Psychological Safety in Aviation

Psychological Safety, Aviation Disasters and Crew Resource Management

This week I’ve been reading “Confronting Mistakes” by Jan Hagen. This is a fantastic book, focusing on the human factors behind a multitude of aviation incidents. Jan highlights how miscommunication and failures to speak up were behind so many aviation disasters, and the foundation for this dynamic was often the perceived hierarchy, power and status that captains held (even if the captain was drunk). In all of the incidents in the book, a common theme is low psychological safety – that people, even experienced and well qualified co-pilots, did not speak up about concerns or mistakes, which either contributed to or led directly to the disaster. 

Jan points out an interesting 1970s study in simulations for flight crews when they were presented with a complex and difficult series of incidents: “…subsequent evaluation of each crew’s performance revealed very different results. Some crews had worked together effectively and landed the plane safely, while others made mistakes that, in normal circumstances, would have put the flight at risk. Each time, the errors could be traced back to poor interaction. In none of the cases were they down to a lack of technical knowledge or insufficient flying ability.“

Cockpit Resource Management

This discovery, and a series of serious incidents including the 1977 Tenerife airport disaster at Los Rodeos Airport, catalysed first the creation of “Cockpit Resource Management” (CRM) by NASA, which was rapidly adopted by the aviation industry and evolved further into Crew Resource Management – incorporating not just the cockpit crew, but cabin and ground crew as well. By the 1990s, CRM had become a global standard for aviation. Between 1983 and 2002, not only did the number of accidents in the aviation sector fall, but the proportion of pilot errors involving reportable incidents also dropped by nearly 70 percent. CRM includes training on speaking up directly, fostering psychologically safe conditions for others to speak up, reducing power gradients, and training in simulations to highlight the importance of (and build capability in) effective comunication. We’ll cover CRM fully in a future article.
 

“The risk of making a fool of oneself (or being yelled at) is much higher if one speaks than if one stays silent. Under these circumstances, it is not surprising that flight attendants are less assertive, more afraid to disagree, and less likely to challenge senior crewmembers decisions.” Krivonos, 2005.

In our psychological safety training and workshops, we dive much deeper into the practices and behaviours taught in CRM, and apply them to non-aviation contexts. Find out more about our training and workshops, and sign up here. Or ask about our corporate and organisational training services.

A lack of psychological safety doesn’t just cause mistakes in the cockpit though. The lack of psychological safety at Boeing, highlighted here by Amy Edmondson, contributed to serious faults in the 737 Max that caused two deadly disasters that could not have been prevented even by flawless CRM in the air.

Tenerife, 1977

The worst single aviation disaster in history actually occurred on the ground. I mentioned the Tenerife disaster above, and this podcast episode of Black Box Down dives into that incident involving two 747s at Los Rodeos Airport on the island of Tenerife. Due to a terrorist attack on a nearby airport, time pressure, confusion, miscommunication, and a failure to speak up at critical moments, the two planes collided, resulting in the deadliest commercial aviation disaster in history. At a critical moment, the Flight Engineer, Willem Schreuder did not speak up about his (correct) concern that the Pan Am 747 was still on the runway that Captain Van Zanten was about to take off from. A key finding from this disaster was that people do not speak up against power gradients, even when lives are at stake.
 

The Kegworth Disaster

This is the earliest reference I can find explicitly to psychological safety and an aviation incident, where a plane crashed on approach to East Midlands Airport because the pilots turned off the wrong engine, even though cabin crew could see the engine was still on fire – but did not raise their concerns. I actually have personal connection to this disaster – I used to manage field trials at the research farm at The University of Nottingham at Sutton Bonington, and on occasion, we’d find a piece of aircraft debris in a field from this crash.
 

Here is a fascinating, gripping and at times harrowing episode of Cautionary Tales by Tim Harford – “The Vigilante and The Air Traffic Controller”. Tim describes a number of incidents where mistakes by ATC result in instructions that conflict with TCAS (the Traffic Collision Avoidance System) issued to pilots, and end in disaster. The episode covers effective communication and speaking up, human error, blame, work-as-imagined vs work-as-done, and how disasters always have multiple causes. There’s an interesting and rather disturbing twist, too.

Go-Arounds

This is another great podcast episode from Aviation Mentors on “What causes Common Aircraft Incidents?”, and the key point here is “You can always go around”. Whether you’re 50 feet from landing or 5 miles from landing, if something isn’t right, you can always go around for another attempt. And I think that’s a lesson that many of us can learn from aviation: not only can we always try again, but we should be really careful (as the aviation industry is) to never shame people for having another go.
 

Finally, I’m kind of addicted to this: Lofi ATC radio. Head over to lofiatc.com and listen to lofi beats overlaid with transmissions from Air Traffic Control from your choice of tower or ground crew across US airports. Great for background tunes for work or study. (Update February 2025: the Lofi ATC site is currently nonfunctional due to being blocked by the ATC feed, but you can donate to help get them back online!)

References:

Krivonos, P.D., 2005, February. Communication in aircraft cabin safety: lessons learned and lessons required. In Proceedings of the 22nd Annual International Aircraft Cabin Safety Symposium.

Comment: Production, Protection, and Psychological Safety in Aviation Operations

Daniel Scalese, Program Administrator at the USC Aviation Safety and Security Program, USC Viterbi, University of Southern California

Aviation has learned many lessons through tragedies, including those mentioned in this newsletter. In response, the aviation industry has developed and implemented the Safety Management System (SMS), which creates standard procedures, methodical processes, and organizational structures to identify, evaluate, and mitigate risks before they result in damage, injury, or loss. An SMS requires full involvement from all stakeholders to succeed, and psychological safety is a precondition to universal participation. The stakeholder that needs the most psychological safety may have the best understanding of the operation’s hazards, and the SMS requires the involvement of that stakeholder.

Yet safety, particularly the mantra of “safety first,” is not a strategy to prevent every organizational failure. James Reason, professor of psychology at the University of Manchester, describes how successful organizations balance production and protection. This newsletter’s case studies show that production at the expense of protection leads to disaster. Conversely, protection at the cost of production leads to bankruptcy. While we prefer bankruptcy to disaster, both represent the organization failing to complete its mission.

This past winter, a near-example of protection leading to bankruptcy occurred in North America. Severe weather grounded most airlines, and when the weather cleared, Southwest Airlines experienced great difficulty resetting their flight operations. Per the airline’s published action plan (https://www.southwest.com/travel-disruption-action-plan/), a combination of insufficient winter infrastructure, inadequate crew-scheduling software, and poor cross-workgroup collaboration kept Southwest on the ground while competitors resumed normal operations. Southwest correctly chose to keep airplanes grounded instead of asking employees to work in freezing temperatures without protection, operating flights without enough crewmembers, and dispatching planes without approval from all relevant workgroups.

The conditions for Southwest’s flight cancellations did not start with falling temperatures and snow. They began with insufficient operational infrastructure, tools, and organizational communication: a lack of buffer, or “slack” – possibly reflecting an emphasis on being “lean” over being resilient? The resulting inadequate production culture and infrastructure could not balance against the airline’s stellar flight safety culture, and the airline’s primary operation, flying passengers, collapsed. Southwest put safety first, and it came too late to keep the airline flying. Would a greater degree of psychological safety across the organisation have helped people raise concerns productivity as well as safety?
 We will likely learn more about Southwest’s mass cancellations in the coming months and years. I hope people in aviation and psychological safety continue to study this case, both for what created the conditions of the failure and the decision to preserve flight safety over production.

Psychological Safety in the Workplace

This is a great article by Kristopher T Blume, fire chief of the Meridian (Idaho) Fire Department, about psychological safety in the Fire Service. Kris also references CRM in aviation as a precursor to the current paradigm of psychological safety in the workplace, and points out that if firefighters do not feel supported by their team, supervisors and peers, they may be less likely to speak up about their concerns or ask for help.
 

I love this set of visuals from Pascal Bornet, particularly this one showing weaving through cones – which is exactly what I’ve been doing a lot of recently on my motorbike lessons. At first, especially moving up to a much larger 800cc bike, I found this nearly impossible. One of the things that I thought of was my 7-month old daughter, who shows no fear of failure: she just tries stuff and persists until she gets it (or gets too tired and needs a nap!). She doesn’t entertain any of the self-limiting beliefs that hold us back as adults. I always imagined that I’d hopefully be inspirational for her, I never actually considered that she’d inspire me. 🙂
 

Are you interested (like me) in psychological safety AND Kaizen? Here’s a great piece from Balachandhiran Sankaran: “navigating the landscape of Psychological Safety and how systemic conditions unlock the keys of Kaizen by aligning our actions towards achieving a common goal of the system”.

Things to do and try:

Adaptive Monitoring and Management

In a one-day workshop this week with a great bunch of folks in Leeds, we discussed in depth how we might measure the progress and change over time of psychological safety and other dimensions of organisational transformation. The consensus was “it’s complex” and that quantitative metrics defined at the beginning of the programme are likely to be minimally useful and possibly even damaging. 

This is where I like to draw on my first degree in ecology, and how to measure change in complex environments over time. I learned ecological monitoring methods such as quadrat sampling, impact surveys, eyeball estimation, line-intercept, line- and grid-point intercept, and Parker three-step. Parker Three-step is a mixed-method (qualitative and quantitative) approach to gathering insights about trends in ecosystems, and acknowledges that when we’re monitoring a changing system, the way we monitor should change too.

This is a form of “adaptive monitoring”: a concept that has been around in ecology since the 1990s. Adaptive monitoring combines with adaptive management (you could consider this “agile ecology!”) “Adaptive management acknowledges that decisions must be made in spite of imperfect understanding of their consequences; likewise, adaptive monitoring is designed to accommodate the unknown objectives and technologies of the future.” 

This diagram from the excellent paper “Adaptive monitoring in support of adaptive management in rangelands” by Sarah McCord and David S. Pilliod shows the relation between research, monitoring, and management in an adaptive context. (Open access paper)
 

I just had to include this too, a wonderful illustration by Amy Perez (@thefunkieflower) on Instagram:
 

This week’s poem:

Flying Inside Your Own Body, by Margaret Atwood

Your lungs fill & spread themselves,
wings of pink blood, and your bones
empty themselves and become hollow.
When you breathe in you’ll lift like a balloon
and your heart is light too & huge,
beating with pure joy, pure helium.
The sun’s white winds blow through you,
there’s nothing above you,
you see the earth now as an oval jewel,
radiant & seablue with love.
It’s only in dreams you can do this.
Waking, your heart is a shaken fist,
a fine dust clogs the air you breathe in;
the sun’s a hot copper weight pressing straight
down on the think pink rind of your skull.
It’s always the moment just before gunshot.
You try & try to rise but you cannot.