Expert pilots process multiple visual cues more efficiently, Stanford and VA scientists find

A new study has found that expert pilots’ brains process visual data more efficiently, allowing them to make better decisions when landing.

- By Bjorn Carey

Landing an airplane is one of the most difficult piloting techniques to master, and the stats show it: 36 percent of all airplane accidents and 25 percent of fatalities occur during the final approach and landing.

New research by scientists at the Stanford University School of Medicine and the VA Palo Alto Health Care System reveals that expert pilots make better decisions during this phase than less experienced pilots because their brains behave more efficiently.

The study, published today in PLOS ONE, could lead to technology and techniques that guide pilots to more efficient brain behaviors in order to improve flight safety.

The researchers rigged a fMRI machine so that 20 pilots — 12 moderately experienced pilots and eight experienced pilots — could operate the controls of a flight simulator while having their brain activity scanned in real time. The simulator mimicked the cockpit of a single-engine airplane, and the pilots were instructed to land at a virtual San Francisco International Airport.

The trial started the pilots at 350 feet of altitude. They were instructed to begin their descent based only on their instrument readings, as would be typical in most real-life flights. Once they reached 200 feet — the altitude at which the Federal Aviation Administration mandates you must be able to clearly see the runway in order to land — the program would display the runway, either clearly or obscured by varying degrees of fog.

The pilots would then need to flash their gaze from the instruments to the runway and back to make a snap decision about whether or not it would be safe to continue the approach.

Better decisions, half the brain activity 

After multiple attempts per pilot, the results showed that the expert pilots made the correct decision about whether to make a landing attempt or abort 80 percent of the time, whereas the moderately experienced pilots did so only 64 percent of the time. Interestingly, the fMRI scans revealed that the expert pilots scored higher while displaying only half as much brain activity.

The flight simulator depicted either a relatively clear view of the runway (top) or conditions that made it too foggy to land (bottom). 
Courtesy of Maheen Adamson

Landing a plane involves constantly scanning instruments as well as the view out the window, said Maheen Adamson, PhD, lead author of the study and a clinical associate professor of psychiatry and behavioral sciences at Stanford. Reduced neural activity in expert pilots indicates that they are able to complete the task at hand with fewer brain resources. She suspects that the brain’s ability to streamline multiple visual inputs is the result of experience.

“The data show that the expert pilot seems to just know what to look for, where to look and when to look,” said Adamson, who is also the director of research and PhD fellowships at the War Related Illness and Injury Study Center at the VA Palo Alto Health Care System. “And we’ve been able to trace that skill back to the caudate nucleus.”

This is an area of the brain involved in regulating gaze as the eyes quickly shift their focus to different fixed objects. The work needs to be replicated to confirm the caudate nucleus’ role in instrument scanning, Adamson added.

Other brain structures or mechanisms might be playing a role in visual processing for pilots. But this work opens the door to pairing fMRI and flight simulators — something that NASA is already doing in limited trials — to test pilots’ mental engagement during various flight maneuvers. Researchers might also be able to design interactive lessons that guide less expert pilots into behaviors that mimic the more efficient brain activity of more expert pilots.

Other implications

“In the future, we could put a trainee in a scanner and see what mechanism in their brain they’re using to get better at the task, and train them to reallocate resources in the brain more efficiently,” Adamson said. “If we are able to train pilots to process instruments and other visual cues more efficiently, you could reduce the likelihood of accidents during landing.”

Although these findings were obtained by testing aircraft pilots, they may also have implications for skilled task performance and aging in general. Prior work in the same laboratory showed how slower speed of information processing predicts worse performance on aviation-related tasks in older pilots. The current research offers an insight into how more efficient brain activity and information processing may provide a countermeasure to loss of speed of processing with aging. 

The study’s senior author is Jerome Yesavage, MD, professor of psychiatry and behavioral sciences, associate chief of staff for mental health at the Palo Alto VA and director of the Stanford/VA Aging Clinical Research Center. Other co-authors are Joy Taylor, PhD, a clinical associate professor of psychiatry and behavioral sciences at Stanford and assistant director of the Aging Clinical Research Center; Daniel Heraldez, a former research assistant at the medical school; Allen Khorasani, a former computer programmer in the Department of Psychiatry and Behavioral Sciences; Art Noda, MS, a database manager at the medical school; and Beatriz Hernandez, a data analyst at the school.

This research was supported by the Sierra-Pacific Mental Illness Research, Education, and Clinical Center; the War Related Illness and Injury Study Center; the Medical Research Service of the Department of Veterans Affairs; and the National Institutes of Health (grants P30AG17824, R37AG12713, P41 RR023953 and 2P41RR009784).

More information about Stanford’s Department of Psychiatry and Behavioral Sciences, which also supported the work, can be found at http://med.stanford.edu/psychiatry.html.

About Stanford Medicine

Stanford Medicine is an integrated academic health system comprising the Stanford School of Medicine and adult and pediatric health care delivery systems. Together, they harness the full potential of biomedicine through collaborative research, education and clinical care for patients. For more information, please visit med.stanford.edu.

2023 ISSUE 3

Exploring ways AI is applied to health care