The definition of psychological safety

The definition of psychological safety is the shared belief within a team that it is safe to take interpersonal risks — to ask a question, admit a mistake, offer a half-formed idea, or challenge the way things are done — without fear of embarrassment, rejection, or punishment. LeaderFactor states the same construct in five words: psychological safety is a culture of rewarded vulnerability. It describes an environment, not a personality trait, and it exists in degrees rather than as an on/off switch.

That precision matters. “Psychological safety” is now used loosely enough to mean almost anything pleasant about a workplace, and a definition that means everything measures nothing. This page defines the term exactly: what it is, where it came from, what it is made of, and where its boundaries sit against trust, comfort, and niceness.

What is the definition of psychological safety?

Psychological safety is the belief, held in common by the members of a team, that the team is safe for interpersonal risk-taking. In practice, that means people expect to be met with respect rather than ridicule when they say something exposing — “I don’t know,” “I made a mistake,” “I disagree,” “I need help.”

Three properties are built into that definition, and every one of them is doing work:

• It is a group-level property, not an individual one. Psychological safety is not confidence, resilience, or a personality type. It is a shared perception of an environment, which is why it is measured at the level of the intact team — the smallest divisible unit of culture.
• It is specifically about interpersonal risk. Not business risk, not physical risk, not risk to the schedule. The exposure in question is social: looking incompetent, ignorant, intrusive, or negative in front of other people.
• It is not binary. From a family to a boardroom to a Navy SEAL platoon, every social unit registers some level of psychological safety. It can be high, low, or somewhere in between, and it is context-sensitive — a person can feel safe in one meeting and unsafe in the next.

Definitions that omit any of the three tend to collapse into vagueness. “A place where people feel comfortable” drops the interpersonal-risk clause. “Being able to be yourself” drops the group level. “We have psychological safety” drops the matter-of-degree.

Where does the term “psychological safety” come from?

The idea is as old as human interaction, but the term itself has a traceable lineage — and knowing it prevents the most common attribution errors.

• 1954 — Carl Rogers coins the term. The psychologist used “psychological safety” to describe conditions that free a person to create, in the context of the therapeutic relationship.
• 1960s — Edgar Schein and Warren Bennis bring it into organizations. They defined it as the climate that reduces interpersonal risk, framing it as a precondition for change and learning in groups rather than a private state.
• 1990 — William Kahn ties it to engagement. Kahn argued that people bring their full selves to work only when they feel safe to express themselves without fear of negative consequences to self-image, status, or career.
• 1999 — Amy Edmondson makes it empirical. Edmondson’s study of work teams established psychological safety as a shared, team-level belief that the team is safe for interpersonal risk-taking, and demonstrated its link to learning behavior and performance. This is the construct most academic literature still uses.
• 2014 — Google’s Project Aristotle popularizes it. After studying 180 of its own teams, Google identified psychological safety as the single most important factor explaining high performance — the moment the term moved from academic literature into everyday management vocabulary.
• 2020 — Dr. Timothy R. Clark introduces The 4 Stages. Clark, founder and CEO of LeaderFactor, published The 4 Stages of Psychological Safety and made two moves the earlier work had not: he defined psychological safety as a culture of rewarded vulnerability, and he argued it is not one general condition but a sequence of four.

One attribution note worth getting right: Amy Edmondson did not author The 4 Stages. Edmondson’s contribution is the validated team-level construct and its link to learning; The 4 Stages of Psychological Safety™ is Dr. Timothy R. Clark’s model. The two are compatible — the stages operationalize the construct — but they are different bodies of work by different authors.

What does “a culture of rewarded vulnerability” mean?

LeaderFactor’s five-word definition is not a slogan; it names a mechanism. Unpack it in two halves.

Vulnerability is any act that exposes you socially. LeaderFactor’s research team catalogued more than 150 such acts at work — asking a question, pointing out a mistake, saying no, apologizing, delivering bad news, speaking with an accent — then narrowed to the 20 most common and surveyed 2,025 individuals across 834 organizations on how much personal risk each one carries. The average score was 4.1 on an 11-point scale, but the spread was wide: giving an incorrect answer rated highest at 5.6, while connecting with your team rated lowest at 2.7. Vulnerability is universal; how vulnerable a given act feels is personal.

Rewarded describes what happens next. Whenever people enter a social setting, they run instinctive threat detection, asking one question: am I safe here? The answer governs behavior. If the environment consistently rewards acts of vulnerability, people offer a performance response — they engage, contribute, and release discretionary effort. If it consistently punishes vulnerability, they offer a fear response — they self-censor, withdraw, and manage personal risk. The objective shifts from adding value to surviving, which is not a small difference.

That gives the definition a mechanism you can act on. Culture forms through exactly two behaviors: model vulnerability yourself, and reward the vulnerability of others. There is no shortcut around it.

What is psychological safety made of?

Operationalized, psychological safety lives at the intersection of two variables: respect and permission.

• Respect is the general level of regard and esteem people extend to one another — valuing a person as a human being.
• Permission is the latitude granted to others to be themselves, participate, influence the group, and act.

Both have to be present. Respect without permission produces paternalism: the well-meaning leader who pats people on the head and tells them not to touch anything. Permission without respect produces exploitation: latitude granted in order to extract value from people who are not themselves valued. Both are common, and both read as unsafe to the people living in them.

Is psychological safety one condition or four?

Definitionally, psychological safety is one construct expressed in four stages. Earlier definitions treated it as a single general condition; Clark’s contribution was to show that the vulnerability people are willing to risk changes as their needs change, so the condition is stage-specific:

1. Inclusion safety — safe to be accepted and belong.
2. Learner safety — safe to ask, experiment, and make mistakes.
3. Contributor safety — safe to do meaningful work with real autonomy.
4. Challenger safety — safe to challenge the status quo and propose change.

The sequence is empirically supported, not merely intuitive. In a PSindex® study of 3,366 employees across more than twenty organizations worldwide, respondents were asked in what order they would engage a new team, with the response options randomized to eliminate display-order bias. Sixty-five percent named inclusion first, 63% named learning second, 81% named contribution third, and 86% named challenging the status quo last — a progression confirmed by a Friedman test (χ²(3) = 7,357.81, p < .001) with strong agreement across respondents (Kendall’s W = .73).

This is why the definition is measurable rather than merely descriptive. PSindex® operationalizes it as 12 items — three per stage, each rated 0–10 — and scores a team from −100 to +100 by subtracting the percentage of responses reporting consistently punished vulnerability from the percentage reporting consistently rewarded vulnerability. A definition that produces a number is a definition you can manage against. The full model lives on our 4 Stages of Psychological Safety skill page.

What is psychological safety not?

The definition is easiest to hold onto next to its most common counterfeits. Psychological safety is not:

• A shield from accountability. It is not diplomatic immunity; standards go up, not down.
• Coddling. Overprotecting people is not the same as making it safe for them to be exposed.
• Consensus decision-making. It does not change who decides.
• Niceness. Superficial collegiality paints a thin layer of nice over a thick layer of fear.
• Political correctness. The construct is apolitical and subscribes to no ideology.
• Unearned autonomy. Autonomy is still earned through demonstrated competence.
• Verbal reassurance. You cannot announce psychological safety into existence; leaders are responsible for conditions, not declarations.

Each of these is worth understanding in detail — see what psychological safety is not for the myth-by-myth breakdown.

How is psychological safety different from trust, comfort, and niceness?

These three near-synonyms account for most of the confusion around the term.

Trust is a belief one person holds about another, usually oriented toward the future: will you do what you said? Psychological safety is a belief about the group, oriented toward the present: is it safe to be vulnerable here, now, with these people? Trust is typically dyadic and deferred; safety is collective and immediate. You can trust a colleague’s competence and still not feel safe raising a concern in front of the team. For the full comparison, see psychological safety vs. trust.

Comfort is the absence of tension. Psychological safety is the presence of enough regard that tension becomes survivable — which is why safe teams argue more, not less. Challenger safety in particular introduces conflict, confrontation, and creative abrasion by design.

Niceness is a behavior; psychological safety is a condition. Niceness withholds hard truths to preserve the mood. Psychological safety makes hard truths sayable. The two often move in opposite directions.

Why the definition matters

A good definition of psychological safety has to pass three tests: it must describe a group, name interpersonal risk specifically, and admit of degree. “A culture of rewarded vulnerability” passes all three — and adds a mechanism, because it tells you exactly what raises and lowers the number.

That is the practical payoff of getting the definition right. If psychological safety means “people are comfortable,” there is nothing to do on Monday morning. If it means “acts of vulnerability are consistently rewarded rather than punished,” the work is obvious: notice the vulnerable acts around you, and reward them. Start with the psychological safety glossary entry for the short form, or the workplace overview for how the definition plays out day to day.