Rutgers, The State University of New Jersey

03/25/2026 | Press release | Distributed by Public on 03/25/2026 13:54

MRI Scans Reveal How the Brain Processes Toxic Workplace Abuse

Rutgers researchers use advanced neuroimaging to track real-time emotional and moral reactions of employees who witness a boss mistreating a coworker

What happens in the brain when someone witnesses toxic behavior in the workplace?

Researchers at Rutgers Business School, in a first-time collaboration with the Center for Advanced Human Brain Imaging Researchat the Rutgers Brain Health Institute, used an MRI scanner to find out.

While most workplace abuse research focuses on direct victims, this study shifts attention to third-party observers: the coworkers who decide whether to speak up, offer support or remain silent.

"We often think of workplace abuse as something that affects only the target," said Nguyen Pham, a doctoral degree candidate in organization management. "What this collaboration allowed us to do was actually see what happens in the brains of the people watching abuse unfold."

By revealing the neural dynamics underlying third-party responses to abuse, researchers said the study - which was presented at 85th Annual Meeting of the Academy of Management Conference in Copenhagen, Denmark, in 2025 - offers new insight into how ethical workplace cultures can be strengthened from within.

While magnetic resonance imaging, a noninvasive diagnostic procedure, is most often used for clinical or basic cognitive neuroscience, Rutgers researchers used an MRI scanner in a pilot grant program to study brain responses in third-party observers witnessing abusive supervision. The program and technical assistance allowed researchers to adapt it for real-time observation of emotional and moral processes in a workplace context.

The functional MRI gave us a sort of window into the potential emotional and moral processes that shape whether someone might stand up to abusers or stand down.

Nguyen Pham

Doctoral Degree Candidate

The researchers found that empathy was the strongest predictor of constructive workplace responses. Although anger was associated with action, particularly constructive criticism, empathic concern was consistently more powerful in predicting supportive behaviors and intentions to stand up to abusive supervisors.

The project, titled "How Third-Party Employees Respond to Abusive Supervision toward Coworkers," is funded through a pilot grant from the center and the institute for Rutgers-New Brunswick faculty. Danni Wang, an associate professor of management and global business, and Chao C. Chen, Distinguished Professor at RBS, are the principal investigators. and Nguyen Pham, a student of Wang, was integrally involved in executing the study, contributing to the data collection, analysis and implementation of the MRI paradigm.

Participants watched video scenarios depicting verbal abusive supervision across academic, sports and business contexts during MRI scanning, allowing direct measurement of neural activity that complements typical behavioral or survey-based approaches.

Pham and colleagues used functional MRI to examine what happens in the brain when individuals witness abusive supervision, which is defined as the extent to which supervisors engage in the sustained display of hostile verbal and nonverbal behaviors by organizational scholars.

Pham said that functional MRI "gave us a sort of window into the potential emotional and moral processes that shape whether someone might stand up to abusers or stand down."

While most workplace abuse research focuses on direct victims, this study shifts attention to third-party observers: the coworkers who decide whether to speak up, offer support or remain silent.

The findings show that witnessing abusive leadership first activates motivational brain regions, including the insula and cingulate cortex, associated with emotional salience, anger, and alarm.

As observers move from watching the interaction to evaluating it, neural activity shifts toward regions linked to empathy, social cognition and moral reasoning, such as the superior temporal gyrus, fusiform gyrus and anterior cingulate cortex. Rutgers researchers said this pattern suggests that third-party reactions unfold in stages, beginning with emotional arousal and transitioning into reflective moral judgment.

The results suggest empathy as a central emotional driver of ethical bystander behavior and suggest that organizations seeking to reduce toxic leadership may benefit more from cultivating empathic concern than from simply amplifying outrage. By revealing the neural dynamics underlying third-party responses to abuse, researchers said the study offers new insight into how ethical workplace cultures can be strengthened from within.

The study also highlights how advanced brain-imaging resources at Rutgers can support interdisciplinary collaborations beyond conventional neuroscience fields. The Center for Advanced Human Brain Imaging Research houses an MRI scanner dedicated to research and provides technical support for task design, sequence optimization, data analysis, pilot scanning programs and workshops.

"By providing scanner access, technical guidance for integrating behavioral tasks into imaging protocols, analysis support and pilot funding, CAHBIR helped make this project feasible," said David H. Zald, a Henry Rutgers Professor of Psychiatry and the inaugural director of the Center for Advanced Human Brain Imaging Research. This collaboration illustrates how access to advanced neuroimaging can extend beyond core neuroscience to inform questions in organizational behavior and ethics.

"The work reflects BHI's vision: When state-of-the-art brain imaging is accessible across disciplines, unexpected, high-impact breakthroughs emerge, connecting neuroscience to pressing real-world issues like fostering ethical workplaces," said Zald.

Explore more of the ways Rutgers research is shaping the future.

Rutgers, The State University of New Jersey published this content on March 25, 2026, and is solely responsible for the information contained herein. Distributed via Public Technologies (PUBT), unedited and unaltered, on March 25, 2026 at 19:54 UTC. If you believe the information included in the content is inaccurate or outdated and requires editing or removal, please contact us at [email protected]