Psychological Safety in the Boardroom: Candour, Not Comfort
Psychological safety is the shared belief, within a team, that it is safe to take interpersonal risks: to ask a question, admit an error, or challenge a superior without fear of humiliation or reprisal. It is not comfort, niceness, or lowered standards. In high-stakes leadership, it is the mechanism that decides whether the information a team already holds ever reaches the person making the decision.
The term has travelled a long way from its origins, and much of what is now said about it would be unrecognisable to the researcher who made it tractable. Psychological safety entered the management literature through Amy Edmondson's 1999 study in Administrative Science Quarterly, which began with an anomaly. Studying error rates across hospital nursing units, Edmondson found that the teams with the best working relationships appeared to make more mistakes, not fewer. The finding was counter-intuitive enough to warrant a second look. On closer inspection, the better teams were not erring more often; they were reporting more, because their members felt safe enough to surface errors rather than bury them. The difference was not competence but disclosure. Where status and fear governed the unit, mistakes went unspoken and uncorrected; where candour was possible, problems were caught and the team learned. Edmondson formalised this into a precise, testable construct, which she defined as a shared belief that the team is safe for interpersonal risk-taking, and tested it in a parallel field study of fifty-one work teams in a manufacturing firm. The result was clear: psychological safety predicted learning behaviour, and learning behaviour, in turn, predicted performance.
That chain of reasoning matters because it locates the value of psychological safety in something concrete rather than sentimental. The construct is not a measure of how warm a team feels; it is a measure of whether people will do the uncomfortable, performance-relevant things that organisations depend upon: ask for help, question an assumption, report a near-miss, dissent from a confident superior. By defining it at the level of the team and supplying an instrument to measure it, Edmondson turned a vague intuition about "good culture" into a variable that could be studied, compared, and connected to outcomes. This is also why the concept has held up better than most ideas in leadership psychology, a field not short of fashionable claims that evaporate under scrutiny. The most comprehensive test to date, a meta-analytic review by Frazier and colleagues published in Personnel Psychology in 2017, pooled 136 independent samples covering more than 22,000 individuals and nearly 5,000 groups. It found that psychological safety is reliably associated with task performance and with the discretionary, citizenship behaviours, such as volunteering information and helping colleagues, that lubricate any complex organisation. The associations are moderate rather than overwhelming, and they operate largely through learning behaviour rather than directly. That is an honest result, not a marketing one: psychological safety is a genuine enabling condition, not a lever that mechanically raises output.
The clearest demonstration of what its absence costs is not found in a spreadsheet but in a catastrophe. In 2003, the Space Shuttle Columbia broke apart on re-entry, killing all seven astronauts aboard. The proximate cause was physical: a piece of foam had struck the wing during launch. The deeper cause, as Edmondson has since argued and as the Columbia Accident Investigation Board documented, was organisational. An engineer named Rodney Rocha suspected the foam strike might be serious and wanted satellite imagery of the wing. In a mission management meeting of some fifty-three people, with senior managers discussing the very problem he was worried about, he said nothing. Asked later why, he described the hierarchical distance: the decision-makers were, as he put it, "way up here" while he was "way down here." The information that might have changed the response existed inside the organisation; it simply never travelled the short distance from the person who held it to the people who needed it. NASA's culture, the investigators found, had come to treat seniority as a proxy for being right and to marginalise dissenting technical voices. This is the texture of leadership under pressure that the data only abstracts: a steep hierarchy, a confident room, and a quiet engineer who calculates that speaking up is not worth the risk. The lesson Edmondson draws, recounted in a 2024 Harvard Business Review discussion of the disaster, is uncomfortable for any leader who prizes decisiveness: the more authority a person projects, the more they must work to make dissent safe, or they will simply stop hearing it.
Candour, not comfort
The most common error leaders make with psychological safety is to confuse it with its opposite. In its diluted, popular form, the term has come to mean a workplace where no one is criticised, deadlines are negotiable, and discomfort is treated as a hazard. Edmondson has spent much of the last decade correcting this reading, noting in interviews such as her 2019 Harvard Business Review conversation that the word "safety" misleads people into hearing coziness, when what she means is candour: the willingness to be direct, to take interpersonal risks, and to say plainly, "I got that wrong." The distinction is not pedantic. In her 2018 book The Fearless Organization, Edmondson sets out a simple matrix that crosses psychological safety against accountability and high standards. Where safety is high but accountability is low, the result is a comfort zone: people speak freely, but little of consequence is demanded or delivered. Where standards are high but safety is low, the result is an anxiety zone, in which people are held to demanding targets yet are too frightened to admit problems, hide bad news, and game the metrics. Neither produces performance. Only the combination of high safety and high accountability creates what she calls the learning zone, in which people can be both candid and stretched. For leaders in high-pressure environments, this is the operative point: psychological safety is not an alternative to demanding standards, it is the precondition that allows demanding standards to be met without the organisation blinding itself to its own failures.
This reframing has direct consequences for how power behaves in a room. Psychological safety is set far more by the powerful than by the powerless, because the climate of a team is calibrated by how its most senior member responds to the first piece of inconvenient information. A chief executive who reacts to early bad news with visible irritation teaches the room, in a single moment, that future bad news should be managed rather than disclosed. A founder who treats a board member's challenge as disloyalty trains the board to nod. A politician who punishes the adviser bearing an unwelcome poll learns to be surrounded by flatterers. Status distance, the very thing that confers authority, is also what suppresses upward candour, and it does so most powerfully precisely when the stakes are highest and the hierarchy steepest. This is the paradox that makes the construct so relevant to senior leadership rather than to junior teams. The behaviours that build psychological safety are unglamorous and specific: framing work as uncertain rather than settled, asking genuine questions rather than rhetorical ones, responding to disclosure of error with curiosity rather than blame, and conspicuously thanking people for views that contradict one's own. None of this requires lowering the bar. It requires the emotional discipline to separate one's reaction to a message from one's judgement of its bearer, a skill that does not come naturally to people who have risen by being right.
It is worth being clear about what the evidence does not say, because the field is now commercially crowded. The market for executive coaching and leadership development is large and growing, valued at roughly 113 billion US dollars in 2026 and expanding at around nine per cent a year, and "psychological safety" has become one of its more heavily marketed phrases. Much of what is sold under the label is thin: an away-day, a poster, a values statement. The construct itself is robust, but it is not a panacea, and the meta-analytic effects are moderate, context-dependent, and mediated by behaviour that has to be sustained rather than announced. Psychological safety will not rescue a strategy that is simply wrong, nor substitute for competence, nor survive a leader whose stated openness is contradicted by how they actually behave when challenged. Its value is narrower and more durable than the slogan suggests: in conditions of genuine uncertainty and consequence, it determines whether the dispersed knowledge a leadership team possesses is pooled or withheld. That is not a soft benefit. In a boardroom weighing an acquisition, a cockpit, an operating theatre, or a cabinet room, the cost of one unspoken concern can dwarf the cost of every training programme an organisation will ever buy.
Leaders who operate where decisions carry significant and irreversible consequence may wish to treat the conditions for candour with the same seriousness they bring to the decisions themselves. The relevant question is not whether a team likes its leader, but whether the person holding a dissenting view in the next high-stakes meeting will actually voice it, and whether the leader's own conduct in that moment will make the next person more or less likely to do the same. Building that capacity is rarely a matter of insight alone; it is a matter of practised behaviour under pressure, and it is the kind of work that bespoke leadership advisory exists to support. The organisations that learn fastest are not those that make the fewest mistakes, but those whose people are willing to name them while there is still time to act.
Frequently Asked Questions
What is psychological safety in leadership?
Psychological safety is a shared belief within a team that members can take interpersonal risks, such as admitting mistakes, asking questions, or challenging decisions, without fear of humiliation or punishment. In leadership, it determines whether critical information reaches decision-makers in time for them to act on it.
Is psychological safety the same as being comfortable or nice?
No. Psychological safety is not comfort, niceness, or lowered expectations. Amy Edmondson, who defined the term, describes it as candour: the willingness to be direct, take risks, and admit failure. It pairs with high standards rather than replacing them, and is frequently misread as permissiveness.
Does psychological safety reduce accountability?
No. The evidence and Edmondson's own framework suggest the opposite. Safety without accountability produces a comfort zone, where people speak freely but achieve little. High performance requires both: an environment safe enough for candour, combined with demanding standards. The two reinforce one another rather than conflict.
What did Amy Edmondson's research actually find?
Edmondson's 1999 studies found that psychological safety predicts learning behaviours, such as seeking feedback and discussing errors, which in turn improve team performance. A striking result was that better-functioning teams reported more errors, because members felt safe enough to surface problems rather than conceal them.
Why do people stay silent in important meetings?
People withhold concerns when the perceived personal risk of speaking outweighs the benefit, especially across steep hierarchies. Status distance, fear of appearing incompetent, and uncertainty about how a leader will react all suppress candour. The result is that information a group holds may never reach the decision-maker.
