For the last few months I have been working on designing a system that enables me to gain insights regarding my moods and emotions. As I outlined in my previous journal entries, this system needs to enable me to monitor my physiological and psychological behavior, and enable me to cross-reference this data with contextual information about my activities. This past week, after a long testing and preparation phase, I finally launched this project and gave it an appropriate title: MoodyJulio.

Today I will share the completed project proposal that I presented to my colleagues earlier this week, followed by detailed information about the final design of the methodology for my data capture and monitoring system. I will also provide an overview of the next step in this project, comprised primarily of data analysis and visualization work. Lastly, I will share some interesting questions that have been raised by the work that I have done so far (several of which are not directly addressed by this project). Enough preamble, let’s get started with my final project proposal:

MoodyJULIO Project Proposal

Tracking and Measurement Methodology The design goal for the methodology that I created was to devise a system that supports “objective subjectivity”. In other words, I wanted to create an objective way to capture my subjective experience of moods and emotions. At the same time, I had no desire to apply the same level of rigor that medical or scientific studies. The system that I created has 3 core components, and 2 fun components. Let’s start with an overview of the core components:

1. Biofeedback Component: the first component encompasses the processes and tools for measuring my physiological response to the world. My focus here is on collecting heart rate and galvanic skin response data, two variables that correlate strongly with physiological levels of excitement. To collect this data I use two small devices that I wear throughout the day: a Polar heart rate transmitter and a home made device (dubbed MyMoody) that captures gsr readings and logs that data along with the heart rate information and a timestamp.
2. Self-Analysis Component: the second component is comprised of the processes and tools for measuring my emotional state. This is the main area where subjectivity comes into play, so I spent a good amount of time thinking about and testing different approaches. The solution I devised leverages the MyMoody device and a smartphone to enable objective-subjective tracking of my emotional states. The MyMoody device prompts me every 40 to 80 minutes, or whenever my heart rate exceeds 95 beats per minute, to log my emotional state. When prompted I submit a short post about my emotional state using my smartphone.
3. Contextual Component: the last core component is focused on enabling me to capture information about the context associated to my physiological and emotional states. The tools associated to this component are a smartphone and/or a computer. When posting updates about my emotional state via my smartphone I will also record my current activity and note who I am with at that moment. At the end of every day I will update my calendar with a detailed overview of my activities during that day.

Now here is a brief description of the two fun components:

1. Social-Analysis Component: this component includes process and tools for capturing other people’s perspective regarding my moods and emotions. The process for capturing this input is supported by an old school paper business card (pictured in the presentation above), and posterous.com blog, and a twitter feed. Here is how the process works: when I talk to someone I request that they submit a report regarding my mood. I give them a card that reminds them of this request and provides an email address where they should submit this information. When an entry is received it is added to the posterous blog and the twitter feed.
2. Picture-Analysis Component: this component aims to provide insights into my moods by capturing images of me when I am using my computer. The process is quite simple, whenever I am using my computer it takes a picture of me every minute using the webcam. These pictures are then stored using sequential files names.

Framework for Describing Emotions In order achieve my goal of objective subjectivity it was important to leverage a framework that enables me to communicate about my emotions in a consistent manner. Rather than attempt to re-invent the wheel, I decided to look for an existing framework that I could leverage. For a while I was leaning towards using Robert Plutchik’s wheel of emotion. However, I ended up deciding to use the HUMAINE project’s Emotion Annotation and Representation Language (EARL) because I find its structure aligns better with my data collection approach and the way I think about emotions. Interestingly, this framework was developed to support emotion-oriented computing. Interesting Questions During this initial process of data collection several questions have been percolating in my mind. These questions are not directly related to my current project but are definitely worth further thought and investigation – who knows, one of these questions may serve as a seed for my thesis project.

• How does the process of categorizing (or rationalizing) our body’s physiological responses into emotion categories impact our experience of life and the world around us. In some ways this is a process of mediation. How does this process affect our relationship with our own mind and body.
• How can we experience the world more directly without the interference of this mediation process? Is it desirable to seeks this more direct experience of the world?
• How does this type of research affect how we experience our emotions. Since emotions are constructs that are largely determined by the stories that we create to rationalize the physiological responses of our body, how does this focus on emotions impact this process (for better or worse).