Understanding the influence of stress on sedentary workers' sitting behavior in screen-based interaction context

Sedentary work has become a significant part of a workday context, and this situation becomes more salient due to the COVID-19 pandemic. Existing studies show that body posture can be a good indicator of emotional states. However, to our knowledge, there are no studies that investigated the relationship between the characteristics of whole-body regions while seated and affective information related to stress for sedentary workers in screen-based working scenarios. This paper conducted a preliminary within-subjects study with eight participants performing three types of screen-based tasks at different difficulty levels for simulating natural working conditions. We developed a rapid posture coding technique to analyze sedentary workers' real-time sitting behavior and deployed multiple methods for continuously detecting their stress conditions. The results indicated that stress conditions and task determinants play an essential role in the postural changes of different body regions while seated. Our preliminary findings provide design implications and recommendations for developing a more unobtrusive health-promoting system in the real-life working context.