Senior Airman David Carbajal
MQ-1B Predator unmanned aerial vehicles sit in a clamshell at night at Kandahar Airfield, Afghanistan, July 31, 2011.
Driving a war drone is a stressful business. Shifts up to 12 hours long are stretches of dullness, watching and waiting, interrupted by flashes of intense activity in which pilots must make life-or-death decisions. Not their own life or death, however.
Pilots may be thousands of miles away from the flying weapons system they're operating. They often head home at the end of the day, as if returning from any other office job, maybe picking up milk on the way. But while at work, their drones' onboard cameras put them in a unique position to watch people being killed and injured as a direct result of their actions.
As psychologists learn more about the mental scarring warfare leaves on drone pilots — caused by long shift hours, isolation, witnessing casualties and those Jekyll-and-Hyde days split between battlefield and home — experts from within the U.S. Air Force are calling for a review of drone pilot selection.
Brad Hoagland, an Air Force colonel and visiting researcher at the Brookings Institution, and a fighter-jet pilot and operations commander of 23 years himself, believes that drone pilots could be picked better, and that existing selection techniques are due to be updated now that the service has accumulated almost a decade of research into the psychological characteristics of drone pilots.
"The thrill of taking off from a runway, flying a mission and then coming back and landing at the end of the mission — that’s very exciting," he told NBC News. "But I think that’s a different type of person who can do that, than someone who is maybe wired to fly an unmanned system from a console 7,000 miles away. It’s a different psychological makeup requirement to execute the mission."
Right stuff, wrong stuff
"I think we are still trying to figure out exactly what the 'right stuff' is," Wayne Chappelle, a clinical psychologist consulting for Air Force Medical at Wright Patterson Air Force Base in Dayton, Ohio, told NBC News. "We have a general idea ... but I certainly think we're probably more aware of what the wrong stuff is versus the right stuff."
The trouble is that spotting the known positive attributes in up-and-coming drone pilots is harder than spotting the negative attributes. To begin with, Chappelle drew up a portrait of the ideal drone pilot from the recorded testimony of 82 drone pilots and their supervisors in a 2011 report.
Good drone pilots, according to Chappelle's findings, have excellent memory for pictures and sounds. They are bombarded with sounds and images from multiple screens through their long shifts, but parse that data quickly, cutting through the noise. They're multitaskers and collaborators.
"These guys are very smart, very bright in a wide range of areas. They are emotionally resilient and highly stress tolerant and very motivated," Chappelle said.
People who have a history of abuse or dependence on alcohol, drugs or other substances, anxiety or depression, and cognitive impairments such as learning disabilities tend to make bad drone pilots.
Although the strengths of a drone pilot differ from the strengths of a manned fighter pilot, Chappelle said the psychological screening protocol for both is the same — and hasn't changed in a decade. "We're still looking at ways to improve and expand upon the screening procedures."
In his research, Hoagland has found that washout rates among undergraduate pilot trainees headed to crafts like the F-16 are traditionally about 10 to 15 percent. But drone pilot trainees exit at 30 percent (though that's down from 45 percent a few years ago). Pilots may drop out, but more often, they fail to meet some flight or academic criteria along the way, Hoagland said.
And when they do graduate, they receive mental health diagnoses at a rate on par with pilots who fly in aircraft, and at much higher rates than other non-pilot Air Force personnel, according to a February 2013 report by the Armed Forces Health Surveillance Center.
NBC News has requested to interview a pilot or pilot instructor at the Holloman Air Force Base in New Mexico, where drone pilots are trained, but to date the Air Force has declined the request without further explanation.
2nd Lt. Logan Clark
A pilot trainee flies an MQ-1 Predator simulator mission at Holloman Air Force Base in New Mexico on Feb. 8, 2012.
In an upcoming report headed to the Pentagon, Hoagland will suggest some fixes for his higher-ups to consider.
For one, though the Air Force has a test called the Pilot Candidate Scoring Method, not all pilot candidates — of drones or manned craft — are given the exam. (The Air Force Academy, for example, only recently started administering it, and only on an "experimental" basis.)
"I can't believe we as an Air Force haven't standardized this," Hoagland says. Once everyone's taking the test, and baseline scores are set, those scores can be mined for indicators as to who might be better suited to fly an F-16 and who might be destined for a drone. "It's a common sense approach."
Also, though it's been standard procedure to assess concentration, attention, psychomotor skills as part of the Medical Flight Screening-Neuropsychiatric test in pre-screened pilots-to-be, that information is not used in the selection process. Tests do weed out the medically and psychologically unfit — Hoagland thinks it would be an easy next step to ask: "Is this person suited for an unmanned or manned system?"
The coming swarm
As the Air Force's drone program grows, so does the importance of pilot selection. What started in 2004 as five drone combat patrols — four aircraft each — will to swell to 65 patrols by 2014. By 2010, Predators had logged more than a million combat hours, more than any other military bird. And today's population of 1,300 combat drone pilots will be joined by 500 more in the next few years.
And as autonomous systems evolve, the capabilities of unmanned craft will, too. The Air Force will shift to a system with multiple vehicles flown in tandem, answering to a single pilot. These "swarm" handlers will have more complex tasks heaped on them earlier in their career.
"In terms of who we need to have, I think we're on a learning curve there," Anthony Tvaryanas, a doctor of aerospace medicine and technical advisor with the 711th Human Systems Integration Directorate at Wright Patterson Air Force Base in Dayton, Ohio, told NBC News.
"If [a pilot is] operating a swarm, what are you looking for in that person? I don't think anyone's looking into those concepts," Tvaryanas said.
"As we get from a pilot in an airplane to a pilot outside the airplane to a pilot controlling 100 airplanes, I think we're approaching the limits of what [prior experience and studies] can inform us. There's a need to look back at training," he added.
Nidhi Subbaraman writes about technology and science. Follow her on Facebook, TwitterandGoogle+ .
First published May 17 2013, 1:15 AM