It’s new habit time again, and this month’s habit is going to be collecting data on my moods. Remember, I’m alternating data collection habits with other habits for the rest of the year, so that I can be full on self quantifying by 2013. So how do you quantify moods, which are inherently subjective?
As I asked myself that very question, I remembered a thread on the Quantified Self forums. There was a very useful post there by Gary Wolf. His first point was to use a three point scale, such as good-neutral-bad. That reduces the subjective noise from inconsistencies in data coding.
He also linked to an article on the site about mood measurement by psychologists. There was a paper in 1980 (James A. Russell, “A Circumplex Model of Affect”) which suggested measuring mood on a two dimensional scale. One dimension is the positive-negative scale we normally think of, sort of the happy-sad scale. The other dimension is a measure of how active, aroused, or energetic the emotion is. So anger would be negative/active, elation would be positive/active, sated would be positive/passive, and depressed would be negative/passive.
I liked this scale, it made sense to me. But as I read the article further I found there are some who don’t like it. They think positive and negative are two different scales, not opposite ends of the same scale. The basis for this is that you can feel happy and sad at the same time, and you can’t capture that on one scale.
This stuck in my craw. Not because I don’t think you can be happy and sad at the same time. Rather, I think it assumes an incorrect model of the mind. I believe in Minsky’s Society of the Mind theory, that rather than a single entity the mind is a collection of mental agents. Each mental agent itself is a collection of other mental agents, until you get down to the minimal mental agents that twitch your thumb. My interpretation of feeling happy and sad at the same time is that one mental agent is happy while another is sad. In that view, positive and negative are still opposites on the same scale. The conflict only comes when you insist on viewing multiple mental agents as a single entity.
So it seems that a two dimensional scale with a characterization of the mental agent involved is the best way to measure mood. But then how do you characterize the mental agents? Well, in addition to being a fan of Minsky, I’m a fan of the Buddha. As a Buddhist, when asked to categorize feelings that include suffering, the obvious answer is “desire.”
So to put it all together, I will measure my moods by using a two dimensional value/energy scale, with three values in each dimension, and a tag for each mood specifying the desire involved in the emotion. I will start by trying to track my extreme moods as I have them. As I expand on the habit I may add to that a thrice daily check of my general mood (after getting up, leaving work, and getting ready for bed). I may also add in tracking of common events that I think affect my mood.
At least, that’s the plan. But quantifying moods is no simple thing, and I intend to adapt my plans based on how the rubber and the road interact.