In Amy Edmondson’s foundational work on psychological safety, one finding keeps surfacing across decades of research: teams full of capable, informed people routinely fail to surface what those people actually know. The information exists. It simply never reaches the table.

The reason is not laziness or disengagement. Before most acts of speaking up, a fast, mostly unconscious calculation runs in the background, weighing what the contribution might cost against what it might earn. Some people are not quiet in meetings because they have nothing to say. They are pricing risk.

That calculation is one of the most underrated forces shaping how teams actually make decisions. The person staring at the table is not checked out. They are doing maths.

The silence is arithmetic, not absence

Researchers who study workplace communication have a name for what is happening in that head. Before most acts of speaking up, a rapid and largely unconscious risk assessment runs in the background, weighing the likely costs of voice against its likely benefits. Scholars describe it as a cost-benefit calculus of voice: when the perceived risk of speaking outweighs the perceived payoff, people lean towards silence.

The costs are rarely about being wrong. They are social. Losing standing in the group, being seen as difficult, creating extra work for everyone, or being remembered as the one who reopened a meeting that was nearly over.

The silence is not absence. It is arithmetic.

The rules nobody wrote down

In 2011, James Detert and Harvard’s Amy Edmondson published research on what they called implicit voice theories: taken-for-granted beliefs about when speaking up at work is risky or inappropriate. These are not policies. They are unwritten rules people absorb and then obey without noticing.

The researchers identified a handful of recurring ones. Do not embarrass the boss in public. Do not go over the boss’s head. Do not speak up unless you have solid data and a fully formed solution. Each sounds reasonable in isolation. Together they form a quiet filter that screens out a great deal of useful input before it is ever spoken.

The striking part was how widely these beliefs are held, and how reliably they suppress even constructive, pro-organisation suggestions. People are not staying silent because they do not care. They have learned, somewhere along the way, that the cost of caring out loud is too high.

The fear is not irrational

It would be comforting to file all of this under insecurity, a problem to be coached away. The research is less flattering to managers than that.

Across field and experimental studies, Ethan Burris found that managers tend to rate employees who raise challenging ideas as weaker performers than those who offer supportive input, and they endorse the challengers’ ideas less. Perceptions of threat and loyalty did the work behind those judgements.

In other words, the quiet calculator is often correct. Speaking up with something that questions the plan can cost you, not because the idea is bad, but because challenge reads as disloyalty to the person being challenged. The colleague weighing whether to talk is not imagining the downside. They have watched it land on someone else.

Who pays the most to speak

Not everyone runs the same equation. The price of voice tends to rise with how little status you hold in the room. Newer hires, junior staff, and anyone who senses they sit lower in the informal hierarchy face steeper perceived costs, because they have less standing to spend and more to lose. Researchers describe this more broadly as employee silence: people withholding ideas, concerns or information despite having something useful to say. Power dynamics matter here, because people are less likely to speak when they believe the risks of doing so outweigh the benefits, especially when their input could be judged negatively by a senior colleague.

This is one reason the quietest person in a meeting may be the most junior, and why their silence is so easy to misread as having nothing to add. The opposite is frequently true. They are paying the closest attention to the social cost, precisely because they can least afford to get it wrong.

What actually changes the maths

This is where Edmondson’s better-known idea earns its keep. Psychological safety, the shared belief that a team is safe for interpersonal risk-taking, is not a soft perk. It is one of the variables that most directly changes the inputs to that internal cost analysis.

When people can predict that a question or a half-formed idea will be met with curiosity rather than punishment, the perceived cost of voice drops and the arithmetic tips towards speaking. Edmondson has pointed out that teams without it can be ruthlessly efficient and still arrive at the wrong answer, because the people who saw the problem coming chose to stay quiet.

The cost of that silence is not abstract. Surveys consistently find large shares of staff holding back input they believe their organisation needs. UK research from the CIPD found that roughly a quarter of employees report high levels of silence at work. Every one of those is a small calculation that ended in “not worth it.”

For the people building teams

For founders and managers, especially in fast-moving startups where the loudest voice often sets the direction, the practical takeaway is uncomfortable but simple. The quietness in the room is data about the room, not about the people in it.

Edmondson’s own prescription is unglamorous: stop waiting for voice to emerge organically and go and get it. Ask people by name. Treat a half-formed concern as a contribution rather than a test of competence. In distributed and remote teams, where the social signals are even thinner, that structure matters more, not less.

Most leaders say they want candour. Far fewer have audited what they actually reward when it shows up. If the people closest to the problem keep choosing silence, the honest reading is not that they lack courage. It is that they have correctly identified what costs them and what does not, and the org chart has been teaching them that lesson on repeat.

So the question is not whether your team has anything to say. The question is what your team has learned about saying it. The quiet ones have already answered. The most expensive ideas in any company are the ones that were fully formed, completely right, and never said out loud — and somewhere in the room, someone is deciding right now whether you have earned the next one.