans returning from active duty endure many challenges to readapt to a civilian life. For some there is the added complication of combat stress, which affects their memory, attention span and other cognitive functions. But how? And for how long?

Neuroscientist Guido van Wingen and colleagues at the Academic Medical Centre (AMC) in Amsterdam monitored a group of soldiers from before their first deployment to Afghanistan until 18 months after their return to civilian life. The idea was to look how combat stress affects brain areas supporting cognitive functions such as memory and attention.

The team used the grid-enabled e-bioinfra science gateway to process and analyse 118 brain scans from 33 soldiers and 26 civilians used as controls. Thanks to a friendly user interface and the computing power of the grid, a workload of several weeks was condensed into two days.

The conclusions, published in the Proceedings of the National Academy of Sciences, show that combat stress impairs cognition by affecting the midbrain and its link with the prefrontal cortex, and that this is largely reversible but could have an impact on future social and cognitive functions.

Troops deployed in Afghanistan

Troops deployed to the Kunar province, Afghanistan in September 2010.

Photo by Staff Sgt. Gary A. Witte. Source: wikicommons

Monitoring the troops

Severe stress in war zones can lead to long-lasting problems in combat troops. The diagnosis of “shell shock” used during the First World War has come a long way and post-traumatic stress disorder is now recognised as a major issue with social implications. But even those fortunate not to experience trauma report memory loss and decrease of attention.

Guido van Wingen and his colleagues at the Academic Medical Centre (AMC) in Amsterdam wanted to know how stress affects the areas of the brain that support cognitive functions. To do this, they tested 33 soldiers before they were deployed to Afghanistan.

The volunteers’ brains were scanned with functional magnetic resonance (fMRI) and diffusion tensor (DTI) imaging techniques. The soldiers also completed cognitive memory and attention tests and filled questionnaires to assess psychological condition. As control, the team also scanned and tested 26 individuals in non-combat roles.

While in Afghanistan the soldiers were embedded as part of a NATO mission and exposed to combat situations, enemy fire and the risk of improvised explosive devices. The tests were repeated shortly after their return from deployment and then again 18 months later.

In total, Guido and his colleagues had 118 scans to analyse, which would take approximately 240 hours of computing time. This is not a lot, but “the logistics to organize so many datasets and process them is complex when using a regular computer,” he said.

To make things easier, Guido used the e-bioinfra gateway to access SURFsara's Dutch e-science grid and analyse the DTI scans with software developed by the AMC’s e-Bioscience group led by Silvia Olabarriaga.

“Silvia’s group has created an incredibly user friendly interface to perform the most difficult analyses with several mouse clicks to upload the data and run the analyses,” Guido says. “And with the computing power of the grid the analyses that would otherwise require several weeks’ time are finished over the weekend.”

Long and short-term consequences

From the psychological evaluation, the team found that combat stress interfered with the soldiers’ capacity for sustained attention shortly after their return from Afghanistan. The fMRI scans of their brains during a memory test revealed that this effect was caused by functional changes in the midbrain. The analysis of the DTI images showed that these were accompanied by structural alterations as well.

In the long run the soldiers returned to their baseline value for attention, which means that the influence of stress on the midbrain’s function and structure is reversible.

The team also found long-term changes in the way the midbrain couples with the prefrontal cortex. “Midbrain-prefrontal cortex coupling is an index for communication between these brain regions,” explains Guido. “This is known to be important for maintaining information in the working memory, for its evaluation and manipulation in order to make appropriate decisions.”

The study, published in the Proceedings of the National Academy of Sciences, shows that while the effects on attention are reversible, combat-stress compromises the link between the midbrain and the prefrontal cortex. This may increase the soldiers’ vulnerability to stress and have an impact on future social and cognitive functions.

Silvia’s group has created an incredibly user friendly interface to perform the most difficult analyses with several mouse clicks to upload the data and run the analyses.

Guido van Wingen

More information

Guido van Wingen’s research pages

The e-Bioscience group is part of the Department of Epidemiology, Bioinformatics and Biostatistics of the AMC


van Wingen et al. 2012 PNAS 109 (38):15508-15513 doi:10.1073/pnas.1206330109 (full text)

Technical details

The DTI data analysis software was developed at the AMC and integrated into the AMC e-bioinfra science gateway described in Shahand et al. 2012. The data was first uploaded by the user to an ftp site, and then transported automatically to the grid storage and back.

The computations were managed by the MOTEUR workflow management system. When the results were ready, they were transferred back to the ftp server automatically.

A robot certificate of the VLEMED VO was used to access the grid resources. Large computation and storage capacity available for this study allowed obtaining results faster and more easily.

Shahand et al. 2012 J Grid Computing 10, 4:725-742. doi: 10.1007/s10723-012-9233-4 (abstract)