Psychological safety concerns the interpersonal consequences people expect when they ask a question, acknowledge an error, disagree, or seek help. Someone may notice a problem and withhold it because speaking feels socially risky. The organization's task includes what happens after that person speaks. Information creates value when the team can examine and use it.
What is psychological safety?
Amy Edmondson's foundational research describes team psychological safety as a shared belief that interpersonal risk-taking is safe within the team (1999). The construct is situated at team level. Members learn what they can say, to whom, and with what consequences through repeated interactions. A formal invitation to contribute is one signal among many.
The risk is social. Asking for help may make someone fear appearing incompetent. Reporting an issue may feel like being negative, while challenging a decision may look disloyal. The person may have the technical skill to identify the problem and still decide that silence is safer. Those anticipated reactions become part of the conditions influencing behavior.
A psychologically safe team still has disagreements, responsibilities, and quality standards. Its members can examine doubts and errors without treating every contribution as a threat to someone's standing. Edmondson and Lei (2014) describe the development of this interpersonal construct across the literature. The practical question concerns how people participate in learning and work when the information is uncomfortable.
Why does silence have an economic cost?
Employees closest to operations often see inconsistencies and emerging risks before senior decision-makers do. Silence delays access to that information. A correctable problem becomes an incident, an unsupported assumption survives a review, or a team continues implementing a decision whose limitations are already known. The cost arises through what the organization discovers too late.
Psychological safety reduces the anticipated social cost of actions such as:
- Reporting an error or emerging risk.
- Asking for clarification before acting on an ambiguous instruction.
- Seeking help when a skill or resource is missing.
- Presenting a minority view.
- Sharing findings from an unsuccessful experiment.
These actions expand the information available for a decision. Their economic value depends on the problem, the response, and what the organization changes. The argument supports investigating the cost of silence in a specific setting; it does not assign a universal financial return to a psychological safety program. A manager can start with decisions or incidents for which earlier information would have mattered.
Why does psychological safety extend beyond accepting mistakes?
An organization may tolerate some errors while discouraging questions, objections, or requests for help. Accepting mistakes covers only part of interpersonal risk. A team also needs to examine how a person is treated for challenging a senior colleague or identifying an impractical deadline. Those situations can carry a greater social cost than admitting a minor error.
The type of failure matters as well. Edmondson (2023) distinguishes preventable failures in familiar operations, failures arising from complexity, and intelligent failures from well-designed exploration in uncertain territory. Bringing an event into view makes learning possible. Accountability still includes preparation, the quality of the decision process, and the response to avoidable departures from established practice.
Disagreement also involves responsibilities. A useful objection supplies information, explains an anticipated consequence, or identifies an assumption for examination. The decision-maker still needs to reach a decision. Psychological safety protects the opportunity to contribute without giving every objection equal evidential weight or a veto over the outcome.
Which workplace conditions matter?
Plouffe and colleagues developed the Psychological Safety Inventory, or PSI, with a broader conception of perceived workplace safety than Edmondson's original team construct (2023). Their five-factor instrument covers interpersonal risk-taking, mutual trust and respect, organizational and structural support, identity and clarity in the context of the team, and supportive leadership. These dimensions describe the instrument's approach rather than five universally agreed components of every psychological safety model.
Trust and respect concern whether people expect their contributions to receive serious consideration. Structural support includes access to resources and information, alongside protection and support when difficulties arise. Confidence in one's place and future within the team contributes another part of that perceived environment. Inclusion and protection from discrimination belong to those organizational conditions.
Employees also observe leaders' reactions. A response to bad news becomes information for the whole group. A manager who investigates a warning and explains the next decision creates a different expectation from one who attacks the messenger. Repeated responses help establish which contributions appear safe.
| PSI dimension | What it covers | Workplace expression |
|---|---|---|
| Interpersonal risk-taking | Perceived ability to speak without disproportionate interpersonal consequences. | Raise a concern, ask for help, disagree, or advocate for a colleague. |
| Mutual trust and respect | Constructive expectations and recognition within the team. | Share an incomplete observation and expect serious consideration. |
| Organizational and structural support | Access to resources, information, protection, and support. | Find help and obtain resources needed to address a problem. |
| Identity and clarity in the context of the team | Perceived importance, security, and stability of one's place in the team. | Feel recognized as a member and sufficiently secure to contribute. |
| Supportive leadership | Perceptions of effective guidance, support, and constructive handling of issues. | Receive a useful response to a concern or request for help. |
Why does psychological safety matter during transformation?
Transformation increases the need to learn while making participation feel riskier. Roles change, some skills become less useful, and decisions may remain provisional. Teams need to try unfamiliar practices while maintaining operations. Leaders depend on incomplete information from the people doing that work.
A user needs to be able to explain that a tool does not fit the workflow. A manager needs to challenge a schedule that overlooks available capacity. A team needs to admit uncertainty about what a new practice requires. When these messages disappear, decision-makers can confuse the absence of warnings with the absence of difficulty.
Psychological safety supports the feedback through which change enablement adjusts implementation. It also matters to adaptive leadership: a leader cannot readily adjust to conditions that others systematically filter out. Information about failed assumptions gives the organization an opportunity to change the design, timing, or conditions of the transformation.
How do you develop psychological safety in practice?
“Speak freely” carries little weight when experience suggests that certain messages will be punished. Teams assess observed consequences alongside stated intentions. An intervention therefore needs specific situations in which useful information is at risk of being withheld. A steering meeting, incident review, or conversation with a supervisor provides a setting for investigation.
The diagnosis examines what people do, which reactions they anticipate, and what rules or management responses support that expectation. Candidate changes follow those findings. A leader might solicit objections before announcing a preference, or a project review might examine invalidated assumptions before discussing individual responsibility. The design needs someone responsible for explaining how concerns will be handled.
Early responses matter because they give participants evidence about what to expect. A defensive reaction can undermine several invitations to be open. Tracking relevant practices helps assess whether conditions are changing: requests for help, warnings raised, objections examined, and issues addressed. A count of comments in a meeting is less informative than evidence about which difficult information reached the decision.
How can psychological safety be measured?
Edmondson's original seven-item scale measures a team climate concerning interpersonal risk-taking. Plouffe and colleagues' 30-item PSI measures individual perceptions across a broader five-factor conception of the work environment. The instruments answer related but different questions. Selection needs to follow the construct and level of conclusion the organization intends to examine.
The PSI validation study reported high internal consistency, including an alpha of .95 for the total score. The authors also noted possible item redundancy at that level. Internal consistency concerns how items relate to one another; it does not establish that a particular program improves performance. The validation involved self-reports from workers in Canada, the United Kingdom, and the United States, with predictive relationships to performance and innovation left for further research.
Repeated questionnaires can track reported perceptions, while interpretation of an intervention's effect depends on the evaluation design. UNREST complements suitable measures with evidence of work: what information was raised, how it was handled, and whether it changed a decision or response. This gives leaders evidence about the behaviors and consequences behind the scores. An average score also needs a justified level of aggregation before it is presented as a team finding.
Psychological safety becomes visible in decisions
People may feel comfortable discussing minor issues while keeping quiet about resources, schedules, or the chief executive's preferred option. A favorable survey result needs to be interpreted alongside those consequential situations. The demanding test comes when someone supplies information that challenges a decision already underway. The organization can investigate and revise its action, or acknowledge the speaker while leaving the issue untreated.
The practical value lies in making contributions usable. UNREST's psychological safety program addresses the behaviors that make warnings possible and the organizational conditions for responding. A campaign that celebrates speaking up while leaders disregard the information leaves the cost of silence largely intact. Progress includes evidence that the organization handles concerns and learns from them.
Frequently asked questions
How does psychological safety differ from mental health?
Psychological safety concerns anticipated interpersonal consequences in a work setting. Mental health includes a much wider range of experiences and conditions. The concepts may be related, but a psychological safety questionnaire should not be interpreted as a clinical assessment of employees.
Does psychological safety lower standards?
Clear expectations and psychological safety can operate together. Earlier discussion of problems helps teams investigate quality gaps and responsibilities. The way a team responds to questions and errors influences whether those gaps become visible in time to address them.
Does it prevent conflict?
It helps people express and examine work-related disagreements. Some conflicts may become more visible as a result. The relevant outcome is whether the team uses disagreement to improve understanding and decisions while dealing with personal attacks or exclusion.
What are warning signs of low psychological safety?
Consistently unanimous meetings, late surprises, few requests for help, and important discussions occurring outside decision forums warrant investigation. Any one of these signals can have another explanation. Observation and confidential accounts help establish what people anticipate when they consider speaking.
Who is responsible for it?
Leaders strongly influence expectations through their reactions and decisions. Colleagues, internal rules, resources, and recognition systems also contribute. Development therefore involves management behavior and organizational arrangements, with responsibility for acting on the information employees provide.
About the author
Maxime Rabéchault is the founder and president of UNREST, a change enablement and organizational development consultancy grounded in applied behavioral science. Trained at Université Paris Dauphine-PSL, the Conservatoire national des arts et métiers (CNAM), University College London (UCL), and Sciences Po Paris, he also teaches as an adjunct at IAE Paris-Sorbonne. His work connects observable professional practices with the organizational conditions that make transformation possible.
References
Edmondson, A. C. (1999). Psychological safety and learning behavior in work teams. Administrative Science Quarterly, 44(2), 350–383. https://doi.org/10.2307/2666999
Edmondson, A. C., & Lei, Z. (2014). Psychological safety: The history, renaissance, and future of an interpersonal construct. Annual Review of Organizational Psychology and Organizational Behavior, 1, 23–43. https://doi.org/10.1146/annurev-orgpsych-031413-091305
Edmondson, A. C. (2023). Right Kind of Wrong: The Science of Failing Well. Simon Element. https://www.simonandschuster.biz/books/Right-Kind-of-Wrong/Amy-C-Edmondson/9781982195069
Plouffe, R. A., Ein, N., Liu, J. J. W., St. Cyr, K., Baker, C., Nazarov, A., & Richardson, J. D. (2023). Feeling safe at work: Development and validation of the Psychological Safety Inventory. International Journal of Selection and Assessment, 31(3), 443–455. https://doi.org/10.1111/ijsa.12434
_11zon.webp)
