1. U N I V E R S I T Y O F S O U T H F L O R I D A Using Interactive Visualizations to Improve Self-Awareness and Management of Everyday Life Stress Tylar Murray, Dr. Andrew Raij Department of Electrical Engineering We propose the combination of smartphone-based sensor systems and interactive data visualizations to 1) track the daily experience of stress and 2) provide personalized feedback on stress. In combining these technologies, we aim to create a closed feedback loop consisting of sensors, a smartphone, and the user. First, the sensors assess the state of the user. Then the smartphone generates personalized feedback based on the sensed state. Next, the user reviews the feedback and takes action to reduce stress, and finally the loop restarts. Introduction Research has shown that everyday stress is one of the most significant factors affecting an individual's health. Theories and approaches to managing stress have become a hot topic of study and are known to affect both mental and physical well-being. Through recent advances in mobile sensing, the collection of stress data throughout everyday life is now possible with the use of unobtrusive, wearable systems. In parallel with these technological advances, there is a growing societal interest in personal informatics and 'quantified self' movements, where individuals use modern tracking tools to quantitatively assess and manage health. Motivating the User Existing technologies allow for the development of data- intensive visualizations and charts for examining collected data. Unfortunately, the majority of users will often find data-intensive displays to be overwhelming and unmotivating. Figure 2: Information flow in the user- personal informatics system feedback loop operates similarly to that demonstrated by UbiFit [2] and others. Figure 1: Showcase of personal sensors. UP wristband, Zeo, Sleep-Cycle App, Phillips DirectLife, Zephyr HxM, AutoSense [1] Figure 3: Reduction of data into a motivating visualization Figure 5: Example feedback displays aim to motivate user by increasing awareness of everyday stress levels Before this vision can become a reality, however, a variety of challenges must be addressed in the fields of human- computer interaction, visualization, sensing, cyber- physical systems, psychology, and behavior change. The Challenges New methods must be developed to merge multiple sensor channels into accurate descriptors of stress and behavior. Visualizations of these descriptors must be carefully designed to help users explore and understand the relationship between their behaviors and stress. In addition, the visualizations must incorporate psychological theories of behavior change, such that when users are presented with feedback, they are indeed motivated to change their behaviors to reduce stress. Lastly, safeguards must be built into the system to prevent users from misinterpreting the feedback, and in turn taking inappropriate or dangerous action in response. Likewise, given the sensitive nature of the data collected and presented by the system, new safeguards will be needed to protect the user’s privacy [2]. Analyzing the Data Basic measurements from simple sensors such as accelerometers, chest band ECG, galvanic skin response, and respiration monitors can be analyzed in real time or in retrospect to help identify meaningful features of a user's day. [3] Over time the user will develop a personal database of historical sensor and analysis data which can be used to create more meaningful visualizations. However, the models which would help determine how best to motivate a user based on this data are still in their infancy. We believe that the use of simple, interactive visualizations are better suited to help motivate positive change, and will allow a user to explore his own personal stress profile and identify and manage stressors. Visualizations need to make use of familiar interface and game paradigms such as scores, rewards, and personalization in order to help motivate users. When these concepts are combined with the aforementioned data analysis and sensor system, an intuitive, minimalistic visualization can provide the motivation for improvement in users' stress management strategies in a fun and fulfilling way. [1] E. Ertin, N. Stohs, S. Kumar, A. Raij, M. al'Absi, T.Kwon, S. Mitra, Siddharth Shah, and J. W. Jeong, "AutoSense: Unobtrusively Wearable Sensor Suite for Inferencing of Onset, Causality, and Consequences of Stress in the Field," In Proceedings of ACM SenSys, Seattle, WA. 2011. [3] K. Plarre, A. Raij, M. Hossain, A. Ali, M. Nakajima, M. al'Absi, E. Ertin, T. Kamarck, S. Kumar, M. Scott, D. Siewiorek, A. Smailagic, and L. Wittmers; "Continuous Inference of Psychological Stress from Sensory Measurements Collected in the Natural Environment," In Proceedings of ACM IPSN, Chicago, IL, 2011. [2] Andrew Raij, Animikh Ghosh, Santosh Kumar and Mani Srivastava, "Privacy Risks Emerging from the Adoption of Inoccuous Wearable Sensors in the Mobile Environment," In Proceedings of ACM CHI, Vancouver, Canada. 2011. (.pdf) (Acceptance Rate = 24%). Times Cited: 2 Figure 4: Data Visualization created using D3 showing a matrix of scatter plots. [4] The correlation between each variable can be quickly and easily determined from the graph as well as the background color. [4] Derived from http://mbostock.github.com/d3/ex/splom.html