Moodlighting Personal Informatics System
About This Project
- UMSI Course
Time of Project
- Jan 2012–Apr 2012
- Adobe Illustrator
- Peter Andrews
- Killian Escobedo
- Jeannette Schroeder
For a personal informatics course, my team and I designed and built an Arduino-based prototype for tracking light exposure. We plotted this data against various mood measures – primarily, level of sleepiness – and created a visualization showing the correlation between these two measures.
The impetus for this project was research on Seasonal Affective Disorder, which indicates that low-light exposure can lead to depression in some people. Our goal was to provide people who may be suffering from depressive symptoms a system for assessing how their daily light intake might factor into their overall mood.
My teammate took the lead on building the prototype and assembling the Arduino code. The images below show the completed device. The device contains two sensors – a luminosity sensor and an RGB-color sensor. We paired the light data with an hourly, one-question survey asking the user to report level of sleepiness.
To the right is an early sketch I made, showing what this visualization and UI might look like.
Below is an image of the final visualization. The top line shows sleepiness; greater amplitude indicates feeling more awake. The bottom line shows luminosity; greater amplitude here indicates greater brightness. The darker the shade of blue, the more blue light the user received. Some studies have found that exposure to more blue light is positively correlated with mood. Finally, the top of the visualization depicts cloud cover.
The data below is real data that was collected over the course of a day. We were happy to see a correlation between level of tiredness and light exposure. Click the image to be directed to an interactive version.
As a final stage in this project, I envisioned what this device might look like if actually brought to market. I explored many options, including a watch, a necklace, an earpiece, and more. All of these were deemed problematic as clothing or hair could easily cover the sensors, rendering the device useless.
The solution that my team and I settled on was a pair of glasses, which I mocked up to further flesh out the concept.
My team and I also presented at a School of Information-run event. The poster of our presentation can be viewed here (PDF).