# Team Exploratory and Exploitative Learning: Psychological Safety, Task Conflict, and Team Performance

**Kostopoulos & Bozionelos (2011) · Upward curve to exploration**

*Kostopoulos & Bozionelos · Primary Research · Team Learning · Group & Organization Management · 2011 · Open access*

Kostopoulos and Bozionelos treat exploratory and exploitative learning as distinct team-level activities, build a five-item measure of each, and test how psychological safety and task conflict relate to them in innovation project teams. The headline finding is a difference in shape: psychological safety related positively to exploitative learning, where only a straight-line term was fitted, and positively but nonlinearly to exploratory learning, in a curve they call 'concave, upward'.

The sample was 142 innovation project teams from 50 ICT and 36 pharmaceutical companies in Greece (65 teams), the UK (42) and Italy (35): 671 members and 142 managers, surveyed after the projects ended. Members rated psychological safety on Edmondson's seven-item scale (alpha .98), and team means were used after agreement checks (mean rwg .86, ICC(1) .36, ICC(2) .92). Managers formally assigned to monitor the teams rated performance on four items, afterwards. Hypotheses were tested with structural equation modelling.

Psychological safety related positively to exploitative learning (beta .21). For exploratory learning both its linear (beta .37) and squared (beta .34) terms were positive and significant; they read this as safety that 'rapidly reinforces' exploratory learning in teams on complex and novel tasks. No turning point is reported. Both learning types related to manager-rated performance (exploratory beta .23, exploitative .21), additively: their interaction was not significant in a separate path model that fitted poorly (CFI .857), which they read as preliminary evidence against the ambidexterity premise. The safety link to exploitative learning was stronger where task conflict was higher (beta .20). For exploratory learning the interaction held only for the linear term, so Hypothesis 3a was not supported, though the discussion later describes both interactions as positive. Task conflict itself correlated negatively with both kinds of learning. Indirect effects of safety on performance were significant and direct effects were not, which they say suggested full mediation; the interaction with task conflict was not mediated.

They name the limits: a cross-sectional design that reduces confidence on causality, innovation teams in knowledge-intensive settings, new learning measures still to be validated elsewhere, and performance ratings with no data showing they were accurate. Whether the findings hold for teams doing routine tasks is left to future research.

We read the curve as running against claims that psychological safety can be too high, for exploratory learning in these innovation teams: no downturn is reported at high safety. Safety also related positively to exploitative learning, a scale that includes routine activities and standardised methods, though these were non-routine teams and the paper leaves routine work untested.

- **Citation:** Kostopoulos, K.C. and Bozionelos, N. (2011) 'Team exploratory and exploitative learning: Psychological safety, task conflict, and team performance', Group & Organization Management, 36(3), pp. 385–415.
- **This page:** https://explore.psychsafety.com/n/kostopoulos-bozionelos-2011/
- **View the source paper:** https://doi.org/10.1177/1059601111405985
- **Interactive map:** https://explore.psychsafety.com/?mode=papers&node=kostopoulos-bozionelos-2011

## Neighbours (19)

- [Psychological Safety and Learning Behavior in Work Teams](https://explore.psychsafety.com/n/edmondson-1999.md) (paper)
- [Exploration and Exploitation in Organizational Learning](https://explore.psychsafety.com/n/march-1991.md) (paper)
- [Psychological Safety: A Meta-Analytic Review and Extension](https://explore.psychsafety.com/n/frazier-et-al-2017.md) (paper)
- [A Strategic View of Team Learning in Organizations](https://explore.psychsafety.com/n/harvey-bresman-edmondson-pisano-2022.md) (paper)
- [High-Quality Relationships, Psychological Safety, and Learning from Failures in Work Organizations](https://explore.psychsafety.com/n/carmeli-gittell-2009.md) (paper)
- [Psychological Safety: A Systematic Review of the Literature](https://explore.psychsafety.com/n/newman-donohue-eva-2017.md) (paper)
- [Psychological Safety: The History, Renaissance, and Future of an Interpersonal Construct](https://explore.psychsafety.com/n/edmondson-lei-2014.md) (paper)
- [Speaking Up in the Operating Room: How Team Leaders Promote Learning in Interdisciplinary Action Teams](https://explore.psychsafety.com/n/edmondson-2003b.md) (paper)
- [A Temporally Based Framework and Taxonomy of Team Processes](https://explore.psychsafety.com/n/marks-mathieu-zaccaro-2001.md) (paper)
- [When Feeling Safe Isn't Enough: Contextualizing Models of Safety and Learning in Teams](https://explore.psychsafety.com/n/sanner-bunderson-2015.md) (paper)
- [Conflict & Holding Environments](https://explore.psychsafety.com/n/psychological-safety-60-conflict-and-holding-environments.md)
- [High Performing Teams](https://explore.psychsafety.com/n/high-performing-teams.md)
- [How Psychological Safety Affects Team Performance: Mediating Role of Efficacy and Learning Behavior](https://explore.psychsafety.com/n/kim-lee-connerton-2020.md) (paper)
- [The Limits of Psychological Safety: Nonlinear Relationships with Performance](https://explore.psychsafety.com/n/eldor-hodor-cappelli-2023.md) (paper)
- [Too Much PS?](https://explore.psychsafety.com/n/can-workplaces-have-too-much-psychological-safety.md)
- [Answers to 20 Questions About Interrater Reliability and Interrater Agreement](https://explore.psychsafety.com/n/lebreton-senter-2008.md) (paper)
- [Nonlinear PS & Performance](https://explore.psychsafety.com/n/a-critique-of-the-limits-of-psychological-safety-nonlinear-relationships-with-performance.md)
- [Two Lines: A Valid Alternative to the Invalid Testing of U-Shaped Relationships With Quadratic Regressions](https://explore.psychsafety.com/n/simonsohn-2018.md) (paper)
- [Kernel Regularized Least Squares: Reducing Misspecification Bias with a Flexible and Interpretable Machine Learning Approach](https://explore.psychsafety.com/n/hainmueller-hazlett-2014.md) (paper)
