Nan Zhao, Asaph Azaria, and Joseph A. Paradiso. 2017. Mediated Atmospheres: A Multimodal Mediated Work Environment. Proc. ACM Interact. Mob. Wearable Ubiquitous Technol. 1, 2, Article 31 (June 2017), 23 pages. DOI: https://doi.org/10.1145/3090096

Abstract

Atmosphere - the sensorial qualities of a space, shaped by the composition of light, sound, objects, people, etc. - has remarkable influence on our experiences and behavior. Manipulating it has been shown to be powerful, affecting cognitive performance, mood, and even physiology. Our work envisions and implements a smart office prototype, capable of digitally transforming its atmosphere - creating what we call Mediated Atmospheres (MA) - using computationally controlled lighting, video projection, and sound. Additionally, we equipped this space with a modular real-time data collection infrastructure, integrating a set of biosignal sensors. Through a user study (N=29) we demonstrate MA’s effects on occupants’ ability to focus and to recover from a stressful situation. Our evaluation is based on subjective measurements of perception, as well as objective measurements, extracted from recordings of heart rate variability and facial features. We compare multiple signal processing approaches for quantifying changes in occupant physiological state. Our findings show that MA significantly (p<0.05) affect occupants’ perception as well as physiological response, which encouragingly correlate with occupants’ perception. Our findings are a first step towards personalized control of the ambient atmosphere to support wellbeing and productivity.