Tracking mood
Experiences from a research about personal data and physical objects
In November 2017 I became a participant in a Personal Data Objects Study by design researchers from Aalto University (Finland) and University of Southern Denmark.
Here, I try to recap my experiences in the study and what I learned my self — being a research subject more than a researcher.
Interviews
On November 15th I had a brief Skype call with the researchers (Mary and Merja). A week later, they visited me at my home and interviewed me about my self tracking activities, devices I used, motivations I had and about physical objects that were most meaningful to me.
In the interview I explained (among many other things) that I’d like to track my mood. The idea is that if I had a record of how I was feeling at various times of day, week or month I could see (from other data I keep track of) which factors affect my mood. Basically, what makes me happy. I have tried using various apps (Exist, Gyroscope, Habit bull, Mood Panda come to mind) but somehow haven’t been able to keep up with a permanent habit of tracking my mood.
Mood tracking challenge
After a few months, Mary and Merja asked me to participate in a Mood tracking challenge. For two weeks beginning on February 23rd I kept a diary of my mood. Their assignment:
We would like to kindly ask you every time that you may be experiencing a mood swing, to note down the following:
• When did you have a mood swing? (date and time)
• What happened? (situation that influenced the mood)
•Who was around when your mood changed? (person/s)
• Where were you? (location you were when the mood swing occurred)
• How did your mood change? (from… to…)
Most of the records in my Google Sheet diary were just kind of realizations like “hey, this is how I’m feeling at the moment” — not big swings. Some of these I recorded immediately. At times I noticed a mood but had no way of recording it at the time. On some days I wasn’t consciously following my mood and had to reconstruct my feelings later as a part of my time tracking.
Testing prototypes
May 25th I met the researchers again and they presented me prototypes of physical objects that could be used for mood tracking. Each came with usage instructions and examples but I was encouraged to use them any way I felt comfortable with and figure out my own ways of using them. I used them during June (during working days) and July (on vacation).
I describe the prototypes one by one and comment on my usage experiences and ideas about them.
Bracelet
The first prototype was a bracelet. It had three types of beads: letters, colors and numbers. The letters divided the day into sections, colors represented basic moods and the numbers the intensity of the mood.
Entering data involved turning two beads for each time interval: selecting the mood and the intensity. Pretty nice and straight forward. Of course, the beads would also turn spontaneously when the bracelet moved on my wrist, so the data I had entered could easily be corrupted. The prototype also required me to manually transfer the information from the beads to a paper or to a spreadsheet each day, since new day’s data always overwrites the previous. A real product based on this prototype would have to be “smart” so that when you entered a mood it would be locked and logged automatically. Actually, a smart bracelet would only need one pair of beads — to select the mood and the level — since the history could be shown via a web page or a mobile app.
The best part of the bracelet was that it was visible and easily accessible. It reminded me of being aware of my mood when I happened to glance on it. I like that kind of reminders. Notifications from my phone or watch feel obtrusive and demanding, and I have a habit of mostly just dismissing them quickly.
The basic moods didn’t suit me very well, since I rarely was “sad”, “angry” or “afraid”. My moods feel more like exited, energetic, productive, creative, refreshed, focused, relaxed, lazy, tired, worried, irritated, disappointed and so on.
After a few days use I also felt that the bracelet was perhaps even too prominent and at times hindered my hand’s placement or movement when I was typing or using a mouse. I’m not into wrist-worn fitness trackers and spent many years without wearing a watch. A device has to be both really useful and satisfactorily good looking to earn a place on my wrist. The best part of the mood tracking bracelet (visibility) is also a downside.
Clicker and Slider
These were two different versions of a clicker. The clicker would just record a timestamp. The slider would also record the intensity (based on the force applied).
Both the clicker and the slider could be worn discreetly or pocketed, allowing for easy accessibility.
The data would (in a real product, not in these mock-ups) be available for later introspection with a mobile app or even a smart watch.
I really like the concept. Often it would be enough just to easily and discreetly record a timestamp to serve as a reminder of an event. The actual content of the event (the mood, intensity and possible description or other comments) could be added afterwards at a suitable time.
The downside of an unnoticeable wearable device is that the wearer tends to forget about it. If I have a clicker or slider hidden or in my pocket, do I remember to use it or do I have to resort to notifications? I tried wearing the clicker attached to my belt and to the top of my pocket in addition to having it in my pocket. It never quite managed to remind me about tracking my mood, but was always easily accessible when I noticed a mood change event by myself. I also tried to attach the slider to my keys. It was slightly too bulky but at least worked as a reminder when I got to work and home and when I left the office for lunch. On holiday time, when I used my keys less, the reminders were naturally less frequent.
Disposable trackers
Disposable trackers were inspired by the researchers’ notion that I lost some of my self-tracking data due to misplacing my trackers every now and then for days or even weeks.
If I carry just a piece of paper with me, there’s not much harm of losing it, as it could be replaced with another piece right away.
As an user interface, paper has interesting properties. The space is of course limited, but so are most digital tools. On paper, it’s harder to edit the data and correct mistakes, but mostly that didn’t pose any problems. Unfortunately, paper suffers from the fact that in order to correlate the data (algorithmically) with other data sources, the data has to be translated into a machine readable format.
Having a piece of paper on the desk beside the laptop or pinned to a cubicle wall with a map pin is also a great way of keeping them in sight (reminding of being aware of and evaluating mood) but not disrupting at the wrong time. On the other hand, portability is somewhat problematic. You can slip the paper to your pocket, but then it becomes easy to forget.
The disposable tracker prototypes had an interesting way of visualizing a day. The round form easily relates to a clock, but a clock (usually) rotates the full circle in 12 hours. With a disposable tracker I would like to visualize the whole day at once.
I tried different variations of 24 hour clocks and found the easiest and most intuitive to be one where midnight is at the bottom and noon on top. Then morning (from 6 A.M. to noon) is in the top left quadrant, afternoon in the top right and evening (from 6 P.M. to midnight) in the bottom right. Morning usually includes morning routines at home, commuting and work before lunch. Afternoon consists of work, commuting back home and often dinner time. Evening is time for family, friends or hobbies.
It occurred to me that if I had to minimize my mood tracking I would assess those four quadrants of each day. In the morning evaluate my readiness after sleep. At lunchtime think about how I have felt and what I have achieved during morning. Coming home or after dinner ponder about afternoon. And finally, before going to sleep reflect on the evening and perhaps on the day as a whole.
Different aspects can be combined to a single view by using colors. In the example photo I have evaluated my physical (body), mental (mind) and emotional (soul) state during a day. By “mind” I mean my capacity to think and handle cognitive tasks, “soul” represents mood. I might be physically exhausted, mentally focused and emotionally anxious at the same time.
Assessing three properties four times a day would be relatively easy and provide data that was easy to process and visualize. However, it’s not granular enough for pinpointing the actual factors that affect one’s mood.
Propeller
The final prototype was a box of plastic propellers and some fishing line.
The idea was to modify the propellers (by e.g. twisting or cutting) to represent different moods and attach them to a fishing line to represent a day.
I tried playing with the propellers — mostly using colors to represent moods. Although I liked the idea, I didn’t really get myself to produce anything meaningful or understandable by twisting and cutting the propellers. I could envision creating a bunch of artifacts attached to a line as a souvenir from a trip or a special time span of few days or weeks (so I could “replay” and remember what I did and how I felt during the event). The weeks when I tested the prototypes weren’t that special and the effort of hand crafting propeller lines seemed to exceed the benefits they would provide.
Summary and ideas
Using prototypes that weren’t “smart” or connected required some imagination and a playful attitude. There were nice concepts and ideas behind all mock-ups, related to the devices’ physical properties, data entry, visualization etc. None of the prototypes solved the (impossible) combination of a tracker being invisible and visible at the same time.
Wall mounted or desktop versions of feedback devices (like e.g. HappyOrNot) placed in the right places could also be a great way of mood tracking. The devices should be a lot smaller (more personal). An alternate way would to place a few NFC tags one could quickly scan with a mobile phone.
The clicker and slider reminded me of a concept I had about 20 years ago. It was mostly related to remotes and smart home control. If someone built an universal remote looking like a lipstick (with integrated display) held upside down, it could be used for variety of purposes — for controlling the volume of your stereo set as well as tracking your mood. Turn the “head” to browse between functions, push it to select, turn the head again to adjust (volume, mood, whatever based on the selected function) and — optionally — push to confirm. Make the device small enough and it could be attached to a key ring. Universal remotes like Xiaomi Mi Smart Home Cube and Bttn already exist. Nokia Thermo has a display — that also functions as a slider — and nice shape (although way too big). Someone please combine these.
Being a participant in the Personal Data Objects Study has been educational and informative. It has sparked lots of thoughts and ideas, only some of which are presented in this article. I’m looking forward to seeing the conclusions and findings of the researchers Mary and Merja.
