Abstract
Environmental Protection Agency (EPA) reports that Americans, on average, spend around 90% of their lives in indoor environments. While Indoor Environmental Quality (IEQ) significantly impacts the occupants’ physical and mental well-being, buildings are not designed to collaborate with occupants, enhance their well-being, or communicate how occupants can reduce their environmental impact (i.e., energy consumption). We believe the recent advancements in affective computing, human-computer interaction, and behavioral theories derived from psychology and sociology can be incorporated to design and operate smart buildings that sense and react to occupants' emotional and behavioral changes. Within clinical psychology, passive sensing techniques have been utilized in combination with intervention strategies to improve patient states for depression and anxiety. While the benefits of digital tracking technologies are widely replicated regarding physical health and behavior, the effects of emotional tracking have not been studied longitudinally in the context of smart buildings. Any benefits from emotion tracking can be incorporated into a dynamic tracking and feedback system to improve its users' general well-being.
Our findings from the literature review reveal that multiple theories include emotions as a major construct affecting occupant behavior and comfort in buildings, which may influence how occupants interact with building systems, impacting buildings' total energy consumption. However, research on longitudinal sensing of occupant emotions and behavior has been far limited to link the emotions and behavior to the different constructs identified in behavioral science theories. Longitudinal emotion and behavior sensing in smart buildings create an opportunity to react to user-specific needs and emotions and increase occupants' emotional awareness by providing feedback on their emotional states' trends.
To address these gaps, in this thesis (1) a literature review is conducted to identify how behavioral theories can explain the role of emotional states and occupant behavior in interactions with building systems. Based on the identified gaps in the literature, (2) a sensing framework is introduced to longitudinally sense occupants’ emotional states subjectively, through daily surveys, and objectively through cameras deployed in the Living Link Lab testbed. Lastly, (3) a set of longitudinal studies are conducted to evaluate the impacts of emotional tracking and feedback on individuals.
Through the longitudinal studies, several hypotheses related to the potential effects of digital emotions tracking on individuals' emotional trends were evaluated. The results of the statistical analyses indicate that tracking daily emotions increases the stickiness of positive emotions. We hypothesize that this effect is due to the difference in recall of positive and negative emotions. Research suggests that memories associated with negative emotions are easier to recall than memories related to negative emotions. This gap in memory recall means tracking of emotions can improve recall of positive emotions. Thinking about and remembering positive experiences, in turn, has shown to benefit an individual’s mental health significantly. The finding that tracking emotions can improve recall and stickiness of positive emotions has implications for dynamic feedback and intervention mechanisms to enhance well-being.
We incorporate a sensing framework to longitudinally sense occupants’ emotional states subjectively, through daily surveys, and objectively through cameras deployed in the Living Link Lab testbed. This platform also incorporates an online dashboard with visualized data and controls for smart devices in buildings. The dashboard will use subjective reports and video data to provide digital tracking and feedback on user behavior and emotions. Furthermore, longitudinal data on emotions and objective measurements of the sensors and cameras can link the identified constructs in behavioral theories to occupants’ behavior and emotions. Finally, the subjective and objective measurements can be combined to verify the external and internal validity of the impact of emotions tracking and intervention strategies on occupant emotion and behavior.
