CHALLENGE & OPPORTUNITY
MIT's Advancing Wellbeing Initiative
The MIT Media Lab's initiative is an ongoing platform that challenges & invites us to create methods of improving wellbeing. The participant is challenged to envision new technologies to reduce stress, illness, or depression.
As part of the initiative our team set out to design, develop, and test a system to help students in times of high stress and emotional strain.
THE IDEA
We created a Mobile App that allows users to self-repot their current & previous moods through simple interactions at the time of unlocking their phone. Through incorporating the micro-questions at the time of unlocking a smartphone, we seek to find if self-reported and self-reflective analysis of individual-based correlations between weather and mood, coupled with an affective feedback system, changes how weather affects self-reported mood, stress, and creativity.
Unobtrusive Interactions
We developed an app with a series of Affective Computing features for the user to track their moods, stress levels and degrees of sensitivity to weather. Upon unlocking the phone, the app prompts the user to answer one of a series of "micro-questions". The system then utilizes the answers to give affective and empathetic feedback.
IDEATION PROCESS
Through exploring a range of quantitative and qualitative means for cataloguing emotion, we implemented a self-reported method for the user to track the fluctuation of their emotions. Rather than a computer deducing what you are feeling, we wanted to create an interface that facilitates the user to "journal" their moods & reactivity to weather patterns. Our cross-disciplinary team felt that promoting a healthy habit of self-reflection would be beneficial to the user & in line with the intent of the Advancing Wellbeing Initiative
We were inspired by the Valence Vs. Arousal method of categorizing emotions as we developed the interface. Given the complexity & dimension of moods, we designed & developed the app to facilitate our understanding of what we feel as the weather changes.
INTERFACE DESIGN & DEVELOPMENT
We then developed a first series of questions designed to directly and indirectly inquire about the user's current mood. A second series of questions aimed to get a sense for how the user felt about the current weather. When generating the many inquiries, it was essential that the question be matched with the proper answering interface method. In order to encourage usage, an Affectively smart system was incorporated - therefore responding empathetically to the given answer.
Feedback based on user's response
Vs.
Coupling questions to gain deeper insight into mood patterns
+
As data is gathered, the system displays a series of visualizations designed to highlight potential correlations such as stress & productivity or mood fluctuations & weather patterns. By incorporating questions asking about moods of previous days, the user was able to reflect on these days with some emotional distance.
USER STUDY & IMPLEMENTATION
The system is comprised of four main parts: database, server, administrative module, and our Android app. Our server queries and stores data from the environmental sensors in the phone as well as from a designated weather database. The gathered information is analyzed and organized through the admin module so that it can be accessed by the user through the app. The participant interacts with the app in order to answer questions, receive feedback, and reflect on his or her data. .
As Released for Android on Google PlayStore
During the initial two weeks, all users were prompted with fourteen different questions per day. All of the questions at this point were inquiring, either directly or indirectly, about perceived mood and stress levels. Subsequently, once users were randomly divided into "Control" (A) and "Weather" (B) groups, each group received a different set of five additional questions.
Questions in Group A group were further asking about mood and stress, while Group B’s questions inquired on the user’s perception of the environment and weather. Furthermore, Group B also received a “refreshing” set of questions through out the last week of the study. These six refreshing questions replaced some of the original fourteen in order to relief the possible repetitiveness of always seeing the same predictable kinds of questions.
Data Visualization for Users
Project Roles:
UX Interface Design / Methods Research / Co-Author
• Collaborated in brainstorming sessions with Affective Computing Lab, MIT Media Lab
• Analyzed relevant research on influence of weather on emotions
• Researched scientific methods of assessing and categorizing emotions
• Tested Emotive Facial Recognition software (collaboration w/Affectiva)
• Executed User Experience (UX) design of app
• Designed flow of UX interface, iconography, and user data visualizations
• Produced imagery, and graphic representations for publications
• Collaborated in written project reports & submittal of conference papers
Project Credits:
Team:
• Ricardo Jnani Gonzalez. Master's of Science in Design
• Sara Taylor. Electrical Engineering, Affective Computing
Collaborating Professor:
• Rosalind W. Picard. Affective Computing, MIT Media Lab
RELATED LINKS