When should Safety intervene on site when it’s not our business? Should we just support our teams by sticking to fixing hazards and unsafe behaviors? Here’s what I mean…
Not long ago, a director lamented, “I want that lead-man to take the foreman position so that he’ll be in line to manage shops, but he says he’s happy where he is.”
It’s no secret here that I’m an active busybody; I owned and ran two businesses and was a reliable rock for my partner and our employees. When I got my Certified Health and Safety Technician certificate from BCSP fourteen years ago, I pledged to honor public safety. When I see unsafe practices anywhere, I get involved and try to stop them. (That’s a Tik Tok if you ever saw one.) So, why wouldn’t I help our teams? You know, for them to excel.
If you’re still reading this then you know how many sides exist in Safety. At the core, it’s about nurturing and engaging workers in our message, building trust like an owner would, if you were one. And like an owner, we are out here to affect better working conditions.
Bosses look for born leaders, employees who know their job and do it well. It doesn’t mean every leader wants to move up, though. They may not see their future like you do. Here’s what they might see in your company’s leadership instead, so take heed.
Managers work overtime unpaid. They might be asked to give up holidays, family events, and their kids’ games. The reality is, today’s foremen are “time poor.”
Sure, foremen get paid more money, but Lead men and women make their own overtime. When extra money is needed a lead can choose when to work it.
Good leads are hard to come by. A good lead has job security. They don’t get fired.
To some people a foreman is low on the management rung. It seems that every day, someone blames them for some new issue, no matter what.
What is inexcusable is that most foremen get blamed for every injury affecting their crew. It’s lazy stuff – the quickest resolution for ‘time poor’ managers to make.
Foremen, managers and directors regularly get fired, sometimes for the smallest offense. To a lead man, when they see it, foreman is a weak position to aspire to.
What is honestly happening here? The answer is there is no unity between the field and management. There should be one team – one purpose — and there are two. In fact, you might say being a foreman is thankless and prone to being dumped on. Why would any sane person take it?
In previous articles I have talked about advancing a business structure called, ‘the trinity business.’ A trinity business has three equal parts that fit well in large, medium, and small construction, manufacturing, and Utility companies. It ends workplace dysfunctions.
Think of the ‘trinity’ as a thick rubber coupling between metal pipes carrying a very precious liquid. Bury these pipes, add frost and heat that expand and contract them; then some heavy vibrations and loads roll over them, even toss in a few low earthquakes. The trinity business still holds, never spilling a precious drop.
There are many reasons why a trinity business succeeds; mainly because the trinity is built on three equal parts: Production, Quality Control, and Safety. Secondly, it is run with respect to each of the three. And finally, it is based on common sense; the kind of common sense and respect for people that can come out of the field.
A trinity business doesn’t cost anything to begin and run, and yet it makes money and keeps your talent from moving on.
The trinity is a field/management system sharing control of the three most crucial drivers, but the glue that binds these together is Psychological Safety. This growing trend is a perfect fit for any business built on mutual respect and common sense. Go to psychsafety.com. Located in Nottingham, UK, Jade and Tom run a very busy shop, full of great purpose. The purpose is how expertly they cut the grizzle and fluff, teaching you what counts in business. They are sensible people; they know that growth, safety or otherwise, comes from well-developed skills like patience and common sense. Based in respect, here is their abbreviated mission statement:
To make the world of work a safer, higher performing, more inclusive and equitable place.
Empower people across the world to foster psychologically safe environments in their organizations and teams.
We’re building a global community of folks who want to create and maintain psychological safety.
A core goal is to amplify the voices of those who are less represented in this field. This includes (but is not limited to) voices of people of color, neurodiverse people, LGBTQ+ people, women in leadership and tech, and more.
We believe knowledge is worthless unless it’s accessible. A key principle of this (Friday) newsletter is to only share content that everyone can access.
Honestly, there is no gimmicky product here for sale.
By now, we all know that our own universal safety missions are a ‘moveable feast’. Safety methods come and go; some with gadgets, some without and all are designed to make us safer. Unfortunately, we deal with sentient human beings. Well-worn ideas top imperfect content, while promised patented products crush careers every season. With all those unpredictable, unsafe choices we humans make every day, success is made in baby steps.
This is why Psych Safety is so attractive to me — it focuses on how we think and function, our physiology, the role of dopamine, hormones, and other natural chemicals, and how our archaic brains reward us when we choose unsafe actions. Those one second calculations we make to bring home the bacon are totally unpredictable. Psych Safety practiced in a trinity structured company may very well be the incoming wave to choose.
Jade and Tom know that it takes ‘a village to raise safe people.’ That’s why they share what they learn with others via their platforms, their training, on LinkedIn, and by mouth. I’ll tell you what a colleague told me recently. He said that in a few years AI will figure us all out finally and chart a course that will make us safe… Is this what you’re waiting for too?
Robert Slocomb CHST, Safety Specialist and book author, works for the water wastewater industry in metropolitan Washington, DC.
360 degree feedback promises that it will offer truth in all directions. It will illuminate blind spots, reveal the development areas people couldn’t see before and unlock potential they didn’t know they had. It promises, above all, to make honesty safe, offering confidentiality and anonymity for the people with less power, so they can finally tell the truth about the people with more. It is an appealing set of promises, but whether it can keep them is another matter.
The Murky History of 360 Degree Feedback
Let’s just clarify first what we mean by 360 degree feedback. Generally, the term refers to a process of gathering feedback about an individual from the people around them, perhaps their manager, but also peers and direct reports, hence the full circle – 360 degrees. This is ostensibly to support their development, to feed into an appraisal, or both. Responses are typically anonymised, often pooled in groups of three or more so that in theory no single rater can be identified (although that’s rarely watertight in practice).
So where does this popular practice actually come from? Rumours abound. I once heard a version involving a KGB sabotage manual, which I’d love to believe as it would make a great story. Unfortunately, I’ve found no evidence for it. The more likely history, referenced widely online, traces an early version to officer selection in the German Reichswehr around 1930, under a military psychologist named Johann Rieffert. Nobody called it 360 degree feedback, but multiple observers would feed into a single judgement about who should make the cut.
The name “360-degree feedback” was coined in the mid-1980s by an assessment firm called TEAMS Inc, who registered the phrase as a trademark, and then spent years trying to enforce their ownership of it until the company was sold. There has always been money in this, and attempts to trademark concepts and practices should always raise an eyebrow, if not a red flag.
None of which stops 360 degree feedback being treated today as a solid, almost scientific HR default. For a practice this murky in its origins, that’s quite a leap, and as we’ll see, neither the evidence nor people’s experiences of it seem to justify the confidence.
Experiences of 360 Degree Feedback
I’ve worked in organisations running 360 degree feedback processes, and I know the anxiety of waiting on anonymised comments, followed, almost always, by one of two deflations:
You’re handed some critical but cryptic line, spend an unreasonable amount of energy trying to guess who wrote it and have to actively resist the urge to go and just ask them what they mean, or:
You get the other thing entirely: bland, school-report style platitudes about being a great colleague, which are technically positive, but don’t really say anything at all.
Both versions arrive by the same route. The 360 degree feedback process tips into a tick-box exercise, and once it has, it doesn’t just fail to start a constructive, developmental conversation; the anonymity built into it often makes sure that conversation can never happen at all.
None of this is unique to me. I have these conversations regularly with clients, people in mid-sized and large organisations where 360 degree feedback has become so completely entangled with performance management or appraisal that all ideas of feedback get drawn into this dreaded annual cycle. It seems like once an organisation has a formal annual mechanism for feedback, that mechanism exerts a gravitational pull over all other feedback instances. The ordinary, day-to-day, low-stakes, useful versions of feedback – like a quick check in after a meeting or small flagging of a blind spot – start to feel like something to save up rather than say now. People batch it all up for judgement day and the process that’s designed to surface feedback ends up suppressing many more useful, and low-threat, day-to-day exchanges.
And then there are the others, the ones just starting out on their 360 degree feedback ‘journey’, asking important questions before they commit: why are we doing this, and what is it actually going to do to psychological safety here?
The Evidence Gap
It turns out that the case for 360 degree feedback is pretty underwhelming, given how embedded all this is. Despite three decades of near universal corporate enthusiasm, nobody has ever managed to show, convincingly, that it does what it’s supposed to do. The Institute for Employment Studies spent a year interviewing organisations running these schemes and combing through the available research, and concluded that widespread adoption of 360 degree feedback reflects “faith rather than proven validity,” which is, it turns out, a rather generous way of putting it. Evidence of actual impact on individual development or organisational performance was, in their own words, “scant.” Similarly, the most authoritative meta-analysis in the field, pulling together twenty-four longitudinal studies, found that improvement in ratings over time, the entire point of the exercise, was ‘generally small’. As the authors put it, “practitioners should not expect large, widespread performance improvement after employees receive multisource feedback.”
That meta-analysis was published in 2005. Two decades on, the more recent literature isn’t full of fresh studies overturning or confirming it, it’s largely quiet, with prominent practitioners in the field noting a “glaring absence” of new research even as usage of 360 feedback keeps climbing. The honest answer to “does this work?” still sits somewhere closer to “a bit, maybe, for some people, under conditions nobody’s fully pinned down” than anything resembling a “yes!”
None of which has slowed adoption down. Which brings me back to the money. Google “360 degree feedback” (a reasonable start for any investigative dive into the practice!) and the first page is dominated by sponsored placements and paid products, all companies selling their own version of a 360 degree feedback tool. Twenty years ago, the IES were already noting the volume of “spin from external providers” surrounding this practice. If there was a lot of spin then, I suspect there’s a great deal more now.
More Raters, More Bias
All feedback is subjective. I’ve written about why that matters elsewhere, so I won’t revisit the whole argument, except to say that the moment you treat a rating as a fact about a person rather than an impression formed by another person, you’re already in trouble.
What’s worth adding in this context is a particular claim some 360 degree tools lean on, implicitly or explicitly: that gathering feedback from many sources cancels out individual bias and delivers something more objective than a single manager’s view. It’s an appealing idea, but averaging several biased judgements doesn’t remove the bias, it just blends it into something that looks smoother and harder to attribute while carrying the original distortions inside it. Gender and racial bias don’t disappear when you collect more ratings; in fact they may well accumulate while becoming harder to see, hidden behind an aggregate score that feels objective and is anything but.
There’s another issue too: 360 asks how a person is performing, rarely about what pressures and constraints they were facing. The forms have feedback for the individual and rarely for the conditions, so the subject of the feedback ends up holding the weight of the system around them.
The Trouble with Anonymity
Anonymity exists in these practices for an obvious reason: people are supposed to feel safer telling the truth if they can’t be identified for it. Aside from the (very real) risk that this reinforces the idea that it’s actually not safe to speak up unless we’re anonymous, the IES researchers did find this working roughly as intended – raters giving upward feedback anonymously were more critical than those who knew their names would be attached. But they also found the protection failing in two different ways. In some cases, managers asked direct reports to provide them directly with “anonymous” feedback, which of course undermines the anonymity from the start. In other cases, raters who had been promised anonymity colluded with each other, with several colleagues agreeing in advance to write identical negative comments about their manager so that no single person could be singled out and blamed. They didn’t trust the protection the system offered, so they built their own on top of it.
There’s a further risk in anonymous feedback – it can easily mask people over- or under-rating for political reasons, which is most likely when the stakes are high. In some organisations, 360 degree feedback has reportedly been folded into the machinery of ‘rank and yank,’ the practice of ranking people by their performance ratings and firing the bottom 10%, a practice which is (thankfully) effectively unlawful in many parts of the world. But where it, or practices like it, still exist, are you really going to risk rating a colleague higher than yourself?
When the IES researchers looked at places where anonymity was removed instead, some people reported being approached and made to feel uncomfortable for giving honest upward feedback. Yet others preferred it that way, worried that anonymity would let others use the process to settle old scores under cover rather than to give honest feedback. “At least the feedback can be challenged if the names are on it,” as one of them put it, and I see their point. We know feedback is most useful when it’s context-rich – when it can be located in specific incidents and relationships – and when it’s part of a two-way conversation. Anonymised 360 degree feedback removes both the context and the conversation.
So take anonymity away, and people fear retaliation for their honesty. Leave it in place, and people suspect the cover is being used to settle scores or game the system. Either way, the distrust is still there. The problem was arguably never about anonymity at all, but that nobody in the system trusted that honest feedback would be received well, used fairly, or kept safe, regardless of whether names were on it.
Which brings us back to psychological safety. What people need in order to offer honest feedback, is consistent, demonstrated, lived experience that speaking honestly won’t cost them anything. That can only be developed slowly, through what actually happens when people speak up – we can’t shortcut our way to it via an anonymous feedback form.
The Costs of 360 Degree Feedback
There is a financial and relational cost to 360 degree feedback which is easy to overlook when it becomes ‘the norm’. We work with a number of large organisations stuck with a frustrating annual ritual, gathering and inputting reams of data into a system most people resent, kept alive less by conviction that it actually works than by a software licence they’re tied to. These platforms cost a significant amount of money, and the decision to buy one was often made a long way up the organisation. The cost is sunk; the plan is the plan, and questioning it means telling someone senior that their expensive choice isn’t working. It’s often easier, on balance, to just keep doing it.
In another organisation, someone told me their HR team was considering bringing 360 feedback in specifically to deal with one ‘bad manager’. The logic, as it was explained to me, was that this was the only way to get that manager some direct, unfiltered feedback from the people reporting to them. Which might be true. But it also means reaching for an entire organisation-wide measurement system, with all its cost, fatigue and risk, to solve a problem that’s really about one person — a siege engine wheeled up to a door that only needed knocking on.
So where does that leave us?
Not, unfortunately, with a tidy verdict. I’m not here to tell you 360 degree feedback never works, or that everyone running it should stop tomorrow. The honest position is that we don’t really know what it does, the evidence has been thin for twenty years and a great many organisations are running these processes at real cost to time, money and trust without properly questioning whether they’re helping or harming.
If there’s a structural point underneath all of this, it’s that candid feedback isn’t a system we install. The IES researchers, after all their interviews and analysis, landed somewhere strikingly modest: that organisations might do better simply reminding people to give honest, regular feedback as a matter of ordinary practice, rather than building elaborate machinery to extract it once a year.
I don’t think the story of 360 degree feedback is over. The tools are still being sold, the licences still being signed, the annual cycles still grinding on in organisations full of people wondering what they’re for. I suspect I’ll be having these conversations with clients for a long time yet. But maybe we can at least take a step back and ask what problem this process is actually meant to solve, and whether this is a good way to solve it.
If this leaves you wondering what good feedback actually looks like, the more personal, everyday kind that doesn’t need an annual cycle or an anonymous form, that’s exactly what we cover in our Delivering Effective Feedback workshop. Full details are here.
Tenerife, the power gradient, and the calculus of voice
On 27 March 1977, two Boeing 747s collided on a runway in the Canary Islands, and 583 people died. It is still the worst disaster in the history of civil aviation, and it happened on the ground, in fog, between two aircraft that weren’t actually supposed to be there.
Neither flight was scheduled to land at Los Rodeos. They were bound for Gran Canaria, until a separatist group detonated a bomb in the terminal there and a warning of a second one closed the airport. All flights headed to Gran Canaria were diverted, and many of them were sent to Los Rodeos on Tenerife: a small airport on a saddle between two mountains, with a single runway, a handful of taxiways, one apron, and a local reputation for being foggy. By the afternoon the apron was full of diverted widebodies, and the taxiways were blocked by the aircraft parked on them. When the all-clear finally came, the planes that wanted to leave could not taxi around to the runway in the ordinary way, because there was no room. They had to taxi down the active runway itself, turn around at the far end, and take off back the way they had come.
By Mtcv for Dutch wikipedia. CC BY-SA 3.0, https://commons.wikimedia.org/w/index.php?curid=1091580
Speaking up on the flight deck
The KLM 747 was piloted by Captain Jacob van Zanten, with First Officer Klaas Meurs beside him and Flight Engineer Willem Schreuder behind. Van Zanten taxied the length of the runway, turned the aircraft 180 degrees, and got ready to depart. Behind them, the Pan Am 747 was still taxiing down the same runway, under instruction to turn off at the third exit.
Van Zanten advanced the throttles. Meurs, somewhat surprised, told him they did not yet have ATC (air traffic control) clearance. Van Zanten brought the power back and, by the report’s account, told him to go ahead and ask for it. Meurs did speak up, and van Zanten did respond.
The ATC tower then passed KLM a route clearance, the instructions for after departure, and it contained the word “takeoff” without actually being a clearance to take off. Meurs read it back and tailed off with the words “we are now at takeoff.” Van Zanten, already releasing the brakes, said two words over the top of him: “We’re going.” The controller, who could not see the runway, answered “OK,” and then, after a pause, “stand by for takeoff, I will call you.” At that exact moment the Pan Am crew keyed their microphone to say they were still on the runway. The two radio transmissions collided, producing a squeal that meant neither were audible. The only word that reached the KLM cockpit cleanly was “OK.”
The tower then told the Pan Am to report when it was clear of the runway, and Pan Am acknowledged. This much was audible in the KLM cockpit, and Schreuder heard it. He then asked the question that should have prevented disaster: “Is he not clear, that Pan American?” Van Zanten answered, emphatically, “Oh, yes.” Schreuder asked again. He was answered again, with the same certainty, and Van Zanten increased the throttle, accelerating the 747 along the runway.
And then both Schreuder and Meurs were quiet. For the ten or fifteen seconds that remained they said nothing, until the fog ahead cleared, at about 160 miles an hour, showing the Pan Am 747 directly in their path, pointing straight at them. Both crews tried to take evasive action. The Pan Am hauled left toward the grass verge beside the runway; whilst van Zanten pulled back hard enough to drag the tail along the tarmac, trying to take off straight up and over the Pan Am. However, Van Zanten had decided to refuel during the long wait, sensibly enough, and combined with the full complement of passengers and luggage, it meant the KLM was heavy, and they didn’t quite make it. The wheels and one engine of the KLM ripped into the Pan Am, and the KLM aircraft hit the ground at speed. The collision killed everyone aboard the KLM and most of those aboard the Pan Am. 583 people died as a result.
Wreckage on the runway of Los Rodeos after the Tenerife airport disaster of March 27, 1977
“But if it really matters, won’t people speak up?”
If we take one thing from this, it is this: people do not speak up with their concerns, and they especially do not speak up again once they have spoken up and been dismissed, even when their own lives and the lives of others are at stake. The interpersonal costs and risks of speaking up loom so large that they outweigh the benefits of doing so.
Willem Schreuder was not junior or inexperienced. He had more flying hours than either of the men in cockpit with him. He had seen the problem and named it, twice. And then, having watched his certainty bounce off the captain’s certainty, he stopped, in an aircraft accelerating along a runway he had good reason to think was occupied by another plane. This is not cowardice or stupidity. It is what the interpersonal cost of voice does to all of us.
Consider the tacit calculus of voice, combined with Prospect Theory: we weigh the almost certain cost of speaking up against a merely possible benefit. The cost of pressing the captain a third time is immediate, social, and certain. You will be the person who second-guessed the chief instructor, out loud, on the basis of a feeling, and if the runway turns out to be clear you will have aborted his takeoff for nothing. The benefit is probabilistic and invisible if it works: a collision that does not happen leaves no trace and earns no credit. We all tend to over-estimate the immediate costs of speaking up, especially against someone senior, so the calculus tips towards the negative and we remain silent. This is all very human.
And notice the form the doubt took. Schreuder did not say “there is an aircraft on that runway, stop.” He asked whether the Pan Am was clear. This isn’t incidental: crews, and especially the less senior, tend to raise concerns in mitigated forms, as hints and questions rather than statements and commands, and the softer the form, the easier it is to wave away (Fischer and Orasanu, 1999). The same calculus that pushes us toward silence also shapes how we say something.
Here’s the really uncomfortable bit though – if the calculus still tips toward silence when the downside is death, in an aircraft, with your own life in the balance, then we should not be remotely surprised that it tips toward silence in a marketing meeting, an ideation session, or a finance catch-up. In most of the meetings and conversations we have, nobody is about to die. The high stakes that might justify the risk of speaking are not even close to this example, which means the silence makes even more sense. Tenerife is not a rare case at the edge of human behaviour. It’s an ordinary case, but with a tragic outcome.
The Power Gradient
The second thing this disaster surfaced, along with the others that the human-factors investigators were studying around the same time, was the gap in power between the most powerful person in the space, and the least. The aviation safety literature of the period called it various things, the authority gradient, the cockpit gradient, the trans-cockpit authority gradient, the power gradient. They are all pointing at the same structural feature: the steeper the difference in power between the person at the top and the person who has something to say, the harder it is to say it. This isn’t only intuitive: it shows up in the data on how status differences shape cockpit communication and coordination (Milanovich et al., 1998). The steeper the gradient, the more that voice is suppressed – it’s much harder to speak up against someone with greater power – power over your future, your reputation, your job, or your status. This is in fact, one of the most important things, perhaps the most important thing, governing whether people speak up.
“Power” is a large and ambiguous word though, so it helps to break it down. Drawing on French and Raven, but trying to make it usable rather than academic, we can talk about four kinds of power.
Formal power (positional power) is the obvious kind, the kind that is described and written down. Role, rank, title, seniority; where we sit in the org chart, how many people report to us, who we report to, which rules and structures are ours to invoke. It is the most legible form of power, and being legible, it is also the easiest to see and to navigate around.
Informal power is much harder to see, because it is almost never written down. It lives in social standing, reputation, popularity, the size and shape of our networks, our reputation, how many people we know and how well. It does its work invisibly, which makes it harder to name and harder to navigate.
Demographic power is mostly the power we did nothing to earn and cannot hand back: height, race, age, gender, sexuality, class, accent. It endows, or otherwise, different degrees of power in different contexts.
Expert power is what we know: our qualifications, our experience, our demonstrated competence in this domain. It is sharply contextual. A brain surgeon in a cockpit has very little of it (probably, maybe they’re also a qualified pilot). An experienced surgeon in an operating theatre has a lot of it.
The Four Types of Power
The point of separating these out is that they do not behave as one thing. They can pull against each other, they can compensate for each other’s absence, and, dangerously, they can stack up.
In van Zanten they all stacked up. His informal power was enormous: he was the face of KLM’s advertising, the man in the in-flight magazine, photographed under the line about the people who made punctuality possible. He was, at least in the world of the airline, a minor celebrity. His demographic power was high as a tall, tanned, silver-haired Dutchman in a captain’s uniform. His expert power was as high as it goes: he was KLM’s chief instructor on the 747, the man who taught other pilots how to fly the aircraft, the airline’s recognised authority on the type. And his formal power matched it, the seniority and standing of the chief of flight training, the man KLM would later try to reach to help investigate the crash, before they realised that he had been at the controls. He had also, only weeks earlier, conducted Meurs’s qualification check on the 747. The first officer beside him owed his certification on the aircraft to the man he was trying to challenge.
Set that out and the gradient in that cockpit is incredibly steep, along every axis.
It is worth being careful here, because the easy version of this story makes van Zanten a powerful tyrant, and the evidence doesn’t support it; in fact rather the opposite. Colleagues who knew him, including Jan Bartelski, a fellow KLM captain and later president of the international pilots’ federation, dispute that portrait flatly. By their account he was studious and introverted but warm, a believer in partnership who insisted his first officers call him “Jaap” rather than “Captain.” This doesn’t weaken the argument; in fact it reinforces it. The power gradient did its silencing work in the cockpit of a man who, by temperament and by stated belief, was trying to flatten it. The gradient is not mainly a matter of personality. It forms between people, out of their positions relative to one another: rank, reputation, expertise, and how readily the room defers. None of that dissolves because the person at the top is, personally, a good sort. We might work to mitigate the gradient, and we should; but the gradient remains.
None of which is to say it was his fault, or that having power is something to feel guilty about. We may possess a great deal of power and that is fine; most of us will, in some room at some time, be the most powerful person in it. What we are obliged to do is recognise and acknowledge it, because the gradient forms whether or not we want it to, and it has effects whether or not we intend them. And the gradient acts on the powerful, too. A reputation built entirely on being on time every time is a reputation with a downside, and a man already badly delayed, watching the duty clock run down, was under a pressure of his own making to go. The punctuality that sold his reputation was pressing on the throttle in his hand on the runway.
What we learned from the wreckage
The lasting answer to Tenerife, and to the accidents around it, was Crew Resource Management. It came out of a NASA workshop on human factors in 1979, which itself was catalysed by the Tenerife disaster, along with others, and was first adopted comprehensively by United Airlines in the early 1980s, and was taken up by nearly every airline in the world within a remarkably short time. It is, with some justification, often called the most successful safety programme humanity has ever created. It is a large part of the reason that getting into an aeroplane is now one of the safest things you can do.
CRM emerged in aviation specifically to reduce catastrophic errors by changing cockpit culture: communication norms, authority gradients, briefing practices, and coordination under pressure. It was quickly adopted in medicine, especially in emergency response and (surprise) surgical teams. It trains people to speak up when something looks wrong, to challenge respectfully, to cross-check decisions, and to share situational awareness. In doing so, it creates a structured environment in which speaking up becomes expected and legitimate (Weller, Boyd and Cumin, 2014). The sense that it is safe to raise a concern is not a mysterious precondition that must somehow appear in advance. It is the result of concrete practices, training, and norms that make voice a normal part of the work.
It began life as Cockpit Resource Management and became Crew Resource Management, as a result of further disaster and our learning from them. Accidents like Kegworth, where cabin crew outside the cockpit could see things the pilots could not and the information never made the journey forward – due to a steep power gradient between cockpit crew and cabin crew, taught the industry that it is not enough to train only the people at the controls. Cabin crew, ground crew, air traffic control, operations staff: everyone in the system needs to be able to speak up to power at the moment it matters, and everyone needs the people above them to listen when they do. The safest crews are the ones where leadership is, in effect, shared: cabin and cockpit prompting and correcting one another rather than information travelling in one direction (Bienefeld and Grote, 2014).
CRM works on the problem from both ends. It scaffolds the act of speaking up against a gradient, giving people recognised, legitimate ways to push information upwards. And it works on the gradients themselves, building in practices that share power and distribute authority so that the slope is shallower before anyone has to climb it. Both are necessary. We cannot only teach the person at the bottom to be braver; we also have to lower the wall. None of this is a switch we flip once. Safety silence persists even in trained crews wherever the local culture quietly licenses it, which is why CRM has to be lived rather than merely delivered (Perkins et al., 2022).
Walking ourselves up the ladder
One of the practices that came out of this world is PACE, a graded assertiveness tool, and it is worth walking through because it is a scaffold built precisely for the moment Schreuder found himself in.
We begin with a Probe: a low-threat question that surfaces the concern without confronting anyone. “Is that red light meant to be on?”, “That Pan Am, is he clear yet?” Schreuder’s question was, in effect, a probe. If the probe does not get traction, if it doesn’t draw attention to the thing we are actually concerned about, we escalate to an Alert, which names the hazard clearly. “Oxygen is at 90% and falling.” or “We don’t have ATC clearance for takeoff.” Meurs’s first objection was an alert. If the alert doesn’t land either, we go to Challenge: stating the problem and proposing a different course of action, which is much harder, because now we‘re not just raising a worry, we’re asking the powerful person to do something other than what they have decided. “I don’t think it’s safe to take off yet. There may be an aircraft on the runway. Let’s hold and check.” And if even that fails, we reach Emergency, an unambiguous command: “Stop. This is unsafe.” This is very interpersonally challenging and risky, but is less so if we’ve walked ourselves up the PACE steps already – and hopefully we don’t need to get to Emergency anyway.
At Tenerife, nobody got past Alert. Two experienced professionals raised the hazard, had it dismissed, and stopped where the ladder becomes more interpersonally expensive.
We see the same pattern far outside aviation. Elaine Bromiley died in 2005 during what should have been a routine operation, when her anaesthetists became fixated on an airway they could not secure and lost track of the time she had spent without oxygen. The nurses in the room could see what was happening. One of them had fetched the kit for an emergency surgical airway and had it ready. They did not speak up and directly challenge the anaesthetists however, because the gradient between nurse and consultant in that theatre was every bit as steep as the one in the KLM cockpit, and the calculus of voice tipped the same way it always does (Harmer, 2005). Her husband Martin, an airline pilot, went on to found the Clinical Human Factors Group, and has spent the years since bringing aviation’s hard-won human-factors lessons into healthcare.
The reason the ladder helps is not that the top rung becomes comfortable. Issuing an unambiguous command to someone who holds your future in their hands is never going to feel safe, and it is not supposed to. The point is that you very rarely have to start there. By beginning with a probe and escalating only as far as you need, you walk yourself up the slope one manageable step at a time, and most of the time the problem is resolved long before you reach the top. The ladder turns one impossible act into a sequence of difficult but possible ones. And it gives the people around you, the ones with the power, a recognised signal that something is wrong before it becomes a fight.
What Tenerife taught us
We have learned an enormous amount from this disaster, as we have from many others. The conditions that produced the silence in that cockpit are not rare or unique to aviation. A steep gradient, a respected expert who has just expressed certainty, a team under time pressure, a concern that gets raised and dismissed once: this isn’t just a description of a 1977 runway. It is a description of an ordinary Tuesday in a finance team, a weekly planning meeting with the boss, or any number of other, comparatively mundane, contexts (with the merciful difference that the stakes are usually lower). The silence itself is not a quirk of individuals but a collective condition, one that organisations produce and reproduce in themselves (Morrison and Milliken, 2000).
That difference is also the danger. Because the stakes are lower, the silence is cheaper, and the scaffolding that aviation was forced to build under the pressure of mass fatalities is scaffolding the rest of us never get round to building at all. If two skilled people would not press a third time with their own lives on the line, we cannot reasonably expect anyone to press in a meeting where the worst case is mild embarrassment, unless we have deliberately made the conditions possible.
Jan Hagen’s Confronting Mistakesis a great book to read alongside this, because it works through the disasters of these decades and asks, of each one, why it happened, what might have prevented it, and what we learned from it. For the fullest blow-by-blow reconstruction of the accident itself, Admiral Cloudberg’s account is the best there is (Cloudberg, 2022). Underneath nearly all of them sits the same thing the human-factors investigators surfaced at Tenerife, decades before psychological safety became a term anyone outside a few corners of academia would recognise. They weren’t using the term psychological safety (because it wasn’t yet in mainstream use), but they were describing the exact same thing. The gradient suppresses voice; the silence risks disaster; and the work is to build the conditions, structural as well as interpersonal, in which the least powerful person in the room can say the thing that needs saying while there is still time to act on it.
Van Zanten’s last two words on the matter were “We’re going.” Everyone else acquiesced.
References
Bienefeld, N. and Grote, G. (2014) ‘Shared leadership in multiteam systems: How cockpit and cabin crews lead each other to safety’, Human Factors, 56(2), pp. 270–286.
Fischer, U. and Orasanu, J. (1999) ‘Cultural diversity and crew communication’. Paper presented at the 50th International Astronautical Congress, Amsterdam, October 1999.
Milanovich, D.M., Driskell, J.E., Stout, R.J. and Salas, E. (1998) ‘Status and cockpit dynamics: A review and empirical study’, Group Dynamics: Theory, Research, and Practice, 2(3), pp. 155–167.
Morrison, E.W. and Milliken, F.J. (2000) ‘Organizational silence: A barrier to change and development in a pluralistic world’, Academy of Management Review, 25(4), pp. 706–725.
Perkins, K., Ghosh, S., Vera, J., Aragon, C. and Hyland, A. (2022) ‘The persistence of safety silence: How flight deck microcultures influence the efficacy of crew resource management’, International Journal of Aviation, Aeronautics, and Aerospace, 9(3).
Weller, J., Boyd, M. and Cumin, D. (2014) ‘Teams, tribes and patient safety: Overcoming barriers to effective teamwork in healthcare’, Postgraduate Medical Journal, 90(1061), pp. 149–154.
Reading the air: high and low context communication in teams
Picture the scene: you’re in a meeting with your new team. You ask what you think is a reasonable question, and knowing glances are exchanged across the room. A couple of people suppress laughter, and the team lead moves on. It’s clear that you’ve said something wrong, but you have no idea what. The rest of the team are picking up on, and sharing, cues you can’t read and you are suddenly and obviously the outsider. There is an implicit communication at play, and you’re not in on it.
This dynamic can be thought of in terms of the context of the communication. The rest of the team are operating with a high level of shared context – they are reading something into your words, and implicitly communicating something back to you, but you don’t have the shared context to understand what that is. This is a concept that anthropologist Edward Hall identified through his work which began in the 1930s, exploring how different groups of people communicate in different ways with their own shared norms, including how people who have known each other for a long time often communicate more implicitly than strangers who have just met. He later came to term this ‘high and low context’ communication, a framing which Erin Meyer brought to a wider organisational audience through her book The Culture Map. The idea of high and low context communication is a framework that has a lot to offer anyone thinking about how teams actually function.
High and low context: what the framework says
In high context communication, a great deal of meaning is carried implicitly. It lives in relationship history, shared understanding, tone, non-verbal signals and what is deliberately left unsaid. The speaker may be saying one thing explicitly while something else entirely is being communicated, and the onus is on the listener to read between the lines. Japan is often cited as a key example of a high context culture; Meyer describes how someone who fails to pick up on implicit cues risks being labelled a kuuki yomenai – literally, “a person who can’t read the air.” Korea, Indonesia and Iran are also broadly high context cultures.
High context communication tends to flourish in societies with long, dense shared histories – populations who have had many centuries, sometimes even millennia, to develop layered, sophisticated codes of meaning and to become fluent in interpreting them with each other.
Read the air: Mammatus clouds form when cold, sinking air (laden with ice crystals or water droplets) descends into warmer, drier air below, creating distinctive rounded lobes. They’re usually found on the underside of cumulonimbus anvils: the spreading top of a large thunderstorm
Low context communication works differently. Meaning is made as explicit and external as possible – it lives in the words themselves, in clear and precise instructions and in expectations that are named directly rather than gestured toward. The US is Meyer’s go-to example of a very low context culture. She explains that in the US, a country shaped by successive waves of immigration, with diverse populations bringing different languages, histories and backgrounds, people quickly learned that if you wanted your message to land, you had to make it as explicit as possible. Australia, Canada and the Netherlands sit similarly on the low context end of the spectrum.
But the framework isn’t only about national culture, it cuts across many kinds of grouping. A tight-knit friendship group who have grown up together will often have developed a highly implicit, almost private communication style that is opaque to anyone outside it. A group of students thrown together in university halls of residence for the first time will need to be much more explicit if they’re going to organise anything at all. The same applies in organisations. A small, long-established, relatively autonomous team will often communicate in a higher context style than a newly convened multidisciplinary group drawn from across different functions.
The problem with high context in diverse teams
And here is where we go beyond something that’s academically kind of interesting into something that’s really important for the way we work…
High context environments favour insiders. They require a depth of shared knowledge and reference that takes time and proximity to build. Anyone who doesn’t share that implicit frame – for instance new joiners, people from different cultural backgrounds, neurodivergent folks, people from outside the dominant professional culture – is operating with less information, greater ambiguity and a greater risk of getting things wrong. They’re having to work harder just to follow what everyone else seems to be understanding effortlessly. And they’re typically far less likely to speak up when they’re uncertain, because, as we know, uncertainty makes speaking up feel more risky.
This is where high context communication becomes an equity issue, not just a style preference. The implicit meanings, unspoken norms and in-jokes that mark the insiders all create gradients of access to conversation that track closely with who was there first, who shares the dominant cultural background and whose communication style is treated as the default.
This connects directly to how teams foster psychological safety. When norms are implicit, expectations are gestured at rather than named and people are expected to absorb the rules by osmosis, the barrier to speaking up rises sharply, especially for those newest to the group or furthest from the implicit reference frame. When we make our communication norms explicit, we widen the circle of people who can understand what kind of participation is welcome and how to make their voice heard.
Multidisciplinary teams
The challenge of communicating across contexts compounds when you bring together people not just from different cultural or social backgrounds but from different professional disciplines. A product manager, a clinician, a data scientist, a policy specialist and a sales lead have all been shaped by the norms, language and implicit assumptions of their respective fields. They each bring a professional context and assumed way of working and communicating that the others aren’t party to. Perhaps the clinician uses clinical terminology, which they assume everyone else will understand, while the policy specialist has a knowledge of the legislative landscape that underpins all that they say but is never made explicit. Perhaps the sales lead starts from the assumption that the team’s goal is purely commercial. None of them are wrong to carry their expertise, and in fact it’s the diversity of their backgrounds that could allow the work they do together to be hugely successful. But when their contextual assumptions go unexamined, misunderstandings and misalignments proliferate.
This is why taking some time to get to know one another’s ways of working, and, even more importantly, to establish some explicit, shared team norms at the outset of a multidisciplinary project matters so much. What this means in practice is a deliberate, intentional move to lower context communication – being as explicit and clear with each other as possible. All the dialogue that flows from there can be built on shared understanding.
When the stakes are highest: crisis and time pressure
The case for practising low context communication in ordinary conditions becomes even harder to ignore when you consider what happens when those conditions break down.
Under pressure, in a crisis, where decision making is time pressured but the stakes are high, we might feel as if we need to communicate more implicitly in order to go faster, but actually the time cost of ambiguity and flawed assumptions can be even higher.
In their work on ‘Crisis-Ready Teams’ Waller and Kaplan described the characteristics of high-performing teams in terms of their early interaction patterns, and discovered that higher performing teams spent significantly more time in what they called the early structuring phase, where they clarified team member’s roles and created ground rules around communicating – what we would call the social contract. They observed:
“While other teams seemed to jump into the problem feetfirst with very little effort given to structuring activities, high performing teams seemed to approach the crisis situation knowing if all team members weren’t on the ‘same page’ from the beginning, coordination would become more problematic as things became more complex.”
What these high performing teams were effectively doing was taking the time to build shared mental models of their work environment and how they could work together. They were explicitly building shared context. What’s more, these high-performing crisis teams also spent more time in what Waller and Kaplan call the information sharing phase – where they pool the information they have individually with each other to augment each other’s understanding. And the authors are keen to emphasise that this doesn’t mean pausing and deliberating before you ‘stop the bleeding’ or take any immediate actions needed as part of the first response phase of a crisis. But it does mean that as we tackle complex and evolving crises, we need teams who can continually monitor the situation, “freely sharing information about the situation without fear of reprisal,” as Waller and Kaplan put it. In other words, the teams need psychological safety.
Making low context communication an explicit group norm in ordinary conditions is partly preparation for these times. Teams that have practised stating expectations clearly and treating explicit communication as a mark of respect rather than an insult to intelligence are much better placed to communicate effectively when clarity becomes critical.
Building the habit: low context as a team norm
In practice this means taking time to build a social contract deliberately, and discussing the value of low context communication with the team from the start, not assuming that clarity will emerge naturally, but treating it as something the group actively commits to.
A happy outcome of this, of course, is that the more we work on making expectations explicit, encouraging information sharing and getting clear about our roles, the more we are actually building a shared context. In time that can mean more of our coordination becomes implicit again, the team reading each other well without needing to spell everything out. The difference is that this is an implicit understanding the group has built, rather than one it took for granted at the start.
A note on the costs
So does this mean the real goal was never low context communication at all, only creating the shared context that lets us work implicitly again? Not quite. In most teams the conditions for that never fully stabilise. New joiners arrive, team boundaries shift and the shared reference frame that makes implicit working reliable is always partly under construction. Diversity of background, discipline and cognitive approach adds to this: the assumptions any group builds will always be more accessible to some members than others. As Meyer puts it, “Multicultural teams need low-context processes” – and we would extend that further: so do teams with diversity of any kind, whether of discipline, background, neurodiversity or anything else.
But it is worth acknowledging that low context communication is not a universal good that everyone will welcome naturally. For those accustomed to high context styles, a move toward explicitness can feel slow, or even patronising: “You don’t need to spell it out – I get it!”.
There is a risk that imposing low context communication without acknowledging this discomfort lands as a power move in its own right, as if the explicit style were simply better, rather than better suited to working with team members who come from different contexts and have different histories. As Meyer describes, we might need to be really honest and upfront about why we’re communicating like this, saying things like:
I’ll put what we discussed and agreed in writing, not because we don’t trust each other but because then we have a clear agreement we can refer back to and catch any misunderstandings or confusions.
We can apply the principles of low context communication even to the process of moving to lower context approaches!
What a team is really developing is the capacity to move between the two styles deliberately: to grow enough shared context that it can work implicitly where that genuinely serves the group, while maintaining the discipline to switch to something more explicit whenever the situation demands it (which it often will). The way in is an honest conversation about how we communicate, and that conversation is itself an act of low context communication. We start by naming what we’re doing and why.
Who gets to decide if psychological safety matters?
Is scepticism about psychological safety a luxury position? What the LinkedIn discourse reveals about who has positive or negative opinions on psychological safety, and why it might matter more than we think.
Psychological safety is one of the most discussed (and critiqued) concepts in the workplace. It generates a vast amount of research papers, conference talks, podcasts, blogs and videos, and an almost continuous stream of LinkedIn posts. But the conversation about psychological safety is also a thing in its own right, distinct from the concept itself. So we wanted to ask who is actually having it, and what do they think?
To do this, we carried out a small exploratory study into the LinkedIn discourse around psychological safety: who is talking about it, what they are saying, and whether patterns emerge around gender and organisational position.
Executive summary
This exploratory study analysed sentiment in 14 LinkedIn threads discussing psychological safety, covering approximately 170 substantive comments from around 200 unique authors across 12+ countries. We coded sentiment and inferred demographics from names, profile photographs, and job titles.
Three distinct discourse communities emerged: an advocacy stratum, an academic-critical stratum, and an inner circle stratum.
The data directionally supports a privilege hypothesis: scepticism about psychological safety appears to correlate with societal advantage. In regard to professional status, senior academics and executives are the most negative groups; clinical and operational safety professionals are among the most positive. With regard to gender, women are substantially more positive (55%) and less negative (9%) than men (42% positive, 24% negative). The single most over-represented group among negative voices is senior men, who account for 46% of negative comments despite being a much smaller share of the dataset overall.
The most significant finding may be structural: the people most likely to benefit from psychological safety (junior employees, frontline workers, those in genuinely unsafe workplaces) are almost entirely absent from the LinkedIn discourse. The conversation is conducted among the relatively privileged, largely in the absence of the people it is ostensibly about.
Methodology
We manually collected 14 LinkedIn threads in which psychological safety was discussed substantively through March 2026 – the criteria were that there had to be more than 15 comments on it, more than a day old at the time of collecting them, and the posts were not explicitly promotional or commercial. The Linkedin threads were selected to represent a range of post authors, sectors, and audience types. This gave us the best chances of analysing the most substantive posts.
Some important context to add. I’m a practitioner, researcher and facilitator in the psychological safety field, with an active (though somewhat reluctant) LinkedIn presence including in threads of the type analysed here. I am, in other words, part of what I’m describing. That seemed worth saying out loud: who gets to write a piece about who is having the conversation is itself a question about who is having the conversation. Also, as a practitioner in this space, I sit in the “advocacy” stratum, and by my own hypothesis, that is precisely the stratum most likely to hold positive views about psychological safety. To manage that bias where possible, my own comments have been excluded from the dataset.
Across the 14 threads we collected approximately 320 comments, of which around 170 were substantive (through excluding tag-only entries, hashtag-only comments, and too-brief affirmations) from around 200 unique authors (after excluding commercial actors – a small number of individuals who used comment threads to promote their own products or services rather than engage substantively). One commercial actor appeared across five separate threads with near-identical promotional content. This itself is an observation that says something about the commercial ecosystem that has grown around psychological safety.
We used Linkedin names, profile photographs, and profile text to infer the various demographics that we’re interested in – gender, seniority and job function. Gender presentation was inferred from typical first name conventions, and profile photos were used to increase confidence. We wanted to look at ethnicity too, and tried name origin as a proxy. However it was, frankly, too crude, even if only treated as directional. So we left it out of the findings. Seniority and function were established from LinkedIn headlines and job titles. We coded comment sentiment into five categories: positive, mixed, negative, contextualising, and affirmation only. Comment sentiment was coded using an AI assistant (Claude, Anthropic), with categories reviewed and verified by the team.
There are some key limitations, of course: LinkedIn itself is a platform of the already relatively privileged. The most marginalised voices (junior employees, frontline and gig-economy workers, manual workers and tradespeople, for example) are much less likely to be on LinkedIn discussing the concept. This is not an incidental limitation. It is, in fact, part of the finding.
Three distinct discourse communities
One of the clearest patterns to emerge was that the LinkedIn psychological safety conversation is not one conversation. It is at least three almost distinct conversations running in parallel, with limited crossover.
The first is an advocacy stratum: these are posts by practitioners, coaches, healthcare workers, HR professionals, organisational leaders and more who are broadly positive about psychological safety and engage with fostering psychological safety as a practical concern. The post authors were predominantly women, and generated little negative sentiment in the comments: negative sentiment was only around 3% of organic comments across these advocacy threads.
The second is an academic-critical stratum: posts by researchers, scholars, and subject matter experts who engage critically with psychological safety – its evidence base, construct validity, and the various claims made about it. This stratum generated a high proportion of mixed and contextualising sentiment, suggesting genuine intellectual engagement and interest in the concept. It also contained the most methodologically rigorous engagement with the concept in the dataset, and several of the critiques raised deserve serious attention. It was also the most negative stratum in the dataset, with approximately 38% of comments expressing critical or negative views. The commenters were predominantly men.
The third is an inner circle stratum: posts by and around the psychological safety field’s most prominent figures: people who have written books and built professional identities around the concept. This stratum, predictably, was almost entirely positive. It generated the highest engagement figures and the least substantive debate. When a concept reaches this mature stage of its life, much of the conversation around it largely functions as a community of agreement, which is neither sinister nor surprising. It is, however, worth noticing.
Three strata, three different conversations, with surprisingly little crossover between them. The advocacy stratum generates warmth, support and agreement. The academic-critical stratum generates intellectual friction. The inner circle generates celebration. It’s the same concept with three almost completely separate conversations.
It’s worth noting that these strata are descriptive rather than fixed categories – individuals move between them, and occupy more than one at once. We use the framing as a way of seeing the discourse, not as a claim about stable groupings of people.
The privilege hypothesis
Our core research question was whether patterns exist linking gender, ethnicity, and position in the organisational hierarchy (a proxy of positional/formal power) to sentiment about psychological safety. We had a working hypothesis that we considered this data might disprove:
People who are most sceptical of PS tend to be those for whom speaking up has been, or is, less costly.
This may be (typically, but not always) men, people from non-marginalised backgrounds, those in secure, senior or tenured positions, and people operating nearer the “blunt” end of organisations rather than the “sharp” end.
The data supports this hypothesis directionally, with some caveats, namely that the demographic categories were inferred rather than self-reported, the sample is small enough that subgroup percentages should be read as indicative rather than precise, and LinkedIn is itself a platform of the relatively privileged – which means the full range of the pattern we’re describing is likely underrepresented in the data rather than overstated. With that in mind, the findings were as follows.
On gender, the pattern was a consistent finding in the dataset. Women were substantially and consistently more positive about psychological safety (approximately 55% positive) and substantially less negative (approximately 9% negative) than men (approximately 42% positive, 24% negative). This gap was widest in the academic-critical stratum, where men were around twice as likely as women to express outright critical or negative views about psychological safety. This is an interesting pattern that supports the privilege hypothesis.
Figure 1: Sentiment by gender. Women and men are roughly equally represented in mixed and contextualising comments; the gap appears almost entirely at the two ends, with women leaning positive and men carrying most of the negative sentiment.
On seniority and positional power, the pattern was the most striking. Using a blunt-end to sharp-end framing borrowed from Human and Organisational Performance (where blunt end describes those furthest from the work, and sharp end those closest) the gradient was very clear. Academics were the most negative group in the dataset at approximately 48% negative. Founders and C-suite executives came next at approximately 20%. In contrast, workers closer to the “Sharp End”, such as clinical NHS staff, operational safety professionals, and frontline practitioners were among the least negative at approximately 6%.
Figure 2: Sentiment by organisational position. The gradient runs the way the privilege hypothesis predicts: positivity rises and negativity falls as we move from blunt end towards sharp end.
At least on Linkedin, it appears that the people who are most sceptical of PS tend to be those with the most institutional power. The people who appear most enthusiastic for it are those whose working lives most regularly involve the difficult and precarious dynamics that a lack of psychological safety describes: very hierarchical environments (where they are typically lower) and less agency in rapidly changing, complex, sometimes high-risk, environments.
It is worth acknowledging one obvious objection directly. Is it possible that the negative sentiment groups are simply more knowledgeable about psychological safety, and that their scepticism reflects greater expertise rather than greater privilege? The data can’t rule this out. However, the advocacy stratum contains clinical psychologists, safety scientists, NHS consultants, and human factors specialists: people with deep professional knowledge of the conditions that psychological safety describes. The academic-critical stratum contains organisational psychologists and management researchers whose expertise is in the evidence base, though not necessarily in the lived experience of the concept. These are different kinds of knowledge, and neither outranks the other. A surgeon who has spent twenty years navigating the dynamics of clinical hierarchy understands something about psychological safety that a management professor who has read every paper on it may not. A nurse who has seen a colleague punished for raising a concern knows a lot that a think-piece cannot capture.
The hypothesis is not that critics are ignorant. The hypothesis is that those who engage most critically with psychological safety tend to be people whose professional lives have not required them to test the concept against personal cost – that this shapes which questions feel urgent to them, and which feel abstract. For the people at the sharp end, psychological safety isn’t abstract, it’s very real, and incredibly important.
Figure 3: The composition of positive and negative voices, broken down by gender and seniority. Broadly, women and non-senior people are more likely to feel positively about psychological safety. Senior men make up 11% of positive comments and 46% of negative ones.
Who isn’t in the conversation
As we mentioned, there is also another angle that the data points towards.
The people that the hypothesis most predicts would be positive about psychological safety: junior employees, frontline workers, and people in genuinely unsafe workplaces, are broadly absent from this dataset. LinkedIn is by its nature a platform of the relatively privileged. A warehouse worker, in one of the threads, described starting a new job and being laid off seven days later. They had heard all the right language about psychological safety from their employer in those seven days and they were one of the very few genuinely front line voices in the entire dataset. Their working life is the kind of life this discourse is supposedly about, but their voice is the kind of voice the discourse mostly doesn’t contain. The people who get hurt by low psychological safety at work are not in this conversation whilst the people who write articles about it (like me) are. The lack of these voices is an important finding in itself.
The popular psychological safety discourse is a conversation among the relatively privileged, conducted almost entirely in the absence of the people it’s ostensibly about.
If scepticism about psychological safety is correlated with more rarely having needed it (as this data suggests), that is not an argument that psychological safety is unimportant, not real, not useful or is invalid. It’s an argument that those doing the most publicly sceptical work are operating from a position of structural advantage that may well be invisible to them.
Reading the pattern
The data shows that the distribution of sentiment across the LinkedIn psychological safety discourse isn’t random, and that the pattern of who feels positively about psychological safety and who does not correlates with structural characteristics in ways that are hard to explain without reference to lived experience. People who operate in environments where speaking up is potentially higher stakes, such as clinical staff, operational safety professionals, people further down organisational hierarchies, are more consistently positive about psychological safety than those who do not. That pattern is consistent across our Linkedin thread analysis, as well as our previous research on socioeconomic background and willingness to take interpersonal risks at work.
None of this is claiming that sceptics of psychological safety are wrong, or that their critiques lack substance. Some of the most intellectually serious engagement with psychological safety in our dataset came from the academic-critical stratum, and several of those critiques around measurement confounds, misinterpretation and misapplication, and the gap between some claims and evidence are valid critiques and deserve engagement. We often find ourselves saying the same thing as the critics.
Neither is it claiming that being senior and male makes someone’s views on psychological safety invalid.
The privilege hypothesis is one explanation, but not the only one. Several alternatives plausibly account for at least part of the pattern. Academia rewards critique, and the professional incentives of research careers may shape how scholars appear to engage with applied concepts. The advocacy and academic strata may also simply differ in their relationship to the concept: practitioners use psychological safety as a working tool, while academics tend more to study it as a body of literature, and people who use a concept practically tend to be more positive about it than people who study it in the abstract. Some scepticism is also likely an understandable reaction to concept dilution (due to the wellness industrial complex, the LinkedIn thought-leadership ecosystem, and the commercialisation of the term) rather than to the genuine underlying construct.
We don’t think any of these alternatives entirely displace the privilege hypothesis. The pattern we’ve identified is consistent with several explanations, not just one. What we are confident about is that there is a pattern worth paying attention to, and that the absence of frontline voices from the discourse is significant regardless of which causal story best explains the rest.
Scepticism about psychological safety, then, may be a luxury position. One can afford to question whether psychological safety matters when speaking up has rarely cost you much.
The question worth asking
What people say about psychological safety on LinkedIn tells us something, but certainly not everything, about what people experience in workplaces. In a subsequent piece, we will explore the different types of scepticism we surfaced: the various distinct critiques, their positions, intellectual grounds, and what psychological safety advocates might learn from them.
For now, the conversation about psychological safety is happening. This isn’t a debate about whether psychological safety is “real” or not – it’s about how much it matters to us. And maybe it matters most to those for whom the stakes are highest – to the people we are least likely to hear from. Which is, of course, the point.
“…the powerful play goes on, and you may contribute a verse.” Walt Whitman
This research was conducted using publicly available LinkedIn posts and comments. All individuals have been anonymised. Demographics were inferred and should be treated as directional only.
Accountability, quite literally, is the ability to give an account: account-ability. An account is a narrative – the telling of an honest story about what happened, why we did what we did, what was known at the time and what wasn’t, told from our particular position, and situated within the context of what was actually happening. To be accountable is to have the ability to give an account. This isn’t a wordplay or a clever rhetorical manoeuvre. It’s the simple etymology of the word – accountability is meant to be about our capacity for honest explanation. To be accountable is not the same as to be punishable.
But accountability has been colonised by blame. In the world of work, almost without us noticing, the meaning of accountability has been displaced by its near-opposite. The tell is the phrase “Someone must be held accountable” (or something similar), and what that usually means in practice is that someone should suffer consequences. It’s rarely used to mean that someone should be given the conditions in which they can honestly explain what happened, but that someone should bear the cost of it happening. Accountability has turned into blame with better marketing.
Roberts (1991) formalised accountability as a relational tension between an account giver and a recipient, and that framing is important. Accountability requires at least two parties, someone willing to give an account and someone capable of receiving it; it’s relational and dynamic. Philip Tetlock’s work on accountability and cognition shows why the receiving end is so consequential. What he found was interesting but unsurprising: when people know who they’re accountable to and can anticipate what that audience wants, the account they give is shaped accordingly – designed to satisfy the audience rather than give the truest description of events as they actually happened. Every performance review, every post-incident investigation, every board report is subject to this dynamic. The account we receive is the account that the teller believed the audience wanted.
Accountability as a construct has been studied extensively across public administration (Bovens), organisational behaviour (Tetlock, Frink & Ferris), philosophy (O’Neill), and safety science (Dekker), and what strikes anyone who reads across these fields is how rarely any of them mean the same thing by it. The kaleidoscopic definitional confusion is part of the core problem with accountability. Even the classic RACI matrix (the popular tool that gives the appearance of clarity while achieving the texture of a spreadsheet) gets accountability wrong. Much of that confusion is semantic and probably unavoidable, but some of it has structural consequences. The most significant conflation is between accountability and responsibility. Ieraci (2007, via Parris, 2025) draws the distinction clearly: responsibility involves doing, accountability involves reporting. Responsibility is about the work itself, and accountability is about the account of the work. They overlap and interact, but they are not the same thing, and treating them as synonyms has problematic consequences.
Misunderstanding Accountability
Accountability and responsibility have been used interchangeably for so long that even legislation gets it wrong. This isn’t just pedantry. The conflation matters because once accountability is understood to mean “being responsible for outcomes,” the logical next step is: if the outcome was bad, the accountable person is at fault. Consequence then follows automatically and the account (the narrative, the context, the conditions that were actually present) gets dismissed before it can be heard. And accountability for outcomes is, in any case, the wrong target. Outcomes are partly beyond any of our control. What we can reasonably ask people to account for is their reasoning, their decisions, and their actions, situated within their context. Not the outcome.
“People aren’t just people, they are people surrounded by circumstances.” ― Terry Pratchett
Karl Weick’s sensemaking work is useful here, and shows why this is even worse than it first appears. Accountability in organisations is almost always exercised retrospectively, after the outcome is already known, and hindsight does much of the work before the inquiry begins. The account becomes a post-hoc justification exercise rather than a genuine rendering – an attempt to prove innocence in a process that has already assumed guilt. Structural bias is at play, and the knowledge of the outcome contaminates the inquiry before it begins.
Parris’s example from maritime law shows both mechanisms operating at once. New Zealand’s Maritime Transport Act assigns responsibility to vessel Masters where it clearly means accountability in the liability sense, and the confusion has structural consequences for who gets blamed when things go wrong. The Master at the sharp end becomes legally exposed for outcomes shaped by systemic decisions made at the blunt end. The legislative conflation and the retrospective framing work together: context and conditions get set aside because the outcome is already known, and the law reaches for the nearest human rather than the most relevant cause.
“…they took the captain to court, even though there was not a thing I could have done to have stopped it, to have known about it, or anything… I was held accountable… But I wasn’t responsible.” Parris, 2025 pp.35
And here’s the crux: responsibility is actually closer to what organisations think they mean by accountability. Responsibility is about roles, obligations, and the act of doing. We can assign responsibility, but we cannot assign accountability – an account is given, not extracted, and every organisational system built around ‘holding people accountable’ rests on a category error with real-world consequences. If accountability cannot be assigned, only taken, then the pressure organisations apply in its name does not produce genuine accounts. It only produces performances of them.
Accountability without context is blame
When we strip the account (the narrative, the context, the local rationality of the person who was there) and retain only the assignment of fault, we haven’t done accountability. We’ve dressed something much older and more primitive in accountability’s clothes.
Blametropism is the organisational tendency to orient toward fault, the way a plant orients toward light, regardless of whether finding fault serves any operational (rather than emotional) use. The organisational or political apparatus of accountability (the investigation, the finding, the consequence) is seldom deployed to truly understand what happened, but instead to locate where the fault should land. The true account is rarely sought; what’s more often sought is a name, a goat that we can expel from the village, laden with our sins. Accountability-as-blame is blametropism in its business suit. It gives the impression of rigour while reliably producing the outcomes of blame.
Marilyn Paul’s systems framing captures the consequences well: blame generates fear, fear generates cover-up, cover-up degrades information quality, degraded information produces more errors, more errors generate more blame. The feedback cycle is self-sustaining and self-defeating, and as reliable as it is depressing. An organisation that uses accountability as a pressure tool destroys its own capacity to learn what’s really happening inside it.
When people learn that giving an account: an honest one, with all its uncertainty and complexity and acknowledgement of error, leads to punishment, they naturally stop giving honest accounts. Instead they give performed accounts – narratives designed to survive scrutiny rather than enable learning.
And the demand for a single account is itself part of the problem. Real situations are experienced differently from different positions within a system. The account of the nurse who administered the wrong medicine, the ward manager who created the rota, the software developer who designed the drug cabinet interface, and the executive who approved the staffing budget are all partial, all genuine, and all necessary. Accountability-as-blame doesn’t just suppress honest accounts. It demands that a complex, plural reality be rendered as a single story with a single author.
Take the cases of Hadiza Bawa-Garba and RaDonda Vaught: both clinicians who made errors in systemically compromised conditions, both of whom were honest about what happened, and both of whom were subsequently prosecuted. In both cases, the logic was that someone had to be held to account. Investigations were conducted, findings were made and severe consequences followed. What was not sought, in any meaningful sense, was a genuine account. Both clinicians were already carrying the burden of what happened before their prosecutions even began, and the prosecutions compounded an existing problem by converting private grief into public liability. The systemic conditions and the context behind the errors (staffing levels, equipment failures, normalised workarounds and more) were noted but set aside. The apparatus of “accountability” produced the outcome that blame always produces: a conclusion that foreclosed rather than opened inquiry.
The lesson that clinicians drew from both cases was not “be more careful.” It was “be less transparent.” That is what accountability without context reliably produces. Onora O’Neill saw the same mechanism from a different vantage point: systems designed to demonstrate transparency often produce performances of transparency rather than the thing itself. The safety audit, the Ofsted inspection and the compliance checklist all create the appearance of account-giving while systematically degrading the conditions in which genuine accounts are possible. The mechanism defeats its own purpose.
An organisation that punishes honest accounts will eventually receive only dishonest ones.
You can’t assign accountability, only take it
The moment an account is compelled, demanded, or extracted under threat of consequence, it ceases to be an account and becomes testimony. Modern societies do not treat confessions given under duress as genuine, and for good reason – what is said under compulsion is shaped more by the compulsion than by the truth. The same mechanism operates inside organisations. Forced accountability produces self-protection, not transparency. A performance of accountability, not learning. It’s worth noting in this light that we tend to call the most common formal accountability mechanism in organisations an “annual performance review”.
“Currently fashionable methods of accountability damage rather than repair trust.” – Onora O’Neill
A genuine account has to be a voluntary act. It requires the account-giver to be willing to be seen: to expose their reasoning, their uncertainty, their errors, and their context. That exposure carries real interpersonal and career risk, and where candour carries significant risk but little benefit, most of us will err on the side of caution and silence. This isn’t a character flaw, it’s a rational response to the environment, and changing that calculus is the only reliable route to better accounts.
The distance between the account an organisation says it wants and the account it actually rewards is navigated through social inference: reading the room, calibrating our degree of honesty to the audience, knowing instinctively how much truth is too much. It’s also a skill that is unevenly distributed in most organisations. For many neurodivergent people, candour is simply how to communicate, not a choice to be weighed against consequences. An organisation that performs a desire for honesty while punishing it in practice fails those workers and risks actively deceiving them.
Cultivating the conditions for honesty is considerably more difficult than demanding it, and considerably more effective. The phrase “holding people accountable” implies a power that organisations do not have. We cannot hold someone else accountable. We can only nurture the organisational substrate in which people are able and willing to give an honest account.
What real accountability looks like
Real accountability is dialogic rather than monologic. Roberts draws on Senge’s distinction between discussion and dialogue: discussion in this sense, is like a debate, with positions defended and a winner declared. Dialogue is different: meaning, nuance, context and information flow freely between parties, with the intention of surfacing something neither party could reach alone. Organisational accountability processes are almost universally structured as a defensive discussion – and a genuine account cannot be extracted through cross-examination. It can only emerge through a more open, reflective dialogue, which requires the account-giver to feel safe enough to think out loud, be uncertain, and say “I don’t know” without it becoming evidence against them. Both giving and receiving are equal components of accountability, and the receiving requires its own kind of discipline: the willingness to be changed by what we hear, rather than confirm what we already believed. Maybe the hardest part is the moment right after the event, when everyone is nervous and the pressure to find a name is enormous, and someone has to be the person who says: let’s slow down, let’s find out what actually happened.
Parris’s typology reinforces this binary. Backward/hierarchical accountability is structurally monologic: it already knows what it’s looking for. Forward/process accountability requires dialogue, because it’s genuinely interested in what the system contributed, which can only be discovered, rarely confirmed.
“But surely there have to be consequences sometimes?” Yes, but consequences and accountability are separate questions. We can impose consequences without ever actually receiving an honest account. And if negative consequences are attached to the account, we will never receive an honest one. Consequences cannot come first without contaminating everything that follows. The desire for consequence after failure is often a desire for justice, and that is a fundamental human impulse worth taking seriously. But retributive justice and learning are different projects, and the accountability mechanisms most organisations use are poorly designed for either.
The question after a failure, Dekker argues, is not just “who is accountable?” but (more importantly) “how do we learn?”, and answering the second question well requires the conditions that make genuine accounts possible. It follows that psychological safety and true accountability are mutually dependent. We cannot give a true account if we don’t feel safe, and accountability as punishment makes honesty unsafe. A genuine account shouldn’t feel like a confession. It should feel like a contribution to our collective understanding, and treating it as the former guarantees we will never receive it as the latter. Building the conditions in which genuine accounts are possible is what accountability, properly understood, actually requires.
Account-ability
In the vast majority of organisations, accountability is expected to flow upwards. The powerful receive accounts; whilst they are seldom required to honestly account for themselves to those ‘below’ them. And I’d argue that the most valuable accounts are actually those that flow sideways, worker to worker, where honesty isn’t calibrated to hierarchy.
We have collectively spent considerable energy perfecting the art of demanding honesty from the people for whom we have made honesty unsafe. “Accountability”, a word that belonged to everyone who needed to give an honest account, was appropriated by institutions that needed a more respectable name for blame. Blame is cognitively cheap, emotionally satisfying, and operationally useless. It produces a very clear answer to the question of what to do next, and it is almost always the wrong one.
However, blame is also often politically useful. Blame doesn’t just emotionally satisfy: it secures power. When the investigation concludes and the name is named, someone’s position is secured and someone else’s is eroded, or ended. The apparatus of “accountability” in this case is functioning exactly as designed, just for purposes that aren’t on the tin.
This is why genuine accountability remains so rare. It asks the powerful to submit to the same conditions they currently use accountability to avoid: to give an honest account of themselves, to be uncertain in public, and to say “I don’t know” to the people they lead. In cultures where blame secures position, a genuine account risks it. The rarity of genuine accountability in organisations is its rational consequence.
Many organisations “hold people accountable” by imposing negative consequences for failure. The organisations that actually handle failure best are the ones that make it safe to give an honest account – they foster account-ability. They are not the same thing. They are, in fact, complete opposites.
References
Bovens, M. (2007) ‘Analysing and assessing accountability: a conceptual framework’, European Law Journal, 13(4), pp. 447–468.
Dekker, S.W.A. (2012) Just Culture: Balancing Safety and Accountability. Farnham: Ashgate. (See also:Dekker (2011) in the map)
Frink, D.D. and Ferris, G.R. (1998) ‘Accountability, impression management, and goal setting in the performance evaluation process’, Human Relations, 51(10), pp. 1259–1283.
Ieraci, S. (2007) ‘Responsibility versus accountability in a risk-averse culture’, Emergency Medicine Australasia, 19(1), pp. 63–64.
O’Neill, O. (2002) A Question of Trust. Cambridge: Cambridge University Press. → Explore in map
Parris, R. (2025) Accountability: A Lived Experience. MSc thesis. Lund University. → Explore in map
Paul, M. (2017) ‘Moving from blame to accountability’, The Systems Thinker.
Roberts, J. (1991) ‘The possibilities of accountability’, Accounting, Organizations and Society, 16(4), pp. 355–368. → Explore in map
Roberts, J. (1996) ‘From discipline to dialogue: individualizing and socializing forms of accountability’, in Accountability: Power, Ethos and the Technologies of Managing, pp. 40–61.
Tetlock, P.E. (1985) ‘Accountability: the neglected social context of judgment and choice’, Research in Organizational Behavior, 7, pp. 297–332. → Explore in map
A few months back, I was preparing for a workshop with a client, and they shared, anonymised, their staff engagement survey data. We both sat on our respective screens, peering at a heat map of red, orange and the occasional shiny green rectangles, and I realised the data itself told me nothing useful for planning the session.
I could see from the data why they had identified psychological safety as an area for improvement, but beyond that the numbers themselves gave me no context, nuance or insight. So I started to ask questions – how long have the team been working together? What are their challenges? How do they communicate with each other? Really quickly, a picture started to take shape. The team had previously worked under a tyrannical manager who had since left, but in whose shadow they were still dwelling. Communication only happened in one-to-one conversations or private slack channels – the whole group chat was largely silent. Lack of clarity about what the team’s real purpose was meant that prioritisation was always contentious and conflict-ridden, and decisions were often undermined by other team members. From a relatively short conversation, I had a much stronger starting point for planning my session. I still didn’t know everything, of course, more always emerges in the workshop, but I had a much clearer picture of where to start.
Psychological safety is a felt phenomenon. It’s something we experience, in context, for reasons that are unique to us and the team. Yes, we might be able to measure headline figures about psychological safety, and we can even dig into some of its different ‘flavours’ – are people more likely to feel safe sharing an idea than asking a question for example? But we still don’t know why until we actually start talking to people.
There’s an assumption baked into a lot of workplaces that numbers are rigour and words are just impressions. People are often apologetic about offering their observations and might caveat them with, “This is just anecdotal,” or “Obviously this is only my take on it.” Which is good practice, of course, to acknowledge the limitations of our perspective and the biases in our take on the world. But I notice we tend to be less inclined to do the same with quantitative survey data. We run the risk of treating an engagement survey completed by only half the team at 5pm on a Friday, while they’re rushing and vaguely resentful of yet another survey, as if it’s more truthful than a set of thoughtful observations and reflections from someone who has worked in the team for years. It’s not, it’s just different, and ultimately it’s just more countable.*
Terry Pratchett put it best: “The best research you can do is talk to people.” Talking to people, and listening to them, with openness and curiosity, gives you things numbers can’t. You might discover that no one really speaks up because they know only the boss’s opinion counts. This means it’s not necessary that anything ‘bad’ will happen as a result of speaking up, but that it’s futile, so why bother. That’s a very useful insight into how the team works. Perhaps someone else tells you they’ve always taken responsibility for speaking up even when it’s hard or unpopular, and when you press a little, it turns out what they actually mean is that they don’t think they’d be fired anyway. This tells us that for this person, the stakes aren’t as high as they might be for others, which is quite different to either the whole team feeling safe or that person being particularly ‘psychologically brave’. None of this gets captured in survey scores, but it’s invaluable for planning next steps for a team.
The point, I think, is that we’re often scared of qualitative measurement because, unlike surveys, which have become accepted as standard practice across organisations, we don’t always know how to do qualitative measurement of psychological safety – and whether it “counts”.
What qualitative measurement looks like in practice
There are rigorous ways we can go about formal qualitative measurement of psychological safety if that’s what we want to do. However, I think it’s also important to acknowledge that we’re already doing it. In every conversation we have with someone about how they’re getting on with colleagues, in every team meeting we’re part of and every interaction we observe, we’re gathering information, rich qualitative data, about the team. If our focus is psychological safety, we can find out a lot this way. All of this helps build our picture of what the organisation is like. It’s not a complete picture, but neither is a survey score.
“Trying to measure the traditional way with metrics is a suboptimal (at best) approach for complex properties or phenomena like safety or culture. A metric will provide you limited information and more often than not mislead you. A better way of ‘knowing’ is to aim for rich descriptions.” – Carsten Busch
While many organisations speak the language of ‘numbers’, and while an alarming statistic might open the door, research suggests it’s qualitative data – people’s voices and stories – that actually shifts attitudes. Numbers justify decisions; stories make us believe in them. In my experience, even with the most quant-driven folks, what really brings a data story to life is actual quotes – using what people have actually said to build a picture of what’s behind the numbers.
If we do want to more formally gather some qualitative data about psychological safety, what could we try? One of the major constraints in qualitative measurement or research is that it takes time – and to put it bluntly, if we get loads of qualitative data, loads of words, it’s going to take a lot of time to analyse well. But there are manageable ways we can integrate more qualitative exploration into our measurements. A few things come to mind, from the ‘easiest’ to the more involved:
If we’re going to do a psychological safety survey anyway, we can at least add an open text comment field – pop a box in to ask people why, or give them an opportunity to share some qualitative nuance behind their answers.
Observe what’s already happening. When we have the opportunity to join team meetings or join sessions, what do we notice? Who’s speaking? Who isn’t? What sorts of questions, ideas and concerns get raised? What happens to them? What kinds of opportunities and channels do people have to speak up? Do they have “think time”? Can they share in writing instead of verbally, or after the meeting as well?
Set up short interviews, being mindful that this will likely be less effective if we’re interviewing people we directly work with or whose career trajectories we have power over. We also need to think carefully from an ethical standpoint about what we will do with the data and make sure we’re clear with our interviewee – can we guarantee anonymity? But if we can navigate all that, we will find a lot out, through asking people questions like:
How safe do you feel speaking up in this team?
What might stop you from sharing your ideas?
Can you tell me about a time that you held something back? etc.
– As an aside on this one – rather than having to start from scratch with interviews, there are also often wonderful rich sources of interview data held by organisations in exit interviews – what have people shared in those that might relate to psychological safety? And if your org doesn’t do them, why not? They’re an ideal opportunity to gain really candid feedback about what it’s like to work for the organisation – from people who are leaving, so don’t have so much to lose.
Focus groups. Again, these take some setting up, and a neutral and skilled facilitator, but on a group phenomenon like psychological safety, bringing together small groups to talk about their experiences of psychological safety in the organisation could generate some really rich data.
The most useful thing you can do before trying to change the conditions in a team is understand what they actually are. A chart or heat map won’t give you that. Talking to people, and listening to what they say, just might. When we listen to understand, rather than to try to close a gap on a dashboard, it’s an act of curiosity and an act of care. And if we don’t care, we shouldn’t ask.
*As a slightly tongue in cheek aside, I’d like to point out the fist of five – where people hold up fingers to indicate how psychologically safe they feel – is considered a valid measure in some quarters. If that clears the bar, I’ll take our qualitative observations with some confidence.
For a long time now, we’ve been thinking about a way to make our nearly 300 psychsafety.com articles more easily accessible and searchable in an organic way. Rather than go to the typical search bar and type in a keyword, we wanted a way that folks could actually explore concepts such as power, voice, and organisational ecological thinking without knowing exactly what they wanted to find before they did so.
This is that tool. Head to explore.psychsafety.com and you’ll find a visual map of all our articles, plus a few from our partner websites. You can filter by theme and author, search by keyword, find connecting articles and topics, and head over to read the article from the map itself.
The line thickness indicates strength of connection, whilst the size of a node indicates how connected it is. And the colours represent the themes involved in that piece. Dotted lines around a node indicate that it’s a “hub” article.
If you’re feeling lucky, just click “Surprise me” on the top bar, and you’ll explore a random node and its connections.
Search by keyword, author, theme or idea.
At the bottom of all our articles, you’ll also find a panel that will take you to the knowledge map and explore its relationships with other nodes in the network.
Then click on a paper, and you’ll not only see its connections to other papers, but the panel will give you a summary of the piece, a Harvard-style citation, related psychsafety.com articles, and more. Plus you’ll find a link to the actual paper, together with an indication of whether the paper is accessible or behind a journal paywall.
Explore.psychsafety.com is still in beta, and we’re always ironing out bugs, adding new features, and building up the papers database, so if you find a bug, would like to suggest a feature or another paper to be added, let us know! We also have an open source version available on GitHub (using classic films as data) so that you can create your own Knowledge Maps.
We’ve previously covered the “calculus of voice”: the mechanism by which we decide whether to speak up, stay silent, soften what we were going to say, or say something else entirely. This mental risk calculus is an ongoing, dynamic process, not a one-time assessment of conditions – we’re continuously updating our predictions within interactions, not just between them. It’s usually fast, unconscious, and affective rather than deliberate.
Some models of this cost/benefit calculation – what Detert and Burris call the “affect-laden expectancy-like calculus” – typically assume an even weighting of the costs of speaking up versus the benefits. However, Prospect Theory, as we’ve discussed previously, shows us that the calculus will tend to lean towards the costs due to our inherent loss aversion. This is not a quirk of particularly anxious individuals – it applies to almost all of us as human beings, to varying degrees.
But there’s another crucial layer missing to this voice calculus concept.
Positive interpersonal predictability
One useful way to think of psychological safety can be as “positive interpersonal predictability”. We take interpersonal risks not because we feel vaguely comfortable, but because we can run a rapid mental risk calculus, which requires being able to predict how our vulnerability will land. Positive interpersonal predictability names the actual mechanism by which we may have high or low confidence that the response will be good enough to justify the risk.
This helps to explain why people don’t speak up even in ostensibly “friendly” environments, because if they can’t predict the response with enough confidence, the safest, rational option is to stay quiet. This is why “psychological safety” as a label sometimes misleads, because the word “safety” imports connotations of threat and protection that aren’t quite right. In practice, we’re often not primarily managing danger: we’re navigating uncertainty. And that helps us realise that ambiguity is one of the biggest destroyers of psychological safety.
Before we speak up with an idea, a question, challenging something, or admitting an error, we attempt to estimate the risk of a negative outcome. And to make this calculation, we need data. That data might take the form of past experiences with the people in the group, for example what we’ve seen happened to others when they took interpersonal risks, or even what we’ve heard happened to them. The data might also be our own past experiences in previous workplaces or our childhoods, and many other things too. In the absence of direct contextual information such as how this person has responded to me in the past, I’ll fall back on more distant, less directly connected information, such as how my last boss behaved, instead. And whilst I may use that less relevant data for my risk calculus, as a result I have much less confidence in the prediction I’ve made.
Ambiguity is one of the biggest killers of psychological safety.
We may, of course, be highly confident of a negative outcome. For example, I remember an old boss I had who would always lay into people for making mistakes or not following procedure. The benefit for me (if you could call it that) in this case was that I had a high degree of confidence in how he’d react to me raising a concern – which meant that I was able to plan and prepare for that response. In this situation, at least I had some agency.
A potentially worse situation to be in is where we have a boss who may either react very well or very badly, but we can’t predict which – because in this case, we can’t even attempt the risk calculus in the first place. The hot-and-cold manager isn’t just unpleasant, they are actively creating the worst possible conditions for interpersonal risk-taking, because their behaviour resists modelling. In these conditions, we can’t build a prediction and our tacit calculus can’t be run. Our default behaviour under unresolvable ambiguity is very likely, don’t take the risk. Ambiguity removes agency because it removes the inputs to decision-making.
Of course, there is a distinction between our confidence in prediction and how accurate that predictionis – someone can be confident about how they’ll be received, and still be wrong. But the point stands – for our prediction to be accurate, we need good data to work from.
This is why telling people “This is a safe space” rarely works. The declaration of a “safe space” doesn’t actually generate meaningful data, and data is what our mental calculus runs on. If anything, the need to announce it might itself imply that if you have to say it’s safe, it probably isn’t. What we need isn’t an assertion of safety; it’s enough consistent observational data to form our own accurate prediction. And frustratingly, ambiguity can persist or even increase in psychologically unsafe teams because people are reticent to question things or ask for greater clarity.
Ambiguity aversion
In 1961, economist Daniel Ellsberg (later famous for the Pentagon Papers) described a thought experiment that illuminates exactly this. Imagine an urn containing 90 balls: 30 red, and 60 that are some unknown mix of black and yellow. We have to bet on which colour we’ll pull out. Most people choose red even though red is outnumbered two to one. Why? Because with red, we know the odds. With black or yellow, we don’t. Given a choice, we’d typically rather know the odds, even if they aren’t that great, than face uncertainty about what they are.
This is ambiguity aversion, which is distinct from, and additional to, loss aversion. It isn’t just that people dislike bad outcomes. We dislike situations where we can’t even form a reasonable estimate of what the outcome might be. The hot-and-cold manager is an Ellsberg urn. The employee isn’t just facing a potentially bad outcome: they’re facing unpredictable probabilities, which, combined with loss aversion, can trigger a more paralysing form of risk avoidance than even a known-negative environment (the “consistent asshole” manager) would.
Essentially, in some cases, we may actually prefer a predictable, slightly negative outcome, over an uncertain, potentially positive one.
The data available for making predictions, and the cost of getting them wrong, aren’t the same for everyone, so it’s worth considering how this interacts with power.
Someone with institutional standing, tenure, and a personal relationship with leadership likely has a good amount of power in an organisation, and they have also accumulated a rich predictive dataset. Someone newer, more junior, from a different cultural background, or without access to informal networks is operating with far less data – not necessarily because the environment is hostile to them, but because they have fewer reference points to make the calculation. More importantly, for that person who also likely has less overall power in the organisation, the stakes of getting the calculation wrong are often higher. Predictability is unevenly distributed, and that uneven distribution maps closely onto existing power gradients. Accessing predictability, it turns out, is a privilege. And like most privileges, it’s often invisible to those who have it.
This is why making norms explicit and adopting low-context communication, where we don’t assume too implicit shared understanding, aren’t just tools for building team performance, they’re fundamentally redistributive. They help transfer predictive capacity from those who already have it to those who don’t.
The 2-Gate Model of Psychological Safety: Increasing interpersonal predictability
If we think about the elements of our mental process when we make these calculations, we effectively weigh up “how much data do we have?”, and “what does that data tell us?”
This framework thus suggests two distinct targets for intervention, which we’ll call “gates” (we’ll see why in a moment):
Gate 1 is ambiguity reduction – making the implicit, explicit. This is about how much data we have and our confidence in it. It lets me know whether I can form a confident enough (or indeed, any) prediction about the result of this interpersonal risk
Gate 2 is shifting the calculationpositive. This is the actual calculation that I run, which lets me know whether that prediction leans positive enough to clear my bar for loss aversion.
If Gate 1 doesn’t open, we don’t even get to Gate 2. And both gates typically have to open in order for us to take the risk.
For Gate 1, our interventions are all about increasing predictability – giving us more opportunities to gather data, and make implicit norms and expectations more explicit. They include artifacts and practices like co-created social contracts and team charters, well defined group norms, definitions of “done”, PACE, Ladders of Participation, Lean Coffee, the Andon Cord. These things don’t necessarily change whether the environment is kind or punishing (we could feasibly have a social contract that says “be horrible to each other”!), but they do make it legible, and more predictable. That predictability is the first precondition for psychological safety.
Gate 2 is making those predictions as positive as possible, reducing interpersonal costs, and increasing the benefits of speaking up. This is categorically harder. We can design in opportunities for better legibility but we can’t enforce positive responses from everyone to interpersonal risk taking. What we are aiming for is repeated demonstrations over time that it is in fact safe for people to take interpersonal risks. In practice, this means not just having a social contract, but making sure the agreements are as supportive as possible of interpersonal risk taking, and that we act decisively if it is breached. It also includes following through when issues are raised so people see that they matter, being mindful of how we respond, appreciating diverging perspectives and contributions, praising people for admitting mistakes, reframing work as learning, and many more. Every time these occur, they provide a helpful, positive social proof data point that updates the next prediction.
Recent experimental work by Artinger and colleagues (2025) provides empirical support for this in another form, showing that low psychological safety significantly increases what they call ‘defensive decision making’ – where employees choose a personally safer option over the one they believe would be best for the organisation. In the large public administration they studied, they estimated this pattern equated to forgone opportunities worth around 10.8% of annual revenue.
Of course, silence produces less data; less data increases ambiguity; and more ambiguity produces more silence. That’s a reinforcing feedback loop, and like all reinforcing loops it accelerates in whichever direction it’s already moving. High psychological safety environments tend, over time, to become safer, because interactions produce data, data reduces ambiguity, reduced ambiguity encourages more interactions. Low psychological safety environments tend to become less safe for exactly the same reason in reverse. This is why deliberate intervention matters: left alone, the system will amplify whatever condition it’s already in.
It also helps explain why Gate 1 and Gate 2 interventions have different characters. Gate 1 interventions are largely structural: explicit norms, and defined expectations and processes. Structural interventions, once in place, tend to be reasonably durable. Gate 2 interventions are more like changing a flow over time, through consistent demonstrated behaviour, which accumulates slowly and erratically and can be eroded quickly. Both matter; but they operate at different speeds and with different leverage.
Empirical work supports the distinction between ambiguity reduction and valence improvement as separate interventions. Kusuma and Etikariena (2023), in an experimental study with 262 employees, found that role ambiguity and psychological safety each significantly affected employees’ intention to engage in innovative work behaviour.
Thinking in a new way about psychological safety
This matters practically because it suggests two distinct failure modes in low psychological safety environments with different remedies:
The first is failing to open Gate 1. This is Ambiguity Failure: we don’t have enough data to perform the calculation. This failure to have enough (or any) data may occur because we’re in a new team, working under unfamiliar conditions or are in an environment consisting of highly unpredictable characters.
The second is failing to open Gate 2 – Valence Failure: we have enough data, but the result is negative or neutral (valence is a psychological term that describes the attraction or repulsion one feels towards something).
Valence Failure isn’t only about clearly negative predictions. Even a neutral predicted response (where the cost and benefit are equal) tends, under loss aversion, to register on the negative side of the bar, i.e. a neutral result is rarely safe enough. The prediction has to lean significantly and confidently positively to clear the bar (and that bar is set differently for each of us, by history, culture, and neurodiversity as well as the immediate context).
A lot of psychological safety interventions treat Ambiguity Failure as if it were Valence Failure – they assume that the problem is that people expect negative responses, when in reality, maybe they just can’t predict the outcome at all. We should therefore remember to ask: “is this an ambiguity problem or a valence problem?”, because the interventions for the two are very different.
This framework proposes something the psychological safety and voice literatures to date have not explicitly done: connect the calculus by which people decide to speak up with the structural conditions that make speaking up possible, and names predictability as the missing mechanism connecting them.
References:
Artinger, F.M., Marx-Fleck, S., Junker, N.M., Gigerenzer, G., Artinger, S. and van Dick, R., 2025. Coping with uncertainty: The interaction of psychological safety and authentic leadership in their effects on defensive decision making. Journal of Business Research, 190, p.115240.
Berger, J., Cohen, B.P. and Zelditch, M., 1972. Status characteristics and social interaction. American Sociological Review, 37(3), pp.241-255.
Ellsberg, D., 1961. Risk, ambiguity, and the Savage axioms. The quarterly journal of economics, 75(4), pp.643-669.
Detert, J.R. and Burris, E.R., 2007. Leadership behavior and employee voice: is the door really open?. Academy of management journal, 50(4), pp.869-884.
Kusuma, A.H. and Etikariena, A., 2024. Effect of role ambiguity and psychological safety on employees’ innovative work behavior intention. Persona: Jurnal Psikologi Indonesia.
“Between stimulus and response, there is a space. In that space is our power to choose our response. In our response lies our growth and our freedom.” Viktor Frankl, psychiatrist and Holocaust survivor
In the gap between thinking something and saying it, everything happens.
The voice calculus mechanism
Before every act of voice (which may of course not be verbal), a rapid, and usually unconscious, risk assessment takes place in the mind. It’s what Detert and Burris (2007), drawing on earlier work by Ashford et al, called an “affect-laden expectancy-like calculus” – a simple cost / benefit calculation wrapped in a delightfully academic phrasing. It’s essentially “what are the potential costs of me speaking up, and what are the potential benefits?” If the costs outweigh the benefits, then we’ll lean towards silence; if the benefits outweigh the costs, we’ll lean towards speaking up.
The costs and risks of speaking up are many: from a fear of losing our status in (or being excluded from) the group, being passed over for a future project, being embarrassed for how it comes out, facing punishment or “retraining” for a mistake, simply creating more work for ourselves or others, and many more.
What the calculus runs on
This is one way to think about psychological safety, not the only way. It is also somewhat reductive, but it is certainly useful. The inputs to the calculus vary enormously between people and contexts, and this variation matters enormously for how we think about the work of psychological safety.
This is especially true when we think about the calculation involved when we need to speak up to challenge something – maybe to raise a concern, or express a different perspective. Our cultural background, including national and religious dimensions, influence what we consider to be “acceptable” challenge, as well as who it is acceptable to challenge. That background will likely shape what we feel is an appropriate degree of deference and what constitutes disrespectful directness. Our neurodiversity may affect how we process social signals, read ambiguity, and assess interpersonal risk; for some this means our calculations come with considerably more intensity, or more difficulty, than neurotypical colleagues might assume. Socioeconomic background also leaves a significant imprint. Even if we’re financially secure now, the lessons we absorbed early in life about who gets to speak, who gets listened to, and what could happen if it goes wrong don’t dissolve when we enter a workplace.
Our past experiences accumulate and form a large dataset for this mental calculus: someone who was punished for speaking up in a previous role, or grew up in a household where dissent was unsafe, might be running the same calculation as someone who has rarely faced humiliation or punishment for the interpersonal risks they took. But these two people are running the calculation with very different data, and with different predictions about the potential costs should it go wrong.
The Psychological Safety Observer Effect
And then there are the signals from the immediate context: what we’ve observed happen to others in this group when they took interpersonal risks; the stories that circulate about what this organisation does with bad news; the unwritten norms about what is sayable here and what isn’t; even the physical arrangement of the room, which can encode hierarchy in ways we absorb without consciously noticing. All this, and more, flows into the calculation.
The calculus is always running
Importantly, a cost-benefit calculation about whether or not to speak up isn’t a one-time assessment we make at the start of an interaction and then stick to. It’s a continuous process of recalibration, updated in real time and shaped by many factors. For example, as we begin to speak, we may observe that our manager has begun to glare at us – it’s quite likely we recalculate and might quickly adjust what we say next or whether we continue at all. We are always looking for more data to help us make, and re-make, this calculation. And we are always, likewise, providing data for others to make their own calculations. How we respond matters.
The calculus, in other words, isn’t only running on what is happening right now. It is running on an accumulated history of signals, some of them decades old, some based on deeper cultural norms and rules, some absorbed from watching what happened to someone else. This is the substrate of psychological safety, and why “just create a psychologically safe environment” is considerably harder than it sounds.
This voice calculus applies to all of us. It’s a fundamental feature of being human in social settings, not a pathology of “overthinkers” or the especially anxious. And as we explore in this article on Prospect Theory, our human calculation is always skewed somewhat to the negative, even if the “objective” calculation would be more neutral.
A note on “voice” calculus
Voice, in this context, is any act of communication intended to share something that matters to the “sender”. An email, a text message, a post-it note, hand signals, a drawing, or even just a look. It includes people who stutter or have speech differences; people expressing themselves in a second or third (or ninth!) language; people using sign language or assistive technology; people (including children) whose most natural mode of expression is non-verbal. The calculus applies to all of it.
Narrowing “voice” to verbally spoken words often excludes exactly the people who most need to be heard, and encodes, maybe insidiously, an assumption about what legitimate communication looks like.
Four outputs, not two
The calculus doesn’t produce a binary output of “speak” or “don’t speak”. There are four possible outputs.
The first is to speak up and to say the thing, as fully and clearly as intended.
The second, opposite, outcome is silence: to say nothing, or to deflect.
The third is softer, far more common, and considerably less discussed in the literature: to soften and dilute what we were going to say. This might be the natural outcome of the calculus, or it might be because the option of silence isn’t possible: such as when a leader asks us directly for a response, when the meeting requires an answer, and especially when saying nothing would itself be highly conspicuous. We might then phrase something as a question instead of a statement, make a much more mild or tentative observation, or wrap something up in qualifiers and apologies. In those moments, instead of silence, the calculus produces softening as interpersonal risk management, and we share a mitigated, more ambiguous, more palatable and less interpersonally risky version instead.
And of course, we might take this third, softening approach intentionally (as in the graded assertiveness practice of PACE) as a helpful way to walk ourselves up to more interpersonally risky communication if the situation warrants it. Or we might do it in a more avoidant way because we don’t feel safe to be more direct.
There is a fourth possible outcome, where we completely change from what we were going to say in order to please the recipient or avoid their wrath; this may even involve telling a lie. It is, fundamentally, simply another form of interpersonal self-protection.
Most of us will recognise the experience of walking into a meeting with the intent of saying one thing and coming out having said a diluted, or even completely different, thing. The gap between those two things – between what we meant to say and what we actually said – is the result of our tacit calculus of voice.
Local Rationality
Local rationality is a useful frame for the calculus of voice. When speaking up, especially against a power gradient, silence or softening is often the locally rational choice: it makes sense, given what we believe we’re capable of, what we know about our position, the risks we face, our focus of attention and our reading of the signals and situation around us.
The first officer who sees the captain making a fatal error and says “Which runway are we aiming for?” rather than “Stop! We need to change course now.” has made a locally rational calculation, in real time, under pressure. If we deem that lack of directness to be a failure, the failure is not theirs. The failure is in the conditions that made softening the outcome of the calculation. Blaming individuals for behaving exactly as the system encouraged them to behave is the wrong diagnosis, and worse, it near-guarantees the same outcome next time.
Reduce the costs and increase the benefits
The voice calculus is psychological. It happens inside us, shaped by our personal history, our reading of the current situation, and our assessment of the likely response. Locating the mechanism there is not the same as locating the responsibility there – we’re not saying this is solely down to individuals to be “psychologically braver”. The calculus is internal; the conditions that shape its output are largely not.
The inputs to the calculus are social and structural as well as individual: what data does a person have access to? What has happened when others spoke up? What norms govern this group? What is the power differential between this person and the target of their voice? Those are questions of the psychological safety substrate: the accumulated conditions that precede any particular interaction and shape what the calculus produces. Psychological safety calculations reside in our minds but it’s the substrate we’re in that can make it more or less likely that people will feel safe.
Interventions that address only the structural conditions, without understanding the mechanism, may miss why people still stay silent in ostensibly “safe” environments. Interventions that address only the individual (coaching people to be “courageous” or training them in speaking up skills) without changing the more systemic dynamics simply ask people to absorb the costs that haven’t actually been reduced.
The work is to change the conditions that make speaking up costly while supporting the people who have to navigate those conditions in the meantime. That means taking seriously both elements of the calculus: the internal mechanism, and the environment that loads it. Understanding this calculus gives us a seemingly simple (but rarely easy!) task in order to foster greater psychological safety: we need to reduce the costs of speaking up for those around us, and increase the benefits.
Ashford, S. J., Rothbard, N. P., Piderit, S. K., & Dutton, J. E. (1998). Out on a limb: The role of context and impression management in selling gender-equity issues. Administrative Science Quarterly, 43: 23–57.
Detert, J.R. and Burris, E.R., 2007. Leadership behavior and employee voice: is the door really open? Academy of Management Journal, 50(4), pp.869–884.
A lot of relationship friction comes from faulty assumptions: two (or more) people each operating on their own mental model of what was agreed, who’s responsible for what, and what counts as acceptable, without ever actually comparing notes. Contracting is the practice of surfacing those assumptions before they become problems.
We contract with people all the time, and not always with a handshake or a signed document. More often we contract through a casual chat over coffee, a WhatsApp message or a quick check in on Slack or Teams. We contract with our window cleaner about when they’ll come and how they want paying, with a colleague about who owns which part of a shared project and with a friend about what time to arrive for a group trip.
The word “contract” has been around for centuries. English contract law traces its roots to ancient Greek and Roman legal traditions, and the concept of binding agreements between parties is as old as organised society. But the idea of contracting as a tool for everyday human interaction – something we might use not just in legal or business agreements, but in conversations, relationships and all kinds of workplaces – owes a great deal to Eric Berne.
Berne, the psychiatrist who developed Transactional Analysis (TA) in the 1950s and 60s, gave us a definition that captures a lot. Berne saw a contract as:
“an explicit bilateral commitment to a well-defined course of action” (Berne, 1966).
Each word in this does important work:
Explicit: This doesn’t necessarily mean contracts have to be written, but what is agreed does need to be spoken or articulated in some clear way. Not assumed and not implied.
Bilateral: As in – both people need to agree on this (we’ll come to multilateral contracts later). And we need to commit to them, which requires opting into an agreement.
Well-defined: The course of action that we agree to should be well-defined; we should strive to be as clear as possible about what we’re going to do.
Crucially, a contract is future focused – it’s about what we will do together, to move forwards.
Berne also identified three levels at which an effective contract operates:
Administrative – the practical logistics of how we work together
Professional – our roles, responsibilities and boundaries
Psychological – the expectations, needs and assumptions each party brings
When I trained as a coach, these three layers became the architecture of every new working relationship. At the start of a coaching engagement, you’d sit down with the other person and work through them together. How do we contact each other? What’s in scope and what isn’t? What does each of us need for this to work?
It sounds quite formal when it’s written out like that. But in practice, it’s a really good, honest conversation, where the administrative, professional and psychological layers naturally overlap and support one another. After a good contracting conversation, both sides leave feeling clearer and happier about how we’re going to work together.
Recontracting
During my training and subsequent coaching, I noticed again and again that most coaching relationships difficulties weren’t really about the work. They were about something that either hadn’t been contracted properly at the start, or a contracting agreement that somewhere down the line had fallen through. It might be a misaligned expectation, an assumption that turned out to be wrong or a boundary that was unhelpfully fuzzy.
The fix, almost always, was to go back to the contract. Not to start over, but to revisit and to either make explicit what had been left implicit, or to revisit the original agreement and see whether it was still working for us both. We called it recontracting.
This might look administrative on the surface: “You’ve cancelled the last two sessions without much notice. Could we agree a cancellation window – two days, say?” Or it might look more psychological: “I notice you seem a bit distracted. Is there anything that would help this time work better for you?”
Either way, the move is the same. We name the thing, agree something new and move forward with greater clarity.
People Problems
In coaching we used to say that most problems could be solved by recontracting, and this is likely true of many, if not most, “people problems” too.
It might sound like a sweeping statement, but think about the difficulties that tend to simmer in teams and relationships: the colleague who’s making unreasonable demands; the friend whose version of “being on time” bears no resemblance to yours; the family childcare arrangement that worked brilliantly until it didn’t, and nobody has quite been able to say so. In most of these cases, something was never properly agreed in the first place, or was agreed and then outgrown. Very often, the gap between what each person expected and what actually happened is the source of the friction, not some deep incompatibility between the individuals involved.
Contracting resolves conflict by surfacing it early enough to actually work with; it’s the opposite of an “avoid it and hope it goes away” approach. The initial discomfort of an honest, respectful conversation (what TA would describe as adult-to-adult) is almost always preferable to the ongoing frustration of unmet expectations.
Social contracts
But what about when we need to agree on a way of working with more than one person? In that case, it’s not just a bilateral agreement that’s needed but a multilateral one. In the workshops and programmes we run, we always use a group social contract. We’d recommend the same for any meeting or group that spends time together. It’s a brilliant way of clarifying, from the start, what we can expect of each other.
Our typical social contract includes administrative elements (“we’ll finish on time”), professional elements (the Chatham House Rule, which sets the terms for what can and can’t be shared beyond the room), and psychological elements (“all questions are welcome”). Some elements sit across multiple layers. “We have a car park” for example is partly administrative, as it’s about how we organise the session, but it’s also psychological, as it signals that even if we can’t follow the thread of a question in the moment, we still value it being asked.
For a longer-term contract we might co-create it with the team, but even in a shorter session, we can suggest one and invite improvements. Once we have an agreed version, we explicitly check that people are happy to commit to it, maybe by asking for a thumbs up. People know what to expect of the space, and what’s expected of them, and that helps us all feel safer.
I saw a great example of this in a workshop recently – the facilitator shared a social contract with the group, and asked whether anyone in the group wanted anything added. Someone raised their hand: “Should we have something about phones?” The facilitator replied, “What would you like?” to which the participant replied “We don’t use them?” The facilitator checked in with the group – someone else suggested adding the acknowledgement that taking emergency calls was fine – and the group agreed and carried on.
People are far more likely to speak up, take interpersonal risks and ask questions in environments where they feel it is genuinely safe to do so. Clear contracting is one of the most practical mechanisms for building that kind of environment. When people understand how a space works, what kind of participation is expected, what happens we disagree and where the boundaries are, the uncertainty and ambiguity that so often produces silence starts to lift.
Contracts as living things
Berne’s three-layer model has been developed extensively by later practitioners, helping remind us that contracting is an ongoing and dynamic process rather than a one-time event. That’s the part that tends to get missed. We too often treat contracts as something you establish at the start and then put away, but relationships change and contexts shift. What worked well six months ago may not serve either party now.
Recontracting then is less a corrective measure than a regular practice. Good working relationships are living things, and the agreements that support them need to be alive too.
Ask for what you need
There’s something that can feel almost radical about taking contracting seriously in our day-to-day relationships. Muriel James and Dorothy Jongeward took the definition a little further and put it as:
“A contract is an Adult commitment to one’s self and/ or someone else to make a change.”
The point about commitment to yourself is important here. The practice of contracting asks us to state clearly what we need, and then to check whether we’re actually aligned with the person opposite us, rather than assuming we are. Crucially, it requires us to be honest about our needs, and not hide behind a covert agenda.
When we do that well, contracting creates the conditions for both of us to do our best work, without the accumulated resentment of a hundred unspoken, then broken, expectations. We tend to reserve that level of explicitness for high-stakes or formal contexts, but it belongs in the everyday too.
A few years ago, I was working for a client in Stockholm and in some free time, I visited the wreck of the Vasa, the world’s best-preserved 17th-century ship. She’s housed in a museum built specifically around her – enormous and (mostly) intact, after spending nearly 350 years under the sea. I’d known the Vasa’s story for a while and had even used it in workshops, but standing next to the thing itself is different. I found myself looking up at this huge beast of a construction, wondering how many people who were involved in building it knew that it would sink.
Understanding why the Vasa sank also gets to something important about how organisations fail.
The Vasa and the King
On 10 August 1628, the Vasa set sail from Stockholm harbour on her maiden voyage as the newest and most expensive ship in the Royal Swedish Navy. Commissioned by King Gustavus Adolphus, she was laden with ornate carvings, towering masts and 64 bronze cannons; one of the most formidable displays of naval power in Europe, and a deliberate statement of Swedish ambition. The crowds who had gathered to watch her depart included plenty of foreign ambassadors; in effect, spies of Sweden’s allies and enemies, there to report back on the launch.
What they witnessed was not what anyone had planned. After sailing roughly 1,300 metres, and still within sight of the king’s palace, a gust of wind caught her sails and she heeled sharply to port. Water rushed in through the many open gunports. Within minutes, the Vasa had sunk to the bottom of the harbour with between 30 and 50 lives lost out of a crew of around 150. The wind that sank her was later estimated at about 8 knots, which is not by any means a strong wind. Subsequent calculations suggest she might even have gone over in a breeze of just 4 knots.
The captain, who survived, was thrown into jail the following morning and a formal inquiry was convened.
Scope Creep and Production Pressure
The Vasa’s instability wasn’t a single mistake. It was the accumulated result of years of changing requirements, “scope creep“, poor coordination, production pressure, and crucially, an organisation in which bad news could not travel upwards.
The ship had started life as something quite different. In January 1625, the King had contracted for four ships: two smaller vessels with keels of around 108 feet, and two larger ones at 135 feet (the keel is the structural backbone running along the bottom of a ship’s hull, from bow to stern, like a spine). Then, in the autumn of that year, the Swedish navy lost ten ships in a storm, and the king ordered the smaller ships to be built first, on an accelerated schedule. The Vasa’s keel was laid in early 1626 as a relatively modest vessel.
What followed was scope creep in its extreme. The king learned that Denmark was building a large warship with two enclosed gun decks (something no Swedish shipbuilder had so far attempted) and ordered the Vasa enlarged to match. The 111-foot keel already in the ground was physically extended: a fourth scarf joint was added, where a traditional ship would have three. The hull was widened, but only in the upper sections, because the keel was already fixed. No formal specifications were drawn up for any of these changes. The lead shipwright, Henrik Hybertsson, who had never built a two-gun-deck ship, appears to have scaled up his original plans by proportion and instinct rather than calculation, partly because, in 1628, there were no known methods for calculating a ship’s centre of gravity or stability characteristics. You found out how a ship sailed by sailing it.
Hybertsson fell seriously ill in 1626 and died in 1627, leaving his assistant Hein Jacobsson to complete the project. At the time of Hybertsson’s death, around 400 people across five different teams were working on the ship with apparently little coordination between them. The armament specification changed repeatedly: the final configuration crammed 24-pound guns onto an upper deck that had been built for lighter 12-pound guns, pushing the centre of gravity higher still. The ornate oak carvings the king had ordered, hundreds of them, added yet more weight above the waterline. You can still see these carvings all over the ship today.
There was also a detail that came to light only after the ship was raised in 1961: the construction teams had used four different rulers, two calibrated in Swedish feet (twelve inches) and two in Amsterdam feet (eleven inches). The resulting asymmetry made the ship heavier to port than to starboard. It is, in a way, a perfect metaphor for the whole project: four groups of people, working on the same ship, without any shared and agreed standard.
The Stability Test
What makes the Vasa story extra impactful is that the failure wasn’t hidden. Shortly before the maiden voyage, a stability test was conducted in front of Admiral Fleming and the ship’s captain. Thirty men ran back and forth across the upper deck. After just three traversals, the test was stopped because the ship was rocking so violently that those present feared she would capsize on the spot.
However, there was no good solution available. The hold had no room for additional ballast; the shallow keel, a consequence of extending the original 111-foot design, had required extra bracing timbers that filled the space. In any case, had more ballast somehow been added, it would have pushed the lower gunports below the waterline. The ship was, structurally, beyond rescue at the point of launch. When we think of sunk cost or plan continuation bias, this example should loom large.
And yet, the launch went ahead. The shipwright and the shipbuilder, it later emerged, had not been present at the stability test and were never told about the results. The boatswain, Matsson, who had raised concerns about the ship’s ballast, was told by Admiral Fleming: “the shipbuilder has built ships before and you should not be worried.” Matsson’s response, recorded in the subsequent inquiry, was: “God grant that the ship will stand upright on her keel.”
Fleming, for his part, had reportedly lamented after the failed test: “If only the King were here.” It’s a small window into the bind these men were in. The king had ordered the Vasa ready by 25 July; the maiden voyage on 10 August was already more than two weeks late. Failure to launch meant facing the king’s displeasure. Was it better to face the certainty of the King’s wrath, or the potential risk of the ship capsizing?
The King’s Power Gradient
The Vasa Museum’s own account of the inquest notes that “it was rarely a good idea to disappoint the king.” That’s an understatement, but it points at something real. When we talk about authority gradients in organisations, we’re often talking about the positional power distance between a frontline worker and a senior manager, for example. The gradient between the people building the Vasa and King Gustav was of a categorically different order. Speaking up against a steep power gradient in a modern workplace is uncomfortable and often carries real risk. Kings were appointed by God. Speaking up against a king, in 1628, was unthinkable.
And yet the dynamic is recognisable. To quote Lars Axelsson, “a problem that stays with whoever discovers it is a problem that remains unknown.” The Vasa’s problems were known – to the boatswain, to the men who ran the stability test, almost certainly to Jacobsson, who had always suspected the Vasa was too narrow. What those problems lacked was a path upward. Every layer of the hierarchy had more reasons to ignore or dismiss the concern rather than pass it on, so the information that could have prevented the disaster never reached anyone with the power and the will to act on it.
The formal inquiry in September 1628 found no one to blame, partly because blame was genuinely difficult to assign: the king had approved all plans and armaments, the shipwright was dead, the admiral had conducted a stability test and still let the ship sail; and partly because the structure of the organisation had distributed responsibility so thoroughly that no single decision looked, in isolation, like the fatal one. That, too, is a pattern worth recognising.
VASA SYNDROME: The organisational pattern in which power gradients, production pressure, and the absence of psychological safety prevent the people doing the work from challenging unrealistic demands, so latent flaws go uncorrected until disaster occurs.
Kessler et al, 2004
Vasa Syndrome:
1: Lack of external learning capability 2: Goal confusion 3: Obsession with speed 4: Feedback system failure 5: Communication barriers 6: Poor organisational memory 7: Top-management meddling
After The Disaster
There is a coda to the story that I find almost as interesting as the disaster itself. Hein Jacobsson – the man who completed the Vasa after Hybertsson’s death – had suspected, even before she was launched, that her proportions were wrong. When he was commissioned to build her sister ship, Äpplet, in 1629, he built her a metre wider. Äpplet went on to serve in the Swedish navy for around 30 years.
The lesson there is not that the builders couldn’t learn. They clearly could, and did. The lesson is what it took to make that learning happen: a catastrophic, public, entirely avoidable disaster, witnessed by thousands. Jacobsson already knew what needed to change. He simply hadn’t felt able to say so with sufficient authority until the failure was so large and so visible that a change of approach was essential. That’s a high price for an organisation to pay for a lesson its own people already knew.
The Vasa Organisational Pattern
When I was in the museum in Stockholm, what struck me most wasn’t the scale of the ship, though she is vast (somewhat bigger than the narrowboats I used to live on!). It was how incredibly grand and unseaworthy she looked, with rows and rows of gundecks, crammed together, apparently very close to the waterline. It looks like a demonstration ship – one designed to satisfy an ego rather than to perform. People had to build it and sign off on it. And the people closest to the work, the ones who could see the issues most clearly, were the ones least able to say anything about them.
That is the pattern we see repeated all too often in organisations, in workplaces where the distance between the people at the sharp end and the blunt end is large, and where the cost of carrying bad news upward feels higher than the cost of absorbing it quietly. Projects and programmes become “greenwashed”, so everything looks on track for the leadership who are monitoring, whilst those at the sharp end are crucially aware of multiple real and potential failures. Psychological safety is the structural condition that allows that knowledge to move vertically, and problems to be surfaced before they become disasters.
There were many component causes of the Vasa’s demise: changing designs, the mixed measurement systems, the too-heavy guns, the impossible timeline, but ultimately the Vasa sank because the people who knew it would sink didn’t feel able to say so to the people who could have done something about it.
(Written in collaboration by Tom Geraghty, Bea Poyton and Jade Garratt)
References:
Axelsson, L. (2006) ‘Structure for management of weak and diffuse signals’, in Hollnagel, E., Woods, D.D. and Leveson, N. (eds.) Resilience engineering: concepts and precepts. Aldershot: Ashgate, ch. 10.
Fairley, R.E. (n.d.) Why the Vasa sank: 10 lessons learned. Oregon Graduate Institute. [Unpublished manuscript.]
Hollnagel, E., Woods, D.D. and Leveson, N. (eds.) (2006) Resilience engineering: concepts and precepts. Aldershot: Ashgate.
Kessler, E.H., Bierly, P.E. and Gopalakrishnan, S. (2004) ‘Vasa syndrome: insights from a 17th-century new-product disaster’, IEEE Engineering Management Review, 32(1), pp. 38–48. Available at: https://doi.org/10.1109/EMR.2004.25008
The cases of Dr Hadiza Bawa-Garba and RaDonda Vaught
Content note: this article discusses the deaths of patients, including a child. Co-created and edited by Jade Garratt and Bea Poyton.
Dr Hadiza Bawa-Garba is a British paediatrician. RaDonda Vaught is an American nurse. Both made errors in complex, high-pressure systems and both were subsequently prosecuted for it.
Their cases are very different, but they do share a consequence: the implicit message sent to healthcare workers was that honesty about error carries criminal risk. What happened to patients in both cases mattered a great deal – the harm was only too real, and so was the grief of the families and the staff involved, the sometimes-called “second” and “third” victims. But what happened to Bawa-Garba and Vaught afterwards may have put far more patients at risk, by eroding something that safe healthcare depends on – the safety and freedom to speak up and share learning from mistakes.
“Underneath every simple, obvious story about ‘human error,’ there is a deeper, more complex story about the organization.” Sidney Dekker
The Case of Dr Hadiza Bawa-Garba
On 18 February 2011, six-year-old Jack Adcock was admitted to Leicester Royal Infirmary. He had Down’s Syndrome and a pre-existing heart condition, and had come in with vomiting, diarrhoea and breathing difficulties. He died that evening.
Dr Hadiza Bawa-Garba, a paediatric registrar (ST6 – which means she was in year 6 of her post-grad training), led Jack’s care. In the subsequent investigation and legal proceedings, a number of errors in the clinical management of Jack’s case were identified. A chest X-ray indicating infection sat unreviewed for over two hours – Dr Bawa-Garba was not notified it was available and antibiotics were started late as a result. Blood test results were held up for over five hours due to IT failures. Jack’s regular heart medication was discontinued by Dr Bawa-Garba due to his deteriorating condition, but this wasn’t documented, so when his mother visited that evening she administered his usual dose (which was, and still is, standard practice for family members to administer medication to children), unaware it had been stopped. This led to the circulatory shock and subsequent cardiac arrest. During his cardiac arrest, Dr Bawa-Garba briefly confused him with another patient and paused resuscitation before restarting.
These were real errors and the consequences are devastatingly real too. But to understand them only as individual failures would be a misread of the situation.
The ward was short-staffed, and the on-call consultant was off-site. He hadn’t realised he was on call that day, a rota failure that left Dr Bawa-Garba with sole responsibility for the entire unit. The IT system was also malfunctioning and nursing communication was fragmented.
Photo by gorden murah surabaya: https://www.pexels.com/photo/a-hospital-ward-7563452/
Dr Bawa-Garba was working in conditions that systems thinkers would recognise immediately, with performance-shaping factors stacked against her at every turn. In such conditions, the errors she made can be seen not as aberrations but as predictable outcomes of the system she was embedded in. This is local rationality – the idea that people’s actions almost always make sense from inside their situation, given what they knew, the tools they had, and the pressures they were under. She was a trainee doctor managing an entire unit alone because of a rota error, with an IT system that was actively delaying critical results, in conditions that would have stretched a far more experienced clinician. Her actions made sense in context, even where they fell short.
In 2015, Dr Bawa-Garba was convicted of gross negligence manslaughter. She was suspended, then struck off, then reinstated following a crowdfunded legal battle and significant professional outcry, and finally cleared to practise without restrictions in 2021. It had been a decade of professional limbo for a registrar who was, by most reasonable analyses, failed by her system before the system failed her patient.
The outrage across the medical profession was substantial, but one element of the case stood out above the others, and it matters enormously for psychological safety.
A few days after the incident, Bawa-Garba was asked to meet Stephen O’Riordan, the duty consultant at the time of the incident. At the meeting she was asked to reflect on the circumstances, setting out what she should have done differently. She was sent home immediately afterwards and told not to come back until she was asked to. These reflective learning notes are the kind of structured, honest self-appraisal that medical training actively encourages. She reflected at length on the points where, as a trainee, she felt she could have managed Jack’s care better – all the things she would do differently given the chance. Reflective notes aren’t just for the author to learn from: their intended purpose is at least in part to facilitate collective learning – so we can learn from each other’s experiences.
But those reflective notes were obtained by the prosecution, legally, because nothing prevented it – and the prosecution QC, Andrew Thomas, pressed Dr Bawa-Garba on the reflections she made after Jack’s death. “List for us, please, all of the mistakes,” he said. Her own reflections became evidence that she knew she had made errors.
There is something epistemologically dishonest in that. Reflective notes are hindsight documents by definition: written after the outcome is known, in conditions nothing like those on the ward that day. After the event, Bawa-Garba had time and space to process what happened, and the knowledge of how things had ended. She also had the support in making those notes of the consultant who should have been on call that day. The Bawa-Garba on the ward, in the moment, had none of that – she was a junior doctor managing multiple sick children in an understaffed, overstretched and underfunctioning unit, without adequate consultant cover. Treating her reflection after that event as evidence of what she understood during it mistakes the nature of the document (and of causality) entirely.
The very act of learning from the event was weaponised against the person who tried to learn from it.
“Blame is emotionally important, not operationally important.” Todd Conklin
The effect on the profession was significant. The fact that no law protected those notes from prosecution use only sharpens the point: the boundary between learning material and legal scrutiny had never actually existed. If honest reflection can be read by prosecutors and treated as admission of culpability, why would anyone write honestly? A survey found that over 80% of junior doctors changed how they write reflections after this case — they rationally self-censored out of fear.
Reflective practice only works when clinicians believe that candour is safe. Once that belief breaks, so does the practice – a completely understandable outcome. If honesty about work as done risks prosecution, silence becomes the safer choice, and a profession that has learned to stay silent is a much less safe one.
The Case of RaDonda Vaught
On 24 December 2017, 75-year-old Charlene Murphey was admitted to Vanderbilt University Medical Center in Nashville with a subdural haematoma. Two days later, she was recovering well and nearly ready to go home, but needed a PET scan before discharge, something she was a little nervous about. To ease her anxiety ahead of the procedure, she was prescribed Versed (midazolam).
Nurse RaDonda Vaught went to retrieve the drug from their notoriously flaky automated dispensing cabinet. When she typed “VE”, no results appeared (because unknown to her, the system required the generic name of midazolam). Vaught activated the cabinet’s override function. This was not unusual – Vaught herself later explained that nurses had been instructed to use the override function when they couldn’t access medications patients needed, and many experts have said that this is a common occurrence at lots of hospitals. Scanning the results, she selected what she thought was Versed, but was actually vecuronium: a neuromuscular blocking agent that causes full paralysis. She administered it to Charlene Murphey, a patient who had been on the cusp of going home. Charlene lost consciousness, suffered a cardiac arrest, and died.
Vaught immediately disclosed her errors. She told the truth.
The Medical Center terminated her employment in January 2018 after an internal investigation, but in fact recorded Murphey’s cause of death as natural causes and did not report the medication error to the authorities. It took a whistleblower and a federal investigation to bring the full picture to light. In 2022 (four years after the event) Vaught was convicted of criminally negligent homicide and abuse of an impaired adult. Her nursing licence was revoked as a result. Although a Tennessee Bureau of Investigation agent testified that the hospital had a “heavy burden of responsibility” for the error, no further action was brought against the institution.
The inversion here is stark: the nurse who admitted her mistake was prosecuted, whilst the institution that actively concealed its more structural and systemic role to avoid scrutiny got away with it. With echoes of Bawa-Garba, Vaught’s transparency became the mechanism of her prosecution.
“There is almost no human action or decision that cannot be made to look flawed and less sensible in the misleading light of hindsight.” Sidney Dekker
As with Bawa-Garba’s incident, context is essential. The dispensing machines used drug naming conventions that made confusion easy. The override function existed precisely because the system regularly failed staff who needed to access medications quickly: indeed, staff had been taught how to use the override for exactly this reason. As far as Vaught could tell, the only way for Charlene to get the medication she needed was to bypass the system safeguards, which were designed to catch exactly this kind of error.
But the dispensing cabinet was only part of it. There was no second-nurse verification required for medications pulled via override. Vecuronium – a paralytic agent with no legitimate emergency use case that would require rapid override access – should never have been available via that route at all. Vanderbilt removed it from the override list after Charlene’s death, a relatively simple safeguard that had always been available, but hadn’t been made. There was also no medicine barcode scanner in the radiology department where the drug was administered, which meant a standard check that would have caught the error at the bedside couldn’t happen.
These are not factors that excuse what happened – Charlene Murphey died, and that matters very much – but they do help explain it, and there is a meaningful difference between explanation and absolution. If we address any one of those systemic factors, the outcome might have been different. If instead we focus only on the individual – blame, shame, retrain or dismiss – none of those conditions change. The next nurse walks into the same ward, faces the same broken system, and the same or very similar thing happens again.
The Observation Effect
Psychological safety is not primarily about our own direct experience. Most nurses will never be prosecuted for a clinical error. But when we observe it happen to someone else – someone who made a mistake, admitted it, tried to learn from it, and was then criminalised for it – it affects how we calculate the risks of speaking up and being open about our decisions and mistakes. We don’t need it to happen to us personally for the lesson to land. Vicarious experience is more than sufficient to alter our behaviour, even if those cases are very rare: we may observe 99 successful speaking up attempts, but if we observe just one where the consequences were severe, that’s the one that looms large.
The Bawa-Garba and Vaught cases sent a powerful signal, at large scale, across entire professions: that honesty about error carries serious risk, and that doing the right thing after a mistake offers no protection. Bawa-Garba’s case also showed that our reflective learning notes can be turned against us.
The result is all too predictable: clinicians become more guarded, errors go unreported, near-misses go undiscussed and reflective practice becomes performative rather than genuine. The very data that organisations need to improve patient safety is systematically suppressed.
The Learn Not Blame campaign, which emerged in the UK partly in response to Bawa-Garba, and the widespread commentary from nurses and nursing bodies following the Vaught conviction, both speak to exactly this: that criminalising error doesn’t make healthcare safer: it produces a culture of silence, which ultimately results in far greater risk to the patients the system is meant to serve.
Learning, or Punishing
RaDonda Vaught went to work on 26 December 2017 to help patients like Charlene Murphey. She went to work to do a good job. So did Hadiza Bawa-Garba on 18th February 2011, the day Jack Adcock was brought into her ward. Neither arrived intending to cause harm, but both were working in systems propped up by workarounds that were well known and under-addressed. When things went wrong, both did what we say we want people to do: they were honest about it.
A reasonable objection at this point: surely individuals must be “accountable”? Surely we can’t simply excuse errors because the system was imperfect?
Of course. And that’s not what anyone serious in this space is arguing. The point is not that we as individuals bear zero responsibility. It is that individual accountability and systemic learning are not the same thing; and treating them as equivalent, by prosecuting individuals as the primary response to systemic failure, produces neither justice nor, more importantly, improvement.
What happened to both Bawa-Garba and Vaught after the incidents speaks volumes about what it can cost a person to tell the truth, and about the gap between the cultures that organisations often say they want and the ones that get built. A Just Culture distinguishes between human error, at-risk behaviour, and recklessness. The vast majority of clinical errors fall into the first category: things that happen to competent, well-intentioned people working in imperfect systems. And somewhere right now, after one of those errors, a nurse is deciding whether or not to file an incident report, and a trainee doctor is wondering whether to write honestly in their reflective log. They are making that calculation in the shadow of, admittedly, rare cases like these, in systems that are intended to enable safe candour, but on occasion, dramatically fail to do so.
Vaught and Bawa-Garba weren’t reckless people. They were people doing their best in imperfect systems, who then did the right thing afterwards. If we want people to speak up, we have to make speaking up actually safe. That means in practice, not just in principle, and it means we need to take care of the next person who tells the truth.
Note: This piece was researched in depth from multiple sources (many included below), but some factual errors may remain. We always welcome corrections and clarifications.
And if we can’t make this case, we often won’t get in the room. Budgets are limited, and if we can’t demonstrate value, programmes don’t get funded – and psychological safety becomes a privilege of only exceptionally well resourced organisations rather than reaching the places that need it most.
I’ve made these arguments myself when pitching, justifying a budget line, or helping a client make the case internally. The evidence is real and it’s worth citing.
So the impulse to frame psychological safety in utilitarian terms: to ask what it’s for, what it produces, what return it delivers, is not cynical. It comes from a legitimate place.
But there’s a problem with it too.
What utilitarian logic actually does
Utilitarianism, in its classical form, aggregates welfare across a population and optimises for the greatest good for the greatest number of people. On the surface, this sounds fair: everyone counts, everyone’s interests are included in the calculation. But the utilitarian framework contains a structural flaw that philosophers have been wrestling with for centuries: the welfare of a minority can always be legitimately overridden if the aggregate calculation favours doing so. This isn’t a misapplication of utilitarianism. It’s how it works.
“A man’s virtue is tested by whether he treats those who are useful to him and those who are not useful to him in the same way.” Simone Weil, The Need for Roots
When we ask whether psychological safety “delivers ROI”, we’re (whether we intend to or not) invoking this logic. We are setting up a calculation in which the value of each person’s ability to speak up is weighed against organisational cost. And once that calculation is running, its conclusions are unfortunately rather predictable.
Who loses?
Helping people speak up is not equally easy across a workforce. For some people – those who are already confident, possess more power, or who resemble the leadership – many of the conditions for speaking up exist more or less by default. Creating psychological safety for these people is low-cost, because the system was largely built around (or by) them.
For others, it’s much harder. Improving psychological safety for someone who is neurodivergent, for a person of colour in a predominantly white organisation, for a woman in a male-dominated team, for LGBTQ+ folks in more conservative cultures, for a manual worker in a professional environment, for people from lower socioeconomic backgrounds, or for anyone with a “marked identity” (something that marks them as different from the contextual norm), this work is often slower, more structural, more disruptive and challenging to existing habits and hierarchies. Essentially, in terms of resources, time and money, it may be genuinely more expensive.
A utilitarian approach can keep finding reasons why this particular work, for these particular people, is not quite “worth it”. Not because anyone is being deliberately cruel, but because the framework itself will keep producing that answer. The least safe people are often the most costly to serve, and in a cost-benefit world, cost wins.
And it’s worth noting that it always seems to be other people who are judged to be receiving “too much” psychological safety – as costing too much. The people (who are almost always people for whom speaking up is already safe) making that argument never seem to volunteer themselves for experiencing less psychological safety.
A different framework: capabilities and rights
The philosopher Amartya Sen argued in Development as Freedom, that human wellbeing cannot be captured by utility measures (such as GDP) alone. What matters is whether people have genuine capabilities (the real and practical freedoms to do and be). Martha Nussbaum extended this to describe central human capabilities that function as threshold conditions that must be met regardless of what an utilitarian argument might warrant. The key point is that these capabilities are not inputs to an equation. They are things that must be in place, for everyone.
I believe that the safety to speak up without fear of punishment or humiliation fits squarely within this framework. It is not (solely) a productivity lever. It is a basic condition of full participation in our own working lives.
“Each person is an end, and none is merely a means to the ends of others.” Martha Nussbaum, Creating Capabilities
This is also what rights-based thinking has always understood. Rights function not as goods to be weighed against other goods, but as constraints on what we are permitted to do regardless of the cost / benefit calculation. We don’t ask whether freedom of speech delivers ROI. We don’t (or shouldn’t) evaluate the right to education, or to bodily autonomy, or to freedom of conscience in terms of economic return. We treat them as things that cannot be sacrificed even when the utilitarian maths might point that way. In the human rights sphere at least, the opposite way lies fascism.
Psychological safety belongs in this category of capabilities. Not as a productivity lever, but as a fundamental condition of dignity at work.
Noting the benefits is not the same as measuring utility
None of this means we should stop talking about what psychological safety produces. We should talk about it. Better innovation, improved quality, reduced harm, greater retention, stronger wellbeing are real and very important outcomes and they’re worth naming.
But there is a difference between noting the benefits of something and making those benefits the justification for its existence. We note the benefits of free speech in that it produces better public discourse, more accurate information, and more accountable institutions, without treating those benefits as the reason we defend it. The defence comes from somewhere more foundational.
And I wonder if the same shift is needed for psychological safety. We can point to what it produces but we should not make production the point.
What this means in practice
Framing psychological safety as a right rather than a utility makes the practice of it more honest. It means we stop asking “what might it cost us to foster psychological safety” and start asking “who are we currently excluding, and what does that cost them?”
It means that the people for whom creating safety is hardest: the people with the least power, who are the least visible or represented, and with the least resemblance to whoever set the norms, are no longer at risk of being deprioritised when budgets get tight or attention moves on. And it means we can hold a clearer line when the utilitarian cost/benefit pressure mounts (as it always eventually does) and someone asks whether all of this is really “worth it.”
The answer is that it’s not a question of worth. It’s a question of what kind of organisation, and what kind of working life, we think people are entitled to.
A while ago I read a post in a parenting forum that struck me for a few reasons. It was by a mum who had taken her daughter to the library, and had watched while her young daughter very politely approached an older woman and asked whether she would move her bag so that she could reach the computer station. For whatever reason, this woman had reacted badly and chastised her – told her off for interrupting – leaving her feeling embarrassed and upset.
I was first struck by how hard it is when we want to raise our children to have a voice, and the world isn’t always kind. But the reason the post had caught my eye in the first place was because the mum finished it by asking: “How can I protect her psychological safety when things like this happen?”
And it made me think about how much we can inadvertently stretch the concept of psychological safety beyond its real meaning, especially if it’s the nearest psychological shaped term we have to reach for. Psychological safety, at least using the definition we work with, is a group phenomenon – a shared belief held by members of a group that that group is safe for interpersonal risk taking. It has to be co-created, it can’t exist unilaterally, and it certainly can’t be retroactively applied to a stranger in a library. Sometimes psychological safety is just not there. It actually wasn’t a safe context for the daughter to ask her question without fear of humiliation, and no amount of reframing or reflection changes that.
But reading through the replies, I noticed one commenter gently point out that there’s another concept – psychological flexibility – that might be more appropriate here. I’m not going to attempt a deep dive into the origins of psychological flexibility here – the concept appears across several therapeutic traditions, though it’s perhaps most associated today with ACT: Acceptance and Commitment Therapy – but I think the commenter was right.
“Psychological flexibility, the cornerstone of ACT, refers to an individual’s ability to remain open to their experiences, stay present at the moment, and engage in actions aligned with personal values, even when faced with emotional or cognitive discomfort.“
Psychological safety is about the environment, but psychological flexibility is about our inner capacity for response – the ability to stay present with difficult feelings and still act in line with our values. This is different from the kind of individual resilience narratives we’re rightly sceptical of in organisational settings – it’s not about toughening up or pushing through. It’s about being able to sit with difficulty without it always defining your next move.
For the daughter in the library, there was no way of pre-empting that interaction. We can’t turn back the clock and ask the older woman to respond with grace, or hold her accountable for the impact of her words. Interactions with strangers in libraries sit rather outside the scope of most psychological safety work.
But developing the capacity to feel hurt, acknowledge it, not catastrophise it, and move forward? That’s something the mum can actually work on with her daughter and it sits much more firmly in the psychological flexibility realm.
And this isn’t just true for children navigating difficult strangers. I’d say the same is true of a lot of the work and interactions we manage out in the real world. We can do everything “right” – build better ways of working, foster more generous and supportive team cultures, model admitting mistakes and asking for help. And there will still be times when a grumpy email lands in our inbox, or someone reacts badly to a boundary we’ve expressed, or feedback arrives that feels deeply unfair. That will likely affect our felt sense of psychological safety in that relationship – of course it will, that’s a reasonable response. But it might also help if we know that we can be ok anyway and still work in a way that’s aligned with our values.
When psychological safety requires flexibility
There’s also a more direct relationship between the two concepts. Building psychological safety means, by definition, that people will speak up more – they’ll share dissenting views, ask more questions and challenge what isn’t working for them. That’s the whole point. But receiving that, with openness and curiosity rather than defensiveness asks something of us too. It’s easy to advocate for psychological safety when people are sharing what you want to hear. The real test is whether we can stay open when someone challenges a decision we were confident in, or says something we’d rather not have heard. Shutting down, getting defensive or signalling even quite subtly that a topic is off-limits is how psychological safety erodes. Staying present with that discomfort and choosing how to respond is, in a very real sense, psychological flexibility in action.
That’s what an awareness of the concept of psychological flexibility may offer alongside psychological safety work. We strive for better environments for people to work in (psychological safety) and know that we might need to support ourselves to find a steadier way of being in whatever environment we find ourselves (psychological flexibility). This isn’t an argument for organisations to focus on individual resilience instead of doing the environmental work – if anything, it’s the opposite. That caveat matters most where the environment is harmful; psychological flexibility is not a substitute for safety, and asking someone to sit with discomfort that is actually mistreatment is a different thing entirely. But for us as individuals, knowing both concepts exist and what each one can and can’t do for us feels like useful knowledge to carry.
Sometimes to understand a concept (such as psychological safety) better, we have to know where its edges lie. The library incident wasn’t a failure of psychological safety – it was a reminder that no single concept can carry the full weight of how we navigate the world. Knowing which tool to reach for matters.
Further reading:
Kashdan, T. B., & Rottenberg, J. (2010). Psychological flexibility as a fundamental aspect of health. Clinical psychology review, 30(7), 865–878. https://doi.org/10.1016/j.cpr.2010.03.001
Anusuya, S. P., & Gayatridevi, S. (2025). Acceptance and commitment therapy and psychological well-being: A narrative review. Cureus, 17(1), e77705. https://doi.org/10.7759/cureus.77705
Our wonderful Deisa Tremarias designed this sticker to reflect the various ideas of people such as Deming (“A bad system will beat a good person every time“) to help remind us that we’re all part of a system that influences our behaviour. Working on improving the system will usually yield greater outcomes than trying to change people.
Why “Fake It Till You Make It” Doesn’t Work for Psychological Safety
I remember early in a new job – one I was excited about and keen to make my mark in – I made a decision to be honest with my new team. I was out of my depth – I’d been asked to design workshops I didn’t have the expertise for, and my worry around messing them up was keeping me awake at night. Rather than bluff my way through it, I admitted I was struggling and asked for help.
The response in the team meeting was tumbleweed; no one spoke initially, then something closer to disdain emerged. What do you mean you don’t know enough? I understood pretty quickly that this wasn’t the kind of team where you did that.
I’d been brave. It hadn’t helped.
Every time we consider speaking up – raising a concern, asking a question, admitting a mistake or sharing an idea – we’re running a quick mental risk assessment, which is often part instinct, part experience and part reading the room. What’s the benefit of saying something? What’s the risk? And when the perceived cost outweighs the potential gain, we tend to stay quiet.
But it’s worth remembering that our calculation isn’t always accurate. Sometimes the risk we’re imagining isn’t real. We hold back on asking what feels like an obvious question, convinced others will think us foolish, only to find, when someone else eventually raises it, that the whole room was wondering the same thing. Or we agonise over telling our manager about a mistake, then discover they’re glad we did. Now they can help us fix it, and work with us on changing the context to make sure it doesn’t happen again. When the imagined risk turns out to be worse than the real one, speaking up rewards us, and often the whole team.
This is where the “fake it till you make it” idea of psychological safety gets its appeal. The thinking goes: if we all just lean into it a little more, act a little braver, we’ll start to build the habit of openness, and safety will follow. There’s a certain attractive logic to it.
But it actually only holds in one specific circumstance: when the risks were never real in the first place. Unfortunately, only too often the risks are entirely real.
We might find we ask the question and our teammates, as we may have feared, snigger, or we admit the mistake and our manager angrily kicks off. If speaking up does carry real consequences, then encouraging people to push through their hesitation doesn’t build psychological safety. It just confirms that their instincts were right to be cautious, and ultimately makes the environment less safe, not more.
This is why anyone working to improve team culture – whether you’re a leader, coach, HR professional, or a facilitator designing a workshop – should be cautious about programmes focused on making people braver. Encouraging people to take interpersonal risks before the environment is genuinely safe is a bit like encouraging someone to abseil off a cliff before you’ve checked the ropes. The amount of courage isn’t the problem. The conditions are.
So where does that leave the “fake it till you make it” idea? Is there any truth in it at all?
Actually, yes – but with an important caveat about who’s doing the “faking”. If we find ourselves in a position of power in an organisation or team, we may also find ourselves in the fortunate position where the interpersonal consequences for us of speaking candidly are relatively low.
Let’s say you’re a senior leader, a manager, or simply someone with significant standing in your team or organisation; you have the credibility and the protection that comes with your position. In that context, you going first (maybe with small, tentative first steps) is valuable. That might look like being the first to ask a basic question in a meeting, narrating your own uncertainty when making a decision, or owning a mistake openly, instead of glossing over it. These acts of modelling actively shift what feels permissible for others, demonstrating that honesty is welcome here.
That’s quite different from imploring people with less power or security to do the same. For them, the risks are real and the protection is thinner. Asking them to be braver without first ensuring the environment is genuinely safe is likely to be exposing rather than empowering.
The real work of psychological safety isn’t about adjusting people’s mindsets or building their courage. It’s about creating conditions where speaking up is genuinely low-risk: where questions are welcomed, mistakes are treated as learning opportunities, and different perspectives are consistently valued rather than punished.
Once those conditions exist, most people don’t need much encouragement to speak up – though it’s worth remembering that speaking up doesn’t look the same for everyone. For some people, voicing a thought in a fast-moving meeting, with no think time and a table full of louder voices, isn’t really an option – not because they lack courage, but because that particular format doesn’t work for them. A team lead who pauses to say “let’s take a minute to think about this quietly first”, or who follows up after a meeting with “I know we haven’t captured everything – please share your thoughts by email”, is creating the conditions for more voices to be heard. The brave thing becomes the normal thing, once we stop assuming that speaking up only counts if it happens out loud, in the room, in the moment.
Until then, instead of focusing on bravery, it’s worth asking what we’re doing to make it safer.
Spend any time on LinkedIn or other social media, and you may well come across posts and articles mentioning neuroscience or brain science, using neuro- as a prefix to invent terms like “neuroleadership”, or including images of brain scans or brain waves. If you google “the neuroscience of leadership”, you’ll find myriad articles, training courses and certifications, and images of glowing brains and neurons firing.
There’s an interesting reason for this: using neuroscience terms and imagery means people are more likely to believe the claims made, even if they’re false. Neuroscience seems clever, mysterious, advanced, intellectual, and objective, whilst sociology and psychology may be seen as more woolly, or at least less backed by “robust”, empirical evidence like brain scans.
The Seductive Allure of Neuroscience Explanations
Some of this neuroscience work is thoughtful, robust, and genuinely illuminating. But some of it, frankly, is entirely made up. And it’s often very hard to tell the two apart.
“It’s a popular fact that 90 percent of the brain is not used and, like most popular facts, it is wrong.” Terry Pratchett
Our human fascination with the brain is powerful and universal. Neuroscience carries enormous epistemic authority. That authority is well earned, but it is also far too easily borrowed.
There’s a name for this. It’s called the SANE effect: the Seductive Allure of Neuroscience Explanations, and was coined in a 2008 paper by Deena Weisberg and others in the Journal of Cognitive Neuroscience. Explanations of psychological phenomena appear more credible when they contain neuroscientific-sounding information. Even completely irrelevant neuroscience information appears to interfere with people’s abilities to critically consider the claims being made. Worryingly, the added neuroscience had a particularly strong effect on bad explanations, masking issues with claims that would otherwise have been obviously false.
And the SANE effect is not just an effect on laypeople – students of neuroscience showed the same biases. Only true experts – people with deep experience and expertise in the field – were largely (but still not always) immune.
This excellent piece on the SANE effect by Christian Jarrett shows that there’s, understandably, depth and nuance to it. Some replications of the Weisberg studies show strong effects, and others weaker ones. Other factors such as longer explanations, visual richness, writing style, and extra detail can all affect perceived credibility, regardless of the actual content. And this points to the underlying mechanism of the bias, because the SANE effect is a natural heuristic; i.e. we rarely have the time or expertise to evaluate claims based purely on their logic and evidence, so instead we rely on cues and shortcuts like technical language, imagery, and reference to fields and mechanisms that sound more objective and robust. We often find explanations satisfying because they look and sound right, not because they are structurally robust.
The Reverse SANE Effect
The issue of course is not that neuroscience-related claims are always wrong, it’s that the presence of neuroscience words and images can make explanations feel more scientific, more authoritative, and more convincing than they actually are. Neuroscience is frequently co-opted to establish unwarranted credibility for the claims being made.
In itself this can backfire – as a result of the massive misuse and co-opting of neuroscience, I’d like to suggest the “Reverse-SANE Effect”, where it works the other way around, and mentioning neuroscience actually reduces the perceived credibility of a claim. I often find myself intuitively dismissing claims precisely because they mention neuroscience or have an image of a brain scan attached. I’m aware that this isn’t useful, or fair, but it’s a result of the overwhelming deluge of neuroscience-associated claims, and it’s something I try to recognise when I’m doing it.
“If thought corrupts language, language can also corrupt thought.” George Orwell
Language Complexity
One of the mechanisms behind the SANE effect is language, but it’s not that clear-cut. In a beautifully titled paper from 2006: “Consequences of erudite vernacular utilized irrespective of necessity: Problems with using long words needlessly.”, Daniel Oppenheimer explored what happens when writers use overly complex language to sound more intelligent. The prevailing belief among students was that bigger words make you look smarter, but they were wrong. Across multiple experiments, essays rewritten with more complex vocabulary were consistently judged as coming from less intelligent authors.
Oppenheimer (not that one) used the term “processing fluency” to describe how when something is harder to process, we tend to evaluate it more negatively. “Needless complexity leads to negative evaluations.” So – if we write as clearly and simply as we can, we’ll be more likely to be thought of as intelligent.
“If you care about being thought credible and intelligent, do not use complex language where simpler language will do.” ― Daniel Kahneman
So, alongside the SANE effect, a fascinating tension appears – complex language can make writing feel less intelligent, whilst scientific-sounding detail can make explanations feel more convincing. Whilst it seems paradoxical, it isn’t: in both cases, we’re not judging the idea itself – we’re judging signals. When an explanation is sprinkled with technical terminology, references to neuro-things, brain systems, or claims of biological grounding (or even “quantum” on occasion – yes, there’s even “quantum leadership”), it invokes expert power and signals academic authority. It also, importantly, raises the perceived cost of disagreement – often on purpose. It’s psychologically harder to disagree with a claim made using complex, neurosciency terminology, especially when we’re not experts in neuroscience.
“Science is the belief in the ignorance of experts.” Richard Feynman
SANE and Psychological Safety
Interestingly, whilst there have been claims that psychological safety itself is a “SANE Effect” term, it doesn’t appear to be the case, at least not measurably so. The effect isn’t triggered by the word “psychological” alone. The SANE effect is specifically activated by neuroscience terminology and imagery, not psychological terminology. In fact, many of the original SANE studies examined psychological phenomena precisely because adding brain-based language made those explanations more persuasive than they would otherwise be.
However, the field of psychological safety is certainly vulnerable to people abusing the SANE effect. We’ve seen an increase in the popularity of psychological safety explanations that lean heavily on neuroscience or terms like “rewiring neural pathways” as language to make certain (often rather wild) claims seem more robust and credible.
What do we do with the SANE Effect?
In fact, it’s worth noting that we all do versions of this at times; Oppenheimer notes that people often use more complex language when we feel insecure – the complex language becomes a shield, a (possibly false) signal of expertise and experience (or at least more expertise and experience than the person we’re making the claim to). Complex language can be a way that we attempt to be taken more seriously. It can also, of course, be a joy: many of us love to indulge in evocative, flowery language, and finding a term that means a very specific, obscure thing can be highly satisfying. I recently came across a wonderful usage of the word oleaginous to describe someone, and it was satisfyingly perfect.
So what do we do about the SANE Effect? We certainly shouldn’t reject neuroscience, and nor should we flatten everything into oversimplified language. What is powerful is to learn to notice when we’re potentially being seduced by it. Sometimes complex language and neuroscience is doing real work, and sometimes the right word is the obscure but precise one. We can, however, check in with ourselves when we come across neurosciency explanations and ask some key questions:
What does the neuroscience actually add? Does it actually deepen our understanding of the claim, help explain the mechanism, or affect what we do in practice? If not, it might just be marketing.
Would the explanation still stand without it? If removing the neuro- or brain reference doesn’t weaken the argument, it probably wasn’t doing much work.
Is this helpful precision, or theatre? Sometimes complexity is necessary. Sometimes it’s just costume – the Emperors’ New Clothes.
Does this language make challenging the claim easier or harder? If an idea feels harder to question because of how it’s framed, it might be a sign that the people making the claim don’t want it to be challenged.
None of this should be interpreted as an argument against neuroscience, expertise, or complexity where it genuinely helps us understand the world. But it is a reminder to remain a little sceptical of ideas that arrive dressed in the aesthetics of robust science. I don’t think the danger is that people will believe nonsense; it’s that weak ideas sometimes get a free pass when they sound neurosciency enough.
Further reading:
Weisberg, D.S., Keil, F.C., Goodstein, J., Rawson, E. and Gray, J.R., 2008. The seductive allure of neuroscience explanations. Journal of cognitive neuroscience, 20(3), pp.470-477. https://pmc.ncbi.nlm.nih.gov/articles/PMC2778755/
Oppenheimer, D.M. (2006) Consequences of erudite vernacular utilized irrespective of necessity: Problems with using long words needlessly. Applied Cognitive Psychology: The Official Journal of the Society for Applied Research in Memory and Cognition, 20, 139-156. https://cahill.people.unm.edu/480-21/Oppenheimer-2006-Applied_Cognitive_Psychology.pdf
Often, when we’re trying to improve how work gets done, we start with principles. We agree what we believe in, or are working towards, at a high level, and then figure out how those ideas translate into day-to-day practice. This piece takes a different route.
Instead of beginning with abstract principles, I want to start with a concrete practice: Learning Teams, from the world of HOP (Human and Organisational Performance). Through the excellent work of Sam Goodman and others, Learning Teams have evolved into a reasonably well-defined way of bringing people together to learn from everyday work.
By looking closely at how Learning Teams are designed and facilitated, we can surface the deeper principles they rest on – principles that reach far beyond pure safety contexts, and have important things to say about learning, power and psychological safety at work.
Let’s dive in…
What Are Learning Teams?
Learning Teams are a way of creating space to learn from events at work, from work as it is done, not work as imagined or prescribed in tidy process maps or policies. They focus on real work – the complex, sometimes messy and often adaptive way that real work gets done, shaped as it is by context, constraints and trade-offs.
The purpose of a Learning Team is to bring together the people who are closest to the work, those at the sharp end, to share their knowledge of how the work really happens, so that the organisation can learn.
Learning Teams can lead to many different outcomes, but there is only one true requirement: that we learn more about “normal work.” If a Learning Team helps an organisation better understand how work is actually done, then it has been a success.
Events with Learning Potential
When we get stuck into the world of organisational learning, it quickly becomes clear that, in theory, we could learn from absolutely everything that happens at work – the good, the bad and the truly horrific.
In practice though, running a Learning Team requires an investment of time, so it simply wouldn’t be possible to do one for everything that happens at work. So the question becomes, which events are most worth learning from?
Rather than automatically choosing the things that went most wrong or had the most severe impact, Learning Teams invites us to consider which events have the most learning potential (as does PSIRF). This might be a failure or breakdown, but it might also be something that went unexpectedly well, or a situation that could have gone wrong but didn’t, perhaps a “good catch”.
We’re often looking for “goldilocks” events that sit in the middle ground. Not so rare that they tell us very little about everyday work (a meteorite strike), but not so common that we’re already well aware of what’s happening. Not so severe that they automatically trigger a formal investigation, with all the emotional, regulatory or legal burden that can bring, but not so trivial that there’s nothing very interesting to explore.
This isn’t a precise science. We can never know in advance exactly what learning a conversation will surface. But when deciding whether an event is a good candidate for a Learning Team, it can be helpful to weigh up factors such as:
the effect on the team
scale and impact
frequency
how generalisable the learning might be
regulatory or legal implications
alignment with existing improvement work
When we explore Learning Teams in our HOP workshops, this is often the point where people start to see themselves in it. The examples of what might trigger a Learning Team vary wildly, depending on the kind of work people do.
For some, it’s a spike in complaints after a shiny new IT system goes live. For others, it’s a team restructure that didn’t quite go as planned. And sometimes it’s something more positive – like a particular store or location suddenly outperforming the rest, leaving everyone asking, what on earth are they doing differently?
It doesn’t really matter whether these are successes or failures, as long as they spark some organisational curiosity. We can see all of these as opportunities to ask better questions about how work is really getting done, and what people are adapting to in order to make things work.
Roles in a Learning Team
Once we’ve chosen an event with learning potential, the next question is to decide who to include in the Learning Team.
There are a small number of distinct roles, each with a specific purpose:
Owner: The Owner is someone with enough organisational authority to get the Learning Team off the ground. They ensure the right people are invited, and they commit to championing any organisational actions that emerge afterwards. Crucially, the Owner opens the session, welcoming people and setting the intent, and then leaves. (We’ll explore why this matters shortly.)
Facilitator: Sometimes referred to as the Coach, they are there to facilitate the learning, not to be the technical expert. They don’t have detailed knowledge of the event being discussed, but they are skilled in holding the process. They set expectations, often through a social contract, keep the conversation on track, and help the group stay focused on learning.
Scribe: The Scribe plays another essential, but often underestimated, role. Like the Facilitator, they are not someone with direct knowledge of the work being discussed. Their focus is on capturing the qualitative richness of the conversation as it unfolds. They manage the “wall of discovery”, where insights and learning are made visible to the group.
Team members: There may be two, three, or more team members in the Learning Team. These are the people closest to the work, who hold first-hand knowledge of how work is actually done at the sharp end and ideally, they were present during the event being discussed. Their primary role is to share that knowledge. They may identify improvements along the way, but that is not their main aim.
Learning Teams may also include people who weren’t directly involved in the specific event, but who do similar work or interact with the same system. This broadens any learning beyond a single episode and helps surface patterns in normal work that might otherwise remain invisible.
Structure of a Learning Team
One of my favourite things about Learning Teams is their intentional structure. They are deliberately designed in two distinct parts, with a pause in the middle.
Before the Learning Team itself, there is some preparation required: being clear about the focus of the learning, identifying who needs to be involved, and making the practical arrangements to bring people together. Then we have the meetings:
Meeting 1: Learning from the work The first meeting typically lasts around 60–90 minutes – long enough to allow for depth, but not so long that we lose focus. If it feels like we need more time than this, it’s often a signal that the event we’ve chosen is too broad and needs narrowing.
The sole purpose of this meeting is to understand what happened, or what happens, in the flow of real work. This is not a space for jumping to solutions or improvements, however tempting that might be. The Facilitator’s role here is to protect that intent, and they may need to be quite firm about it! This is in deliberate contrast with many “traditional” investigations, which tend to focus on a single incident and work backwards to identify where things failed. Learning Teams spend proportionally more time understanding how work usually succeeds, what people were adapting to, and why actions made sense in their context at the time.
“Soak time” Between the two meetings, it’s helpful to allow some “soak time” – often around a day. This pause gives people space to process and reflect on what emerged, and sometimes realise what was missing or left unsaid in the first conversation.
Meeting 2: Making sense and improvements The second meeting brings the group back together. The Facilitator supports the group to recap what was learned in Meeting 1, invites them to add any new reflections, and only then opens up the conversation about possible improvements or changes. This meeting is usually a similar length to the first.
Outputs of Learning Teams
As already touched on, the only essential outcome of a learning team is an improved understanding of normal work.
Sometimes, depending on the event, this learning will naturally surface areas where things can be improved, and maybe even some ideas for improvement. But improvements are a possible by-product of Learning Teams, not their primary purpose.
A Learning Team that ends with richer insight, a more nuanced understanding of work, and perhaps the ability to ask better questions about how work happens, has still been successful – even if it doesn’t produce a tidy list of actions.
Evidence supports this shift in emphasis. Interestingly, a comparative study in a large NHS hospital found that Learning Teams generated more actions overall, and a substantially higher proportion of system-focused improvements, than conventional root cause analyses, possibly because they explore work across multiple contexts rather than compressing learning into a small number of technical or individual “root causes”.
Reducing Power Gradients
The success of a Learning Team hinges on psychological safety. Do the people in the team feel safe to be honest about what happened? If not, we’re unlikely to learn very much.
This sits at the heart of why the Owner of a Learning Team will usually step out after the introductions. The Owner is, by definition, someone with greater organisational authority – they have the clout to convene the Learning Team and the responsibility to act on what emerges afterwards. But their presence in the discussion itself can unintentionally introduce a power gradient that makes candour harder.
This isn’t because of bad intention, or a failure on their part to reduce the power gradient. It’s a very pragmatic recognition that hierarchy shapes behaviour. When someone with formal authority is in the room, people are more likely to soften their accounts, or steer away from what feels risky to say.
There’s also a more practical reason for the Owner leaving. The Owner typically doesn’t hold first-hand knowledge of the work at the sharp end; their role in the Learning Team is to enable learning and respond to it, not to be the source of it. By stepping out, the Owner helps create the conditions for more open, real conversations, where the people closest to the work can speak with fewer filters.
A different way of learning
In many organisations, when something happens that we want to learn from, we default to an investigation. A group is brought together, often with more senior people in the room, and the focus quickly turns to understanding what went wrong, often who did wrong, and what needs to change.
Learning Teams take a different approach.
Rather than being judged from on high, the people who do the work are treated as the ones who know most about it. Time is spent first on understanding what work looks like in practice, before determining whether anything needs to be fixed. We put a bit of space between learning and decision making.
“I learned very early the difference between knowing the name of something and knowing something.” ― Richard P. Feynman
This doesn’t mean Learning Teams replace formal investigations altogether. Some events will always need a formal response. But if we’re serious about learning from everyday work, not just from catastrophic failures, then we need approaches that make it easier for people to speak honestly about how work actually gets done.
That, for me, is what Learning Teams really offer. They invite us to take a better stance, one that starts with curiosity, treats people at the sharp end as experts in the work, and accepts that understanding has to come before improvement.
Taking it further
Interested to learn more? Join our next HOP Fundamentals workshop, where we explore Learning Teams and much, much more.
Robbins, T., Tipper, S., King, J., Ramachandran, S.K., Pandit, J.J. and Pandit, M., 2021. Evaluation of learning teams versus root cause analysis for incident investigation in a large United Kingdom National Health Service Hospital. Journal of Patient Safety, 17(8), pp.e1800-e1805.
Login
