Psychological Safety on the Flight Deck

Crew resource management (CRM) is aviation’s global gold standard for training pilots on crew collaboration. Since its inception 50 years ago, it has evolved to incorporate new academic research and new safety models.

In the 1990s, an updated version of CRM emerged to include an integrated safety model called threat and error management (TEM). TEM is a conceptual model designed to identify, mitigate and trap errors on the flight deck. It assumes pilots always speak up, admit mistakes, ask for help and share safety concerns. However, my research has shown these baseline assumptions are not always correct. When pilots do not speak up about safety concerns, the CRM/TEM safety model degrades to little more than hopeful rhetoric.

Two of the assumptions built into the CRM/TEM model are that captains will successfully foster a collaborative dynamic on the flight deck, and that both the first officers and captains will (consistently/unfailingly) share safety concerns. We train pilots on the imperative necessity for these behaviors, but rarely measure the flight deck microculture to see whether they are actually occurring.

Recent academic research reveals that 93 percent of first officers feel compelled to adapt to the culture style established by the captain.¹ This is no surprise since leaders are expected to set the tone of a workplace.

The research also revealed that 75 percent of first officers report they shift from a safety voice (clearly communicating a safety concern) to a muted safety voice² (hesitating to report a safety concern or suggesting — rather than directly reporting — a safety concern), according to the tone established by the captain.³ In an even more alarming statistic, 57 percent of first officers report having felt silenced by the captain after sharing a safety concern.⁴ 

As a pilot and academic aviation safety researcher, I hypothesized that an underlying cause of this reduction in safety voice is due, in part, to a lack of psychological safety on the flight deck⁵ and set out to determine if psychological safety might be a missing element in CRM/TEM training. In a study of more than 800 industry pilots, I found that a reduction of flight deck psychological safety dramatically reduced safety voice. I used various aspects of psychological safety (for example, a pilot’s ability to admit mistakes or ask for help) to measure its level and to determine how the perception of crew dynamics affects safety voice.⁶

Additionally, I reviewed the U.S. Federal Aviation Administration’s (FAA’s) recommended CRM training curriculum and discovered that we, as an industry, initially built safety models and systems on the assumption that psychological safety was omnipresent in flight deck microculture. I found that pilots in general have a major training gap involving interpersonal skills — the necessary tools to help build psychological safety. 

Training Gaps

 In 2004, the FAA introduced guidelines to train pilots in collaborative safety and interpersonal skills. Important concepts such as “ways to behave to foster crew effectiveness,” “strategies to handle conflict” and “external influences on interpersonal communications” debuted in an advisory circular.⁷ Nevertheless, my research shows many pilots do not feel they have been trained on these important interpersonal skills. A survey of more than 800 professional pilots revealed that only about 50 percent of airline pilots believe they received training on the FAA recommendations.⁸ 

The advisory circular encouraged CRM facilitators to assess a pilot’s competency in CRM by having pilots demonstrate “the usefulness of showing sensitivity of other crewmembers’ personalities and styles.” Since U.S. airline pilots are required to receive CRM training, I asked them whether they ever had to demonstrate this facet of CRM. Fifty-one percent said they had not, and 85 percent said such training might enhance safety.⁹

A more disheartening finding was the emotional response to the word sensitivity in an aviation safety survey. One pilot, reflecting the opinion of many others, wrote “Sensitivity towards others? Is this an airplane or a therapy session? Cockpit demands respect of others not sensitivity.”

Another pilot did not complete the survey because the word was too upsetting, evident from their comment, “I stopped at the word “sensitivity … that does not contribute to good CRM.” Fear of interpersonal skills becoming too “touchy-feely” was a prevalent theme of many pilots’ comments, as seen here: “Feelings, gender, race, religion etc. don’t belong in the cockpit. That’s why they have checklists. Do your job and we’ll get along fine.” The emotion exuding from these comments indicates that this isn’t just a training gap – there’s a culture issue that needs to be addressed.  

The fear of having to display sensitivity is often used to discredit collaborative safety. This phenomenon is not new. Similar sentiment was well documented in groundbreaking research on airline pilots in 2005 by Karen Ashcraft, a communications professor at the University of Colorado.¹⁰ She hypothesized that the resistance to collaborative safety in the early days of mandated CRM resulted from a perception that collaborative safety was emasculating and a threat to power structures. The persistence and prevalence of some professional pilots’ inability (or unwillingness) to grasp the interdependency of crew collaboration and overall safety is a missed opportunity for CRM initiatives and an increased safety risk in crew operations.

S-Frame Solutions

We can initiate solutions that are either i-frame (i for individual) or s-frame (s for system).¹¹ I-frame approaches assume that individuals may choose to alter their cognition, affect and/or behavior for moral, altruistic or personal gain. Such is the work of behavioral and cognitive scientists. Recent studies show much of the failure of diversity training programs stems from a reliance on i-frame approaches, hoping individuals, en masse, independently choose to become more inclusive.¹² They usually don’t.

Instead, we must approach the training of pilots’ interpersonal skills through s-frame interventions. Policymakers and social psychologists use s-frame interventions to change the rules of the game – requiring a behavior change, which eventually leads to a shift in how we think (cognitive) and how we feel (affective) about something.

Rather than waiting for individuals to change, we must implement a strategy to require change. Our s-frame solution will fill the training gap while increasing the efficacy of CRM/TEM by removing the erroneous assumptions on which the models were built. We cannot wait for a collective epiphany of pilots to individually seek enlightenment on collaborative safety.

My recommendation is that regulators require enhanced interpersonal skills training to include the following concepts: psychological safety, the triad of bias (cognitive, affective and behavioral), interpersonal communication and resiliency. This training should occur early on (as early as commercial pilot license) and must occur at all levels of professional pilot development (initial, recurrent and upgrade). Furthermore, we must operationalize these concepts by requiring competency-based training in which pilots are required to demonstrate their ability to create psychological safety. Without re-inventing the wheel, we can integrate these concepts into frameworks and systems that already exist.¹³

Further justification for an s-frame solution is recent academic research revealing that U.S. pilots, in general, score lower than the general public on emotional intelligence traits.¹⁴ Here’s why this is critical: Emotional intelligence is a strong predictor of safety performance.¹⁵ The industry needs a mandatory s-frame intervention on collaborative safety.

Two necessary aspects of an s-frame intervention are to establish the terminology that is adequately representative; and to utilize a vernacular that radiates familiarity as a tool to elicit buy-in. In this light, I advocate that we position the flight deck as a sociotechnical system in all future human factors discussions. The flight deck consists of a socio, or social, aspect (emphasizing the human role) and a technical aspect (focusing on the airplane and its technology). The health of the socio impacts the health of the technical. The interdependency of a healthy socio on the functionality of the technical is evident from recent aviation news. In a display of pro-safety decision-making, Alaska Airlines Flight 1080 returned to the gate at Washington Dulles International Airport in July 2022 when the two pilots had a dispute and determined they could not fly together.¹⁶ The socio conflict impacted the ability to operate the technical. The event illuminates a need for better interpersonal skills training integrated into pilot training.

When pilots display emotion-laden rejection of collaborative safety or deem CRM too touchy-feely, they are acting in an unsafe manner. The flight deck is a sociotechnical system, and it’s time the industry does a better job of training the socio aspect of safety.

The FAA now requires airline captains to receive leadership training, but with no mandated syllabus, pedagogical approach or methodology, it is likely the plight of leadership training will end up similar to that of CRM. Important concepts like growth mindset, emotional intelligence and psychological safety are, at best, mentioned in ground school. More realistically, these concepts go unnamed and untrained, but their ethos may be plastered on a poster hanging somewhere in the halls of a pilot training facility. We can do better than this.

The fear of collaborative safety or the touchy-feely-ness of interpersonal skills reflects a culture of hyper individualism and a lack of understanding of sociotechnical systems. It is a culture laden with emotion that rejects change, thrives in the status quo and feels threatened by diverse thinking or by diversity itself. It would be irresponsible to assume a culture shift will occur naturally (despite the common temporal argument of wait a generation). We need an s-frame intervention to nudge this culture in a new direction. We can do this without reinventing the wheel.

Our industry loves quantification — we are numbers and stats people. We measure what we care about (number of long landings, go-arounds, hours flown and safety reports, for example). But we are not adequately quantifying culture. There are nuances of CRM, such as those suggested in the CRM advisory circular, that must be measured. The current metric of CRM success — often a checked box at the end of a simulator ride or a “good CRM” comment written on an assessment — is insufficient. 

The upsurge of human factors (HF) specialists over the past decade indicates the industry is moving in the right direction. From integrating HF into aircraft certification to designing a taxonomy of cognitive processes, HF specialists are uniquely positioned to be the vanguard of the inner workings of the socio dyad of our sociotechnical system. We ought to be thinking about the structural facets of crew dynamics: How does power sanction exclusionary behavior? How does psychological safety influence the flight deck microculture? What is the role of individual resiliency in building crew psychological safety? The answers to these questions, I believe, will enhance safety.

Aerospace is such a thrilling industry because we are continuously pushing for the next technological advancement. Let us now push for the next sociotechnical advancement by spotlighting the need for enhanced interpersonal skills training. As a pilot, I, for one, am on board with enhancing safety. How about you?

Image: Yakobchuk Viacheslav / Shutterstock.com

A version of this article was previously published in the August 2023 issue of the Royal Aeronautical Society’s Aerospace.

Kimberly Perkins is a captain on a Gulfstream 650 and a doctoral researcher at the University of Washington. Her research focuses on enhancing aviation safety systems through human centered design and engineering with an emphasis on cognitive and behavioral science. She is a fellow of the Royal Aeronautical Society and serves as an academic adviser on the CAE Human Performance Excellence Council.

Notes

1. Perkins, K.; Ghosh, S.; Vera, J.; Aragon, C.; Hyland, A. “The Persistence of Safety Silence: How Flight Deck Microcultures Influence the Efficacy of Crew Resource Management.” International Journal of Aviation, Aeronautics, and Aerospace Volume 9 (Issue 3, 2022): 6.
2. Noort, M.C.; Reader, T.W.; Gillespie, A. (2021). “Safety voice and safety listening during aviation accidents: Cockpit voice recordings reveal that speaking-up to power is not enough.” Safety Science Volume 139 (July 2021), 105260. 
3. Perkins et al.
4. Ibid.
5. Psychological safety is a team condition in which members of the team feel that it is safe (in other words, they won’t be viewed as incompetent or receive retaliation) for interpersonal risk taking, such as asking for help, admitting mistakes and sharing ideas. It is a team condition in which individuals share safety concerns and feel that they are valued members of the team.
6. There was over a 50 percent increase in the number of pilots who strongly agree that they can admit mistakes and ask for help on the flight deck when the pilot perceives team synergy (that is, getting along) with the other pilot.
7. FAA. Advisory Circular 120-51E, “Crew Resource Management Training.” 2004.
8. Perkins, K.; Ghosh, S.; Hall, C. (Under Review). “Interpersonal Skills in a Sociotechnical System: A Training Gap in Flight Decks,” 2022 academic research conducted at the University of Washington.
9. See U.S. Federal Aviation Regulation Part 121.405.
10. Ashcraft, K.L. (2005). “Resistance Through Consent? Occupational Identity, Organizational Form, and the Maintenance of Masculinity Among Commercial Airline Pilots.” Management Communication Quarterly Volume 19 (Issue 1, 2005): 67–90. 
11. For more on i-frame vs. s-frame see: Chater, N.; Loewenstein, G.F. “The i-Frame and the s-Frame: How Focusing on the Individual-Level Solutions Has Led Behavioral Public Policy Astray.” SSRN Electronic Journal. 2022.
12. Study of 985 anti-bias interventions revealed training does little to reduce bias; a meta-analysis of 426 studies found unconscious bias training has little effect on altering behavior. See Dobbin, F; Kalev, A. Getting to diversity: what works and what doesn’t. The Belknap Press of Harvard University Press. 2022.
13. For example, line operations safety assessment (LOSA), line oriented flight training (LOFT) and advanced qualification program (AQP)
14. Dugger, Zachary; Petrides, K.V.; Carnegie, Nicole; McCrory, Bernadette. “Trait Emotional Intelligence in American Pilots.” Scientific Reports12, no. 1 (2022): 15033–15033. .
15. Bates, S.G. Emotional intelligence and safety culture in business aviation (Order No. 30250161). Available from ProQuest Dissertations & Theses Global. (2777980712). 2023. 
16. List, Madeleine. “Alaska Airlines Pilot Clashes with Copilot, Walks Off Plane ‘Fuming,’ Passengers Say.” Miami Herald, July 22, 2022.