Research

Red Cell Lab offers realistic terrorism lessons for future intelligence analysts

In the Red Cell Lab, students battle biases to improve intelligence.

Students study maps in the Red Cell Analytics Lab at Penn State's Information Sciences and Technology Building on the University Park campus. Credit: Credit: Patrick Mansell / Penn StateCreative Commons

UNIVERSITY PARK, Pa. -- Whether intelligence analysts are trying to predict the next moves of an insurgent group or determining how to best deliver aid after a hurricane, an excess of information can often cause just as many problems as a lack of it.

Red Cell Analytics Lab, a laboratory in Penn State's College of Information Sciences and Technology, uses cutting-edge technology and the latest analysis to turn information into intelligence during fluid, complex situations that are as timely as today's headlines.

The name "Red Cell Lab" refers to teams of military personnel, often called "red cells," that are trained specifically to test the effectiveness of American military tactics. During the Cold War, for example, the U.S. Navy created the Top Gun school to train aviators against other American pilots who had been schooled to use the tactics and strategies of Soviet and Warsaw bloc pilots. The success of these efforts has continued with current programs that train groups of American soldiers to fight like Afghan and Iraqi insurgents; these groups then train fellow U.S. troops that will be deployed in those combat areas.

Similarly, Red Cell members at Penn State investigate threats and opportunities in a range of scenarios and test the effectiveness of possible responses. The Red Cell Lab has collaborated with the Marine Corps Warfighting Lab, the Office of the Secretary of Defense for Intelligence, the Penn State Office of Emergency Management, Boeing Corporation and the U.S. Army War College.

Col. Jake Graham, professor of practice in information sciences and technology, said that while the lab can easily adapt its research to military, counterterrorism and security missions, lab members are also using their expertise to analyze possible outcomes in emergency and natural disaster relief operations.

The lab features advanced technology for information visualization, including 3-D visual displays. Researchers can even print out three-dimensional models using a 3-D printer. This allows researchers to physically hold and examine models of data, rather than just observing the information on a video screen.

In one recent project, Red Cell students helped the military understand how insurgents might use unconventional tactics to defend against aircraft and helicopter attacks. Graham said Red Cell students are also devising a game that will analyze how communities will cooperate -- or compete -- for resources during natural disasters.

"Every professor tries to bring real problems and challenges to the classroom," he said. "What I try to do is bring those real problems to life."

Uncovering Biases

Researchers and students have collaborated with the staff of Raytheon Intelligence and Information Systems, a defense contractor, to design a series of games that helps detect judgment biases. Since the Sept. 11 terrorist attacks, Graham explained, the intelligence community has studied bias as a serious handicap in making decisions based on intelligence. For example, many critics blame the failure to detect warning signs that Al Qaeda was preparing to attack the United States on the biases of intelligence agents and flawed problem-solving methods -- heuristics.

"It isn't just the 9-11 attacks that we're studying," said Graham. "There were several national intelligence failures, including the determination that there were weapons of mass destruction in Iraq, that were fraught with heuristic errors and biases."

Another mission for the lab is to investigate how to best integrate human intelligence with computer-based intelligence, often referred to as hard and soft sensor analysis. As computer sensors, cameras and other intelligence-gathering devices become ubiquitous, enormous amounts of data are now available to analysts. However, Graham said the availability of data does not always translate to intelligence. The way people process information is distinct from how computers do it, he said.

For instance, a camera that shows a person walking down the street cannot easily ascertain the motive of that person. Human observers, on the other hand, may offer information that is clouded by their own biases and less concrete than that given by hard sensors. "For example, you may ask a person the distance of an object and they may tell you it was 'a ways away,' " Graham said. "That could mean different things to different people."

In the lab, researchers can run complete experiments to offer new ways to leverage information from both types of sensors -- human and computer.

Graham also has developed robust game-based tools for intelligence research. The Synthetic Counter Insurgency, or SYNCOIN, is a dataset of fictitious but realistic messages and intelligence reports about insurgent activity in Baghdad.

Recently, intelligence analysts at Raytheon used the SYNCOIN data to design a game-based exercise to determine if detecting confirmation bias was possible. The researchers at Raytheon said they expect the game may one day be used to develop training techniques.

Taking the Pulse

As the tumultuous, grass-roots uprisings of the Arab Spring seem to indicate, social networks like Facebook and Twitter increasingly play a central role in communicating goals and organizing responses to everything from popular unrest to natural disasters.

Red Cell students tested have how people may use Twitter during an emergency, such as the touchdown of a tornado. Graham said the teams experiment with both positive uses of the social network -- distributing aid and medical care -- and negative uses -- identifying targets for possible looting.

David Hall, dean of the College of Information Sciences and Technology, said the Red Cell Lab goes to the heart of the College and the University's mission, not only providing research and service to the community, but involving students as researchers.

"What Jake has done here at Red Cell Lab is interesting," Hall said. "The students are the researchers and, in a way, they also become the research subjects."

Matthew J. Lesniewski, a research assistant in the Red Cell Lab, said his experience has given him opportunities he probably would not have had at other universities, including working with Penn State Emergency Management on how social media may influence how emergency workers respond to a disaster.

"This semester, we were able to participate in a joint Emergency Operations Center (EOC) exercise where men and women from a number of Penn State offices and Centre County emergency services responded to a mock disaster," Lesniewski said. "In doing so, we learned a great deal about how our social media technologies play a role during a crisis."

Lesniewski said the flexibility of the Red Cell Lab learning environment allows him to study current situations in near-real time.

"If I see a trend or pattern forming in current events, or if I happen upon an interesting case study that offers a challenging analytical vignette, I have the freedom to investigate with the full support of the Red Cell Lab," Lesniewski said.

Last Updated July 28, 2017

Contact