Pedestrians, Trucks, and Signs: The Elements Most Stressing Drivers

Driving stress is not solely due to traffic or haste. An innovative study by the Universitat Oberta de Catalunya (UOC) has highlighted the significance of road scenery and visual elements as key factors directly impacting drivers' well-being and health, potentially contributing to a significant number of accidents.

The UOC study examines driver stress using only visual data, without relying on physiological sensors or maneuver records. "We drive in a visual context, and urban scene conditions matter as they affect the driver's stress level," explains UOC expert Cristina Bustos, emphasizing that this aspect had not been centrally considered in research until now.

The open-access paper Analyzing the Visual Road Scene for Driver Stress Estimation presents the findings of the research led by Cristina Bustos, a researcher from the Artificial Intelligence for Human Well-being (AIWELL) group, part of the UOC's Digital Health, Health, and Well-being research unit.

The research also involves faculty from the Studies of Computer Science, Multimedia, and Telecommunications, including Àgata Lapedriza, AIWELL's lead researcher, and Albert Solé, a researcher from Complex Systems (CoSIN3). Javier Borge, CoSIN3's leader, as well as Neska Elhaouij and Rosalind Picard, researchers from the Media Lab at the Massachusetts Institute of Technology (MIT), also participated.

To conduct this large-scale analysis, the research team employed an artificial intelligence (AI) model capable of simultaneously evaluating traffic conditions, pedestrian presence, and urban elements in real driving environments. Various machine learning models were used, from individual image analysis (such as SVM or CNN) to video studies (like TSN), to process this vast amount of visual information.

According to the researcher, this approach allows understanding how the driving environment affects and helps predict driver stress. "We have empirically demonstrated that analyzing the visual environment provides valuable contextual information about the road environment, such as traffic density, urban landscape, or pedestrian presence," details Bustos.

This new perspective, complementing other traditional data sources, is crucial for better understanding the factors triggering driving stress. Moreover, it opens the possibility that urban design and road planning could directly impact reducing driver stress and consequently improving road safety. The UOC study demonstrates, for the first time, that the visual context is a relevant and actionable information source for this purpose.

This study and its conclusions can guide the design of urban infrastructures and policies aimed at reducing stress-inducing factors. For instance, they could inspire improvements in signage, traffic management systems in congested areas, or the design of safer intersections. "By identifying the most stressful elements, urban planners and traffic authorities can take measures to mitigate these effects, contributing to greater road safety," explains Bustos.

There is also the potential to develop driver assistance systems that monitor the environment in real-time and can alert the driver or activate safety mechanisms when potentially stressful conditions are detected.

"For now, there is no immediate plan for practical application, as the study was conducted with a limited number of drivers. However, the results provide a promising foundation for continuing research in this area and exploring its implementation in driver assistance systems," indicates Bustos.

The next steps in the research line involve expanding and diversifying data, exploring multimodal models that integrate other non-invasive data types (such as vehicle information), and refining AI interpretation techniques to better understand the underlying mechanisms in stress detection.

In 2024, there were 1,040 accidents on Spanish roads, along with minor collisions and other driving issues. Behind these accidents are elements such as speeding, adverse weather, and substance abuse, but also distractions and stress situations that can be mitigated by improving infrastructure design, driver assistance technologies, and road safety policies.