When Does Negative Feedback Improve Employee Creativity?

We might imagine that employees would improve their performance if they receive negative feedback. With regard to creativity, research has yielded conflicting results; some studies demonstrate that negative feedback improves creativity, while others find that it has no effect on—or even dampens—creativity.

The authors of this study (Kim & Kim, 2020) suggest that an important factor in how employees respond to negative feedback is direction of feedback flow (i.e. supervisors to followers, followers to supervisors, or peers to peers). They examined how people respond in each of these three scenarios. The researchers also posit that people respond to negative feedback in one of two ways: “task processes” or “meta-processes.” Task processes refers to the acts of working to improve based on feedback and meta-processes refers to the psychological state where feedback recipients feel threatened by the feedback.


The researchers conducted two studies, a survey of 225 product development employees in a Korean health food company, and a laboratory experiment of 356 U.S. college students in a simulated work scenario. The researchers found that responses to negative feedback were related to feedback flow. When negative feedback was given by followers to supervisors, supervisors were more likely to improve their creativity. When negative feedback was given by supervisors to followers, or by peers to fellow peers, the feedback recipients were less likely to improve their creativity.

The researchers also found that in the flow from followers to supervisors, supervisors responded with task processes; that is, from their employees’ feedback, they took active steps to improve their creativity. In the scenarios of feedback from supervisors to employees and from peers to other peers, feedback recipients tended to use meta-processes; that is, they had a psychologically negative response that kept them from improving.

The authors explain that the power differential between supervisors and their employees may lead to the difference in feedback response. Supervisors may be more task-focused and because they may be less concerned with social relationships with their staff, they may take criticism as an opportunity to improve. In the flow of feedback from supervisor to follower, feedback recipients may become inhibited from taking action to improve since they’ve received criticism from someone who has power over them. With negative feedback from peer to peer, given competition between peers, feedback recipients may feel like a peer is seeking to undermine their work.


Given the research findings, the authors recommend that organizations institute formal processes for followers to provide critical feedback to their supervisors. Because supervisors may take such feedback less personally and use it to improve their creativity, the result could be more creative solutions organization-wide.

With supervisor to employee or peer to peer feedback, the authors recommend that feedback be avoided while employees are in the middle of a creative task, as it could lead to a decrease in creativity. Another suggestion is for organizations to offer follow-up sessions to their employees after they receive the negative feedback. In these follow-up sessions, employees could work through their negative reactions and gain an understanding of the gap between their creativity and expected standards. This could help them focus on ways to improve their creativity.

Lastly, the authors suggest that peer to peer feedback be given with “temporal feedback” rather than “social comparison feedback.” The former compares past performance with current performance within one employee and the latter compares performance between employees. Temporal feedback can reduce the sense of competition and instead motivate employees to continually improve their own work.


Kim, Y.J., & Kim, J. (2020). Does negative feedback benefit (or harm) recipient creativity? The role of the direction of feedback flow. Academy of Management Journal, 63(2), 584-612.