Background
I started quantifying and recording my mood as a self protection strategy while dealing with depression. What can be measured can be managed, or so my thinking went. The general idea of the Quantified Self was something I felt held a lot of value, and at the time I was enjoying exploring that concept in other avenues of my life. It felt like applying some of the QS principles to my mood might be of benefit.
Overview
Researchers are endlessly inventive, and there are any number of ways that clever people have come up with to categorize and quantify mood. I didn’t do anything like a thorough review of the literature, but I did examine several models before choosing one. For my purposes, simpler was better. I wanted something that would be easy enough for a layperson like myself to understand, simple enough to be readily available in the moment, and wouldn’t require much effort to record (the more difficult it is to do something, the less likely you are to stick with it). Besides simplicity, the other characteristic I looked for was objectivity. Because of the nature of the thing, there was inherently going to be some subjectivity, but I wanted to minimize that as much as I could. What I ended up using was the circumplex model, developed by Posner, Russell, and Peterson.
The gist of the circumplex model is that you can measure mental affect along two axes: valence (that is, the positive or negative character of how you are feeling) and alertness/activation/physiological arousal. These two characteristics of feeling are driven by two fundamentally different neurophysiological systems. Read the paper linked above for an easy to read description of the circumplex model, or read this paper by Russell for a more in depth treatment. A short summary: the mesolimbic system is associated with pleasure, reward, and negative emotions; the reticular formation is associated with physiological arousal. They are more or less independent of each other, and the two associated characteristics together give a sufficient basis for representing any emotional state.
They don’t present a method for quantifying valence and activation, but the circumplex model is simple enough that I just rolled my own. The way I did this was to take each axis, valence and activation, and scale it from 1 to 9. Most of the time our emotions live somewhere in the middle of the spectrum, and so it makes sense to give that region more space. Because of this, the scale is non-linear–the numbers are clustered towards the middle of the subjective experience. Meaning, a valence 5 is closer (in subjective feeling) to a valence 6 than a valence 7 is to a valence 8. This whole project is an exercise in pinning the objective to the subjective, and I had to do that by feel. Each value has a particular feeling associated with it; it was surprisingly easy to get a qualitative feel for each–after a little practice, I was able to stop, take a look at the quality of my, and know what value of activation and valence I am experiencing in that moment.
Knowing where I am mentally, in the moment, is very valuable, but what I wanted was a system that would let me see what my experience was like over longer periods of time. In concrete terms, the way I use this model is like this: at the end of each day, I characterize where I have been emotionally using the scale just described. For each of the valence and activation axes, I record a higher bound and a lower bound for that day’s experience (I don’t count transitory moments–just to cut down on some of the noise–I need to be feeling a particular way for a few minutes at least for it to count), and also record which third of this range I have spent the most time in (so, the upper third, the lower third, or the middle third). This is used as a sort of summary value, or best estimate of what the day was like.
Usage
Measuring and tracking my mood provides a general barometer of how my life is going, emotionally. The further back in time things are, the fuzzier and more inaccurate memory becomes. However, the numbers don’t lie. Going back to the original purpose of this project, I have set a couple of numerical thresholds, based on experience. When my mood dips below one of those thresholds, that is a concrete signal that I should take action to improve my mood. This is more powerful than it might seem at first. It can be very easy to ignore warning signs. It is easy to tell myself that my subjective experience isn’t really so bad, or tell myself that I can put up with things for a while, and put off doing anything about it. It is much harder to ignore objective numbers, especially if you can use them as a Schelling Fence.
The reason I started measuring and recording my mood came from a desire to control–I wanted a way to know what I was feeling in the moment, and predict how I was going to feel in the future so that I could take action if necessary. It has been successful in that regard, but, perhaps surprisingly, the most valuable outcome of this project is the insights that I’ve had about myself.
Insights
- The most obvious insight to come from this project takes the form of a sense of perspective. One of the core tenets of Buddhism is that all things (including emotional states) have anicca; all things are impermanent. This is self-evidently true, but our minds are trained not to live this way. To know that everything is impermanent is to live in line with reality, but it is profoundly unhelpful for predicting the future, or doing any kind of planning. The predictive processing model of the mind, which I am a fan of, is the polar opposite of this. So, we don’t live like that, for the most part. This is especially true of negative mental states. In the middle of a depressive episode, it can often feel that what you are experiencing is what you have always felt, and will always feel in the future. That is simply the nature of the beast. Even if you know intellectually that this is false, it can be hard to get that felt sense of impermanence to arise. However, looking at a graph and seeing concrete proof that you felt differently in the past, and will feel differently in the future, can be a helpful reminder of the impermanence of mental states. It also becomes apparent how much of your mood is just… noise. Some days you feel a certain way, and that isn’t necessarily for any reason. There is a lot of variance from day to day; your mood has a large random component to it, and seeing and realizing this fact makes a negative mood easier to accept. It becomes more like the weather–just something that happens.
- On the other hand, sometimes it is apparent what is driving your mood. The second big insight is understanding the magnitude of the effect that specific, important events on the mood. In the screenshot below, I’ve highlighted two periods of time. It isn’t always the case, but in these particular instances I know precisely what drove my emotions. Because I know the cause, and because I have a good idea of the magnitude of the effect, I can alter my behavior towards encouraging or discouraging those causes in the future, in a proportional way.
- Something fascinating that I’ve noticed is that much of my affect is cyclical–there are some strong frequency components to my mood. I am not at all sure why this is, and would really like to know. I would expect there to be multiple reasons for this, but the way I interpret that is that my mood has momentum, or inertia. Meaning, once it is trending in a certain direction, it is likely to continue until the attraction to the mean pulls it back to normal.
- On a more practical note, an interesting thing I learned about myself is that I am much more likely to get traction with a new project, or a new habit–or do anything that is outside my comfort zone, really–when my mood is high and my activation is high, or spiking (see the figure below for an example). There is typically a small window when these conditions hold true, usually a few days. I’ve found that saving that big project, or waiting to implement a big, anxiety-producing change in my life, until a day when these two conditions are met, is very helpful.
- Finally, being so aware of my mood has led to a much greater awareness of how my emotions work, and how they respond to the outside world. When you examine your mental state often, you come to see how particular things affect it. You begin to see and internalize the cause-effect relationships between external stimuli and internal emotions in a way that you can’t really do when you are not paying attention. This is true across the board–everything from understanding how a bit of mess causes a slight negative reaction, to realizing how fundamentally important deep connections with friends and family are. It is hard to overstate how important that internalization is. When you have internalized the cause-effect relationship between a particular stimulus and your emotions, adjusting your behavior around that stimulus becomes a natural thing. When you have internalized that fire is hot and painful, you don’t need to mentally weigh whether or not you should put your hand in it, you simply react in a manner that properly takes the expected valence from putting your hand into the fire into account. This is important, because no one has the mental effort available to consciously adjust their life to take into account all of the thousands of things that affect your mood. When you internalize those effects, you can do so naturally. It’s the difference between a virtuoso playing a piece on the piano by feel, and a novice playing that same piece from sheet music with a metronome. One is effortless, flowing, rich, and adaptable to the circumstances, and the other is difficult, brittle, lacking depth, and cannot change. The former is clearly a better way to live your life, and quantifying my mood has helped me move in that direction.
