The Structure Beneath Psychological Safety: Why a Speaking-Up Culture Needs More Than Culture

• The Most Influential Idea in Team Performance
• What Psychological Safety Actually Is — and Is Not
• The Reframe: Safety Is Downstream of Structure
• The Structural Substrate Was in the Data All Along
• Why Culture-First Initiatives Fade: The Mechanism
• The Nordic Dimension
• The Synthesis: Structure and Culture as Partners
• Conclusion
• References

Psychological safety is real, well-evidenced, and important. It is also largely a consequence of structure — which is why culture-first efforts to create it so often fade

The Most Influential Idea in Team Performance

Few concepts have shaped how organisations think about teams as profoundly as psychological safety. The term was defined by Harvard’s Amy Edmondson in a landmark 1999 study as a shared belief that a team is safe for interpersonal risk-taking. It moved from academic literature to global prominence in 2012, when Google’s Project Aristotle studied 180 teams over two years and identified psychological safety as the single strongest predictor of team performance — more important than team composition, resources, or individual expertise.

The evidence is genuinely strong, and it deserves to be taken seriously. Research by Edmondson and James Detert found that around 85% of employees have, at some point, withheld important information from a manager because they feared the consequences of speaking up. Meta-analyses have since confirmed the link between psychological safety and learning, innovation, and performance across sectors. This is not a soft idea with a hard reputation. It is a rigorously supported one.

And yet a familiar pattern recurs in practice. An organisation reads the research, runs psychological-safety workshops, trains leaders to be more open, and sees a genuine improvement — which then, over months, quietly fades. The initiative was not wrong. But something underneath it did not hold. The Bridgium research with 28 innovation leaders suggests what that something is, and it is fully consistent with Edmondson’s own findings: psychological safety is real, but it is largely downstream of structure.

What Psychological Safety Actually Is — and Is Not

Any serious engagement with this idea has to begin by stating it accurately, because psychological safety is widely misunderstood — a point Edmondson herself has been at pains to make. Psychological safety is not comfort: psychologically safe teams can hold high standards and exchange difficult feedback. It is not consensus: safe teams disagree openly, and the absence of disagreement often signals suppressed dissent rather than harmony. It is not a ‘nice’ culture: members can be blunt and direct. And it is not bilateral trust between individuals.

Edmondson is precise on this last point in particular: psychological safety is a property of the team context — a shared belief about what is safe to do here, with this group. That precision matters enormously, because it points directly at the structural question. If psychological safety is a property of context rather than of individual personalities or leadership warmth, then the decisive question becomes: what shapes the context? And that is a question about structure.

“Innovation without recognition becomes invisible work.”
— CEO · IT Services · Finland

The Reframe: Safety Is Downstream of Structure

Here is the crucial link, and it comes from Edmondson’s own research stream. In their work on Implicit Voice Theories (2011), Detert and Edmondson showed that employees hold tacit beliefs about when speaking up is safe and worthwhile — and that these beliefs are formed through experience, not through exhortation. People learn whether it is safe to speak by observing what happens when they, or others, do.

This is the hinge of the whole argument. If voice beliefs are formed by experience, then what generates the experience? The answer is structure. Whether raising an idea leads to a visible response, whether contribution is recognised, whether there is a predictable path from observation to action — these are structural conditions, and they are what teach people, over time, whether speaking up is worth the risk. Psychological safety is the felt result of that accumulated structural experience.

Berger and Luckmann’s foundational work on the social construction of reality (1966) describes the same mechanism at a deeper level: shared beliefs about ‘how things work here’ are built up through repeated experience until they become taken-for-granted reality. A team’s sense of whether it is safe to speak is exactly this kind of constructed reality — assembled, interaction by interaction, from what the structure has consistently rewarded or ignored.

This is not an argument against the culture-first work — the leadership behaviours Edmondson describes are genuinely valuable, and they matter. It is an argument that those behaviours need a structural foundation to hold. A leader can model curiosity and invite ideas sincerely, but if the ideas then enter a void — no follow-up, no recognition, no visible path forward — the team’s implicit voice theory will, over time, override the leader’s encouragement. The structure teaches a more durable lesson than the leader’s words.

The Structural Substrate Was in the Data All Along

There is a striking detail in Google’s Project Aristotle that is often overlooked. Psychological safety was the top factor — but it was one of five. The others were dependability, meaning, impact, and, tellingly, structure and clarity. The structural substrate was sitting in the research alongside psychological safety from the beginning; the popular retelling simply foregrounded the most human-sounding factor and left the structural ones in the background.

The Bridgium Innovation Flow framework makes the relationship explicit. It identifies three structural conditions that determine whether people contribute — and each maps directly onto what generates the experience that becomes psychological safety.

“The biggest struggle is not the idea itself, but knowing who needs to be convinced.”
— Innovation & Sales Leader · Technology & Enterprise Services · Finland

This quote names the Predictability condition precisely — and shows why psychological safety without structure is incomplete. A person can feel entirely safe to speak, yet still not contribute effectively if there is no visible, reliable path from their idea to a decision. Safety removes the fear of speaking; structure provides the route that makes speaking worthwhile. Both are required, and the second is the one most ‘psychological safety’ programmes leave unaddressed.

Why Culture-First Initiatives Fade: The Mechanism

The fading of psychological-safety initiatives is not a mystery once the structural mechanism is clear. It follows a predictable sequence.

A leader introduces psychological-safety practices sincerely. Team members, encouraged, begin to speak up more — raising ideas, voicing concerns, admitting mistakes. This is the initial gain, and it is real. But each of those acts of speaking up now enters the existing structure. If that structure responds — ideas are considered, contributions recognised, a visible path forward exists — the new behaviour is reinforced and psychological safety deepens. If the structure does not respond — ideas disappear into the void the Bridgium research repeatedly documents — then the accumulated experience begins, quietly, to teach the old lesson again: speaking up costs effort and returns nothing. The implicit voice theory reasserts itself, and the felt safety recedes to match the structural reality.

This is why the most sincere leadership commitment to openness cannot, by itself, sustain psychological safety. The leader is working upstream of behaviour; the structure is working downstream of it, delivering the consequences that actually shape belief. When the two align, safety compounds. When they conflict, structure wins — because structure is what people ultimately experience.

The Nordic Dimension

Nordic workplaces are often assumed to have high psychological safety by default, and there is truth in this. Low power distance, consensus orientation (samförstånd), and high institutional trust (luottamus) create conditions where speaking to a senior colleague is culturally normal. This is a genuine advantage.

But it can also mask a structural gap. High cultural trust can create the appearance of psychological safety — people are polite, relations are good, no one seems afraid — while the structural conditions that turn safety into actual contribution remain absent. A team can feel safe and still not innovate, because feeling safe to speak is not the same as having a predictable path for what happens when you do. The Nordic risk is a comfortable, trusting environment in which the Silence Tax nonetheless operates, because ideas, though freely raisable, still lead nowhere. The cultural advantage is real, but it does not remove the need for structure — it can simply make the missing structure harder to see.

The Synthesis: Structure and Culture as Partners

1. Keep the culture work — and give it a foundation. The leadership behaviours Edmondson describes remain valuable and should continue. The addition is to build, underneath them, the structural conditions — legitimacy, predictability, connectivity — that make the experience match the encouragement. Culture opens the door; structure ensures there is a room behind it.
2. Make speaking up reliably lead somewhere. The single most powerful thing an organisation can do for psychological safety is not another workshop but a visible, reliable path from contribution to response. When people see that raising something consistently leads to consideration and often to action, safety becomes self-sustaining, because the structure keeps confirming it.
3. Recognise contribution structurally. Recognition is part of the legitimacy condition — the signal that contribution matters. Building recognition into how the organisation works, rather than leaving it to individual leaders’ goodwill, makes the safety-generating experience consistent across teams and durable across leadership changes.
4. Diagnose the structure when safety fades. When a psychological-safety initiative loses momentum, the instinct is to redouble the culture work. The more effective response is to examine the structure: what happened to the ideas people raised during the initiative’s successful phase? If they entered a void, that void — not a lack of leadership sincerity — is what needs fixing.

Conclusion

Psychological safety is one of the best-evidenced ideas in modern management, and nothing in the structural view diminishes it. Edmondson’s research is rigorous, Project Aristotle’s finding is robust, and the cost of its absence — the 85% who stay silent — is real. The structural perspective does not compete with this. It completes it.

Psychological safety is a property of context, and context is largely built by structure. People decide whether it is safe to speak from the accumulated evidence of what happens when they do — and that evidence is generated by the structural conditions of legitimacy, predictability, and connectivity. This is why culture-first initiatives produce real but fading gains, and why the durable path to a speaking-up culture runs through structure. Build the conditions that make contribution reliably lead somewhere, and psychological safety follows as their by-product — sustained not by exhortation but by experience.

The question for any leader investing in psychological safety is therefore not only ‘how do we make people feel safe to speak?’ but ‘when they do speak, does our structure reliably prove them right?’ The answer to the second question determines whether the answer to the first one lasts.

The Bridgium Innovation Flow Checklist helps assess the structural conditions beneath a speaking-up culture:
bridgium-research.eu/innovation-checklist-2026/
Full research report:
bridgium-research.eu/innovation-report-2026/

References

1. Edmondson, A.C., “Psychological Safety and Learning Behavior in Work Teams,” Administrative Science Quarterly (1999). JSTOR
2. Edmondson, A.C., The Fearless Organization, Wiley (2018). Publisher
3. Detert, J.R. & Edmondson, A.C., “Implicit Voice Theories: Taken-for-Granted Rules of Self-Censorship at Work,” Academy of Management Journal (2011). JSTOR
4. Google re:Work, “Understand Team Effectiveness” (Project Aristotle). Read
5. Frazier, M.L. et al., “Psychological Safety: A Meta-Analytic Review and Extension,” Personnel Psychology (2017). DOI
6. Berger, P.L. & Luckmann, T., The Social Construction of Reality, Doubleday (1966). Publisher
7. Blau, P.M., Exchange and Power in Social Life, John Wiley & Sons (1964). Publisher
8. Kerr, S., “On the Folly of Rewarding A, While Hoping for B,” Academy of Management Journal (1975). JSTOR
9. Weick, K.E., Sensemaking in Organizations, Sage Publications (1995). Publisher
10. Gallup, State of the Global Workplace: 2024 Report, Gallup Inc. (2024). Read
11. Newman, A. et al., “Psychological Safety: A Systematic Review of the Literature,” Human Resource Management Review (2017). DOI
12. Bridgium, How Innovation Happens: Research Report, Albi Marketing Oy & Digitune Oy (2025). Read