Psychological Safety Is Not the Same as Trust

The Theoretical Distinction

Edmondson (1999) explicitly distinguished psychological safety from trust in the introduction of the construct to the organizational literature. She noted that trust is typically conceptualized as a dyadic phenomenon, characterizing the specific belief of one person about another specific person's reliability, intentions, and competence. Psychological safety is a team-level climate variable characterizing the shared perception of the team as a whole that interpersonal risk-taking is safe within the team context. The distinction has structural implications: you can trust a specific colleague without feeling safe speaking up in your team if other members or the leader create conditions that penalize communication. You can feel safe in a team climate without trusting any specific team member with confidential information about yourself, because the safety operates at the team norm level rather than at the individual relationship level.

Mayer, Davis, and Schoorman (1995) articulated trust as composed of three distinct dimensions: ability, the belief that the other party has the competence to do what they are trusted to do; benevolence, the belief that the other party cares about one's outcomes and will act in one's interest even when not monitored; and integrity, the belief that the other party adheres to an acceptable set of principles that includes following through on commitments. The research has consistently found that these three dimensions predict different behaviors and are differently sensitive to violation: integrity-based trust is hardest to build and hardest to repair once violated, while ability-based trust can be established more quickly and recovered more readily through demonstrated capability improvement. None of these dimensions maps directly onto the psychological safety construct, which describes team-level climate rather than beliefs about specific others' attributes.

The practical organizational significance of the theoretical distinction is most visible in failure cases. An organization that diagnoses a speaking-up problem, where employees are not raising concerns, sharing ideas, or reporting problems, as a trust problem and responds by building interpersonal trust between individuals may improve dyadic relationships without addressing the team-level norms that are actually suppressing communication. The improved trust means employees are more comfortable disclosing to their manager in private but remain unwilling to raise concerns in team settings where the relevant climate norms have not changed. An organization that diagnoses the same problem as a psychological safety problem and invests in the specific leader behaviors that establish team-level safety norms is addressing the actual causal mechanism through which collective communication is being suppressed.

What Each Construct Explains Independently

Edmondson (1999) found that psychological safety predicted team learning behavior across a sample of work teams, independently of team composition, organizational context, and established predictors of team performance. Subsequent research has consistently replicated this finding across diverse organizational settings, decision-making contexts, and cultural backgrounds, establishing psychological safety as one of the most robust predictors of team learning and one of the most consequential for teams whose performance depends on the collective sharing of partial information, the raising of concerns about potential errors, and the expression of perspectives that challenge current approaches. The Google Project Aristotle research, while not peer-reviewed, corroborated the academic finding by identifying psychological safety as the single strongest predictor of team effectiveness across a large and diverse corporate population.

Trust, as Colquitt, Scott, and LePine (2007) demonstrated in their meta-analysis, predicts a different but overlapping set of outcomes: task performance, organizational citizenship behavior, and counterproductive work behavior, above and beyond established predictors including personality, justice perceptions, and job characteristics. The predictive pattern of trust outcomes is more strongly weighted toward individual-level compliance, commitment, and risk-tolerance in dyadic transactions, while the predictive pattern of psychological safety outcomes is more strongly weighted toward collective-level communication, learning, and innovation. Both constructs independently predict important organizational outcomes, and the combination of high levels of both produces organizational performance that exceeds what either alone can explain.

The interaction between the two constructs produces the most organizationally important effects. High psychological safety without high interpersonal trust produces team environments where people are willing to speak up but are not committed to following through on the coordination that speaking up is intended to produce. The team generates ideas and raises concerns, but the low trust means that commitment to collective decisions is weak and follow-through on shared agreements is unreliable. High trust without high psychological safety produces strong dyadic relationships and high individual commitment, but collective communication is constrained by the team-level norms that make raising concerns, acknowledging uncertainty, and challenging established approaches socially costly, regardless of the trust any specific pair of individuals shares.

Diagnostic Implications for Organizational Assessment

The diagnostic implication of the theoretical distinction is that assessment of organizational communication quality requires measuring both constructs independently rather than treating either as a proxy for the other. Survey instruments that combine items from both constructs into a single scale produce a conflated score that cannot guide targeted intervention because it cannot identify whether the primary constraint on collective communication quality is the team-level safety climate, the dyadic trust quality, or both. Diagnostically valid assessment requires separate measurement of team-level psychological safety, capturing the shared perception of interpersonal risk at the team level, and dyadic trust, capturing specific beliefs about specific relationship partners' ability, benevolence, and integrity.

The organizational conditions most predictive of psychological safety are primarily determined by team leader behavior: the leader's response when team members raise concerns or make mistakes, the leader's modeling of intellectual humility and willingness to be wrong in front of the team, and the leader's explicit and consistent invitation of input from all team members regardless of status. The organizational conditions most predictive of trust are determined by a wider range of factors including leader follow-through consistency, organizational fairness perceptions, and the accumulated evidence about the reliability of specific individuals. This means that interventions for the two constructs are partially distinct: safety interventions focus primarily on specific leader facilitation behaviors, while trust interventions focus more broadly on consistency and integrity across a range of organizational behaviors.

The assessment instrument design implications are specific. Items that probe psychological safety should describe team-level norms rather than specific relationship qualities: questions about what happens when mistakes are raised in team meetings, whether it is safe to disagree with the leader in front of the team, and whether team members are penalized for raising concerns. Items that probe trust should be anchored to specific relationship partners and specific dimensions of trustworthiness: questions about whether a specific leader or colleague reliably follows through on commitments, acts in one's interest when unobserved, and adheres to stated principles consistently. Instruments that conflate both constructs into a generic interpersonal climate assessment cannot produce the diagnostic specificity needed to guide targeted intervention.

Intervention Implications

Organizations that discover low psychological safety in their team diagnostics should invest primarily in the specific leader behaviors that establish team-level safety norms: ensuring leaders consistently respond to team member communication attempts with acknowledgment and genuine engagement rather than dismissal; ensuring leaders model intellectual humility by acknowledging their own uncertainty and mistakes in team settings; and ensuring leaders actively attribute contributions to the specific team members who made them rather than accepting undifferentiated credit for collective outputs. These behaviors are specific, observable, and developable through targeted behavioral coaching and structured feedback. Generic culture initiatives that address psychological safety as a cultural value rather than as a set of specific leader behaviors produce less consistent and less durable improvements.

Organizations that discover low trust in specific dyadic relationships or across their leadership population should invest in the behavioral consistency and integrity-related practices that trust research most consistently identifies as trust-building: ensuring leaders follow through on explicit and implicit commitments; ensuring leaders explain the reasoning behind consequential decisions including those that some parties will experience as unfavorable; and ensuring that leaders apply stated principles consistently across organizational members rather than selectively. These interventions are less amenable to behavioral training alone and more dependent on organizational design choices, including accountability structures that make follow-through visible and performance management systems that assess leader consistency.

The most organizationally productive intervention strategy is one that assesses both constructs, identifies the primary constraint on collective communication quality in specific teams and units, and concentrates investment in the interventions most targeted at that constraint. Organizations that invest in both safety and trust simultaneously, recognizing that the interaction between the two constructs determines collective performance at a level neither alone predicts, produce stronger and more durable improvements in team communication quality, learning behavior, and performance than those that address either construct in isolation. The practical organizational goal is not high trust or high safety but the combination of both that produces the collective communication environment where the most important organizational information is raised, heard, and acted upon consistently.