Workplace stress is a pervasive issue, with one in three employees in Switzerland experiencing its detrimental effects. However, often employees are unaware of the toll it takes on their physical and mental health until it's too late. That's why researchers at ETH Zurich are working on a groundbreaking solution: using data and machine learning to detect workplace stress in real-time, just by analyzing how people type and click.

Led by study author Mara Nägelin, a mathematician who conducts research at the Chair of Technology Marketing and the Mobiliar Lab for Analytics at ETH Zurich, the team discovered that typing and mouse behavior can be better predictors of workplace stress than heart rate. Through carefully designed experiments that simulated office tasks as closely as possible, the researchers observed 90 study participants as they performed various work-related activities, such as scheduling appointments and analyzing data, while monitoring their mouse and keyboard behavior, as well as their heart rates.

The findings were astonishing. Stressed individuals exhibited different typing and mouse behavior compared to their relaxed counterparts. They moved the mouse pointer more frequently and less precisely, covering longer distances on the screen. Their typing was erratic, with frequent pauses and mistakes. In contrast, relaxed individuals took shorter and more direct routes with the mouse and had fewer but longer pauses while typing.

READ MORE | Heavy workloads keep employees from taking breaks

These observations align with the neuromotor noise theory, which suggests that increased stress levels impair the brain's ability to process information and affect motor skills. According to psychologist Jasmine Kerr, who co-authored the study with Nägelin, "Increased levels of stress negatively impact our brain's ability to process information. This also affects our motor skills."

The potential applications of this research are immense. By detecting workplace stress early on through typing and mouse behavior, companies could intervene and provide support to employees before it escalates. However, the researchers are cautious about the ethical implications of workplace stress detection. They emphasize the need to protect employees' data and ensure responsible handling of sensitive information.

READ MORE | Bad managers raise the risk of office bullying

"We want to help workers identify stress early, not create a monitoring tool for companies," says Kerr. The team is currently testing their stress model with data collected from Swiss employees who have voluntarily agreed to have their mouse and keyboard behavior, as well as their heart rate, recorded directly at their workplace using an app. The same app also regularly asks employees about their subjective stress levels. The results of this ongoing study are expected to be available by the end of the year.

As workplace stress continues to be a pressing concern for employees and employers alike, the potential of using data and machine learning to detect stress early on offers hope for creating healthier and more supportive work environments. With responsible data handling and a focus on employee well-being, this innovative approach could revolutionize how workplace stress is managed and prevented, ultimately benefiting the health and productivity of employees in the long run.