A newly developed experimental software system can help humans make decisions more accurately and quickly in ultra stressful situations such as battlefields, US scientists have claimed.

Researchers at Penn State University said that their R-CAST software has proven its worth in a military command-and-control simulation which involved intelligence gathering, logistics and force protection.

When time pressures were normal the human teams functioned well, sharing information and making correct decisions about the potential threat, according to the researchers.

However, when the time pressure increased, the teams' performances were found to suffer. Because there was no time to share information, the teams made incorrect decisions about whether to avoid or attack the incoming aircraft.

"This is the first test of the R-CAST architecture and it shows that software agents can play an essential role in helping human partners make the right decision at the right time," said Xiaocong Fan, a post-doctoral scholar at Penn State's School of Information Sciences and Technology and lead developer of the software.

In the simulation, team members had to protect an airbase and supply route which were under attack by enemy aircraft. The scenarios were configured with different patterns of attack and at different tempos.

The situation was complicated because team members had to determine whether the aircraft were neutral or hostile. Furthermore, two team members were dependent on the third whose role was to gather information and communicate it to them.

Shuang Sun, a doctoral student in information sciences and technology, and co-author of the project, explained: "When the teams don't know if the incoming aircraft is the enemy the defense team can't attack, and the supply team takes action to avoid the incoming threat which causes a delay in delivery. These decisions lower the performance of the whole team."

When the information gatherer was supported by the R-CAST software, the information was gathered and shared more quickly. As a result, the human teams were better able to defend themselves from enemy attack and deliver supplies without delay.

The researchers also learned that in contrast to human teams whose performance suffers from increased tempos, the software enables human teams to better maintain their performance at an acceptable level.

The software is based on the recognition-primed decision model which posits that people make decisions based on their recognition of similarities between past experiences and current situations.

Earlier research showed that this model addresses situations in which there is little or no time for extensive reasoning.