When Amy Edmondson first went looking for psychological safety, she wasn't looking for it at all. As a Harvard doctoral student in the late 1990s, she set out to show that the best hospital nursing teams made fewer medication errors. The data said the opposite: the better teams reported more errors, not fewer. Her hypothesis looked dead on arrival - until she realized the better teams weren't making more mistakes. They were just willing to talk about them. The struggling teams were sitting on theirs.

That single counterintuitive finding is the whole idea in miniature, and it's why "psychological safety" is worth rescuing from the buzzword pile it's currently sliding into. Like most popular phrases, it's at risk of being stretched until it means nothing. So let's be precise. It does not mean a workplace free of pressure, hard feedback, or high standards. It means something narrower and far more useful: that people can admit a mistake, ask for help, or raise a concern without fear of being humiliated or punished for it.

In high-pressure environments, that distinction isn't academic. It's the whole thing.

Pressure and safety are not opposites

There's a tidy assumption floating around that demanding workplaces have to trade away safety, and that safe workplaces must be soft. It's tidy, and it's wrong. When Google ran its multi-year Project Aristotle study across more than 180 teams and 250-odd variables, it expected the answer to high performance to be talent - the smartest people, the best mix of skills. What it actually found was that who was on a team mattered far less than how the team worked together, and that psychological safety was the single strongest predictor of effectiveness, ahead of dependability, structure, individual IQ, and experience.

The teams that perform best under pressure tend to be the ones where people feel safe enough to flag problems early, surface risks, and ask the awkward question before a small issue becomes an expensive one. Safety isn't the absence of pressure. It's the presence of trust under pressure - which is precisely why Edmondson's later work on high-risk settings, including the Columbia shuttle disaster, found that safe teams catch ambiguous warning signs before they escalate, and unsafe ones don't.

Edmondson maps this onto two axes: psychological safety and accountability. Low safety with high accountability doesn't produce diligence - it produces an anxiety zone, where people are held to a high bar but too afraid to say when something's going wrong. High standards without safety don't get you performance. They get you a very quiet room full of hidden problems.

And hidden problems don't stay small. When people don't feel safe, they go quiet, and in a high-pressure setting silence compounds fast: risks stay buried, errors stack on errors, and the person closest to the problem - the one who could have flagged it first - is the least able to. The chain is predictable, which is also the good news, because a predictable chain can be broken.

How silence compounds under pressure - and where psychological safety breaks the chain.

What psychological safety is - and isn’t

Because the phrase gets thrown around so loosely, it helps to draw the line clearly:

The thing it is most often mistaken for is comfort. It isn't comfort. It's what lets a team have the healthy friction - the honest debate, the pointed question - that actually improves the work.

What erodes it

Psychological safety is easy to lose and slow to rebuild. It rarely collapses in one dramatic moment; it erodes quietly, one reaction at a time. A single dismissive response in a meeting can teach a dozen people to stay silent next time - and they will remember it far longer than you do. It erodes when:

• Mistakes are met with blame rather than curiosity.
• Speaking up is rewarded in theory and penalised in practice.
• Leaders react defensively to bad news.
• Pressure is allowed to curdle into a culture of fear about consequences.

What builds it

It's built less through policy than through repeated behaviour - and, uncomfortably for anyone hoping to delegate this, mostly the behaviour of leaders. Responding to bad news with "tell me more" instead of "who did this." Admitting your own mistakes out loud. Treating questions as a sign of engagement rather than a gap in someone's competence. Following through when someone raises a concern, so people learn that speaking up actually changes something rather than vanishing into a void.

None of these are grand gestures. They're small, daily signals, and their cumulative effect on how a team functions under pressure is enormous.

Four signals leaders can send this week

1. Lead with "thank you for flagging this" before jumping to problem-solving - reward the disclosure, not just the fix.
2. Name one of your own mistakes, and what you learned from it, out loud.
3. Treat a question as engagement, not as evidence someone doesn't know enough.
4. Close the loop: tell people what actually changed because they spoke up.

Why it belongs on the leadership agenda

Psychological safety is not a wellbeing nicety that lives in a separate box from performance. In Google's data, teams with high psychological safety had lower turnover, brought in more revenue, and were rated effective by leadership roughly twice as often. It underpins the early disclosure, honest feedback, and fast problem-solving that high-pressure work runs on.

It's also tightly bound up with mental health. A workplace where people can be honest about how they're actually doing is a workplace where strain gets noticed and addressed before it hardens into a crisis. The same silence that buries a project risk buries a struggling colleague - and both surface later, at a much higher cost.

Medaca helps leaders build workplaces where mental health and high performance coexist rather than compete. Reach out to explore what that looks like in your organization.