Psychological safety appears in almost every conversation about high-performing teams. It features in leadership competency frameworks, team development programmes and culture surveys. Most organisations say they are working toward it, and many believe they already have it.
In practice, genuine psychological safety is rarer than it’s claimed, and part of the reason is that the concept, while well-defined in research, has become somewhat blurred in application. It gets equated with comfort, or friendliness, or the absence of conflict. These things are not quite the same, and confusing them tends to produce the appearance of safety without its substance.
A more concrete definition
Amy Edmondson’s research defines psychological safety as the shared belief among team members that the team is safe for interpersonal risk-taking. Let us break that down, because each element matters.
Shared belief: not a policy, not a leadership statement, but a genuine lived experience that’s shared across the team. You can have a leader who believes in psychological safety, a team that has been told they have it, and individuals who do not actually experience it. The shared belief is what matters, and it’s established through experience, not declaration.
Safe for interpersonal risk-taking: specifically, the kind of risks involved in being honest at work. Raising a concern. Disagreeing with a decision. Admitting uncertainty or error. Asking a question that might reveal ignorance. Challenging an assumption. These are the acts that psychological safety makes possible, and they’re the acts that are suppressed when it’s absent.
It isn’t the absence of challenge, nor the presence of uniformly positive interactions. Psychologically safe teams can disagree, challenge, and address difficulty, what they don’t do is require their members to manage around the truth to avoid social consequences.
What psychological safety feels like from the inside
One of the most useful ways to understand psychological safety is through the felt experience of its presence and its absence.
When it’s genuinely present, you don’t need to prepare carefully before raising a concern. You don’t calculate the political risk of asking a question that might reveal you don’t know something and you don’t experience a low-level anxiety about whether disagreeing will affect your standing. You can be straightforward because the experience of this team has taught you that straightforwardness is safe here.
When it’s absent, all of those calculations are present, often invisibly. You filter what you say before you say it. You watch the room for signals about what’s welcome and what isn’t. You manage your presentation of yourself to minimise risk. You’re present in the meeting but not fully available to the conversation, because part of your attention is occupied with managing your safety.
Most people can identify teams where they’ve felt genuinely safe, and teams where they haven’t. The quality of thinking, of conversation, of collaborative output is usually very different between them.
Why it cannot be created by policy
The most important thing to understand about psychological safety is that it can’t be created by organisational policy or cultural statement alone. It’s produced by experience: specifically, by the accumulated experience of what actually happens when people take interpersonal risks in this team.
If a team member raises a concern and it’s received openly, incorporated into the conversation, and responded to with genuine engagement, the nervous system learns: this is safe. If a team member raises a concern and is dismissed, made to feel they have overstepped, or subtly managed, the nervous system learns the opposite, regardless of what the values statement on the wall says.
This is why psychological safety is so dependent on the nervous system regulation of the people in authority. The responses of leaders in moments of interpersonal risk are the data from which psychological safety is built or eroded. A leader who is regulated enough to receive challenge without becoming defensive, who can sit with uncertainty without projecting threat, who can acknowledge error without self-punishment that frightens everyone around them: that leader creates safety through their behaviour, not their intentions.
The individual and the collective
Genuine psychological safety is built at two levels simultaneously.
At the individual level, it requires each team member to develop enough self-awareness and relational skill to contribute to safety rather than undermine it: to raise concerns rather than manage around them, to stay present in difficult conversations rather than withdrawing, to repair when things go wrong rather than avoiding what is broken.
At the collective level, it requires patterns of interaction that accumulate into a shared experience of safety, not a programme, an away day, and not a values exercise: the daily accumulation of how people actually treat each other in the ordinary moments of work.
Neither level produces safety on its own, and neither can be shortcut by claiming to have what isn’t there yet.
