Fitness band fitness tracker mi band 5, featured image for blog.

Fitness bands, the quantified self, and being gaslit by technology

After years of rejecting modernity (using fitness trackers) and embracing tradition (being mindful of how my body feels and what it’s telling me) for my fitness journey, I finally caved into the siren call of technology last October and got myself a fitness band. I’d gotten into at-home workouts owing to the pandemic, and I needed something to keep me going, to hold me accountable, and to keep a record of my daily activities automatically.

I chose the Mi Band 5, because it was one of the most affordable options out there, and I wanted to test the waters before I committed to something more advanced. The reviews were pretty great, if not stellar. In terms of features, it offers the usual fare. It can track your steps, your sleep, your heart rate, and even your “stress level” using some sensors and math.

Setting up the device was easy enough. I was able to pair the band with my phone through that companion app. After setting it all up, I could see that it could count my steps, show me my heart rate, give me a “stress level” rating out of 100, and even measure how long I slept. So far so good. But you’ve read the title of this blog, and are probably wondering why I think I am being gaslit by technology.

It all began quite subtly, starting with the step counter. While the band tracks things passively, you can get it to actively monitor you by selecting one of several “workout” options. Selecting a workout allows you to record your progress and view summary stats after the workout ends. I started off by selecting the “walking” workout before going on walks. It works well enough for about 90% of the time, except for a weird quirk: sometimes the workout tracking pauses itself thinking I have stopped walking, while I am still walking. This has happened enough times for it to make me wonder how much of my walks are really recorded.

Now let’s move on to the sleep tracking. The sleep tracking gets a lot right. What it tells me about when I go to sleep, when I wake up, and how long I slept for is in line with my personal observations. It also shows me the different phases of sleep I went through in the night, and some “tips” to improve my sleep quality. One of the recurring tips it keeps showing me is that I should get more “deep sleep”. Now, I have always been a light sleeper, but I always felt like I got a good night’s sleep more often than not. Except now this fitness band comes along and tells me that I don’t get enough “deep” sleep even though I slept for a good 7 to 8 hours. Have I been wrong all along? Have I never truly had proper sleep in my entire life? Is that a problem I didn’t know I had? The gaslighting is taking hold now.

The tone of this messaging comes across as mocking, especially seeing how I do “exercise to keep fit”.

The biggest issue I have with this fitness tracker is how it utterly fails me if I try using it to track a workout that isn’t just walking. I have been working out at home for over a year and a half now. I alternate between HIIT (High-Intensity Interval Training) and Weightlifting. What I find when using the fitness band to track my heart rate during these types of workouts is a huge discrepancy between what the band reads, and what I am actually experiencing. I could be in the middle of a very intense session, and the heart rate monitor either completely freezes up, or says I am in a “relaxed mode”. This reflects in the final workout summary, where it shows a drastically different reading to the one I expect.

So how do I know that it’s the band that is wrong, and I’m not just making up these accusations of gaslighting against it? I wouldn’t have known if it hadn’t been for this YouTube video by the quantified scientist:

This is a YouTube channel run by a postdoctoral scientist in Vienna, Austria. He uses a rigorous methodology and reports on in-depth analyses that he conducts on wearable devices. It’s a unique and immensely valuable perspective on wearable technology in a space dominated by tech YouTubers and media outlets that go over the same talking points; basing their reviews on specifications and light real-world use.

For me, the two key takeaways from this review were:

  1. The band recognizes deep sleep for the most part, but it records too little of it, when compared to a more advanced sleep tracker.
  2. The band quite often misses increases in heart rate, and is not quite suitable for tracking weight lifting exercises, because it uses an optical heart rate sensor.

If I hadn’t seen this review, the discrepancy between what the tracker was telling me and my own personal observations would have kept me in a conflicted state of mind. Now that I do know the limitations of this technology, I have a couple of options before me. I could get better trackers (such as a chest strap to measure heart rate more accurately). Or, I could make do with the fact that “the quantified self” is nothing more than a silhouette, or a chalk outline on the pavement.

I’m sure getting better trackers will add some more details to the silhouette, that adding more things to measure will fill out that chalk outline with some more artistic shading. In the past year, I developed an adversarial relationship with the quantified self, because it didn’t record me well enough, and even if it could, it would always leave something out. Like an asymptote— it can come infinitesimally close to a complete picture, but it will be fully realized.

This gap between reality and the numbers presented to me by the tracker left me with nothing but dissatisfaction. What truly brought me back to being happy with myself is the “qualitative” aspect of things. Being cognizant of how I felt. I feel a lot better now than I used to feel 18 months ago. I feel more limber. I don’t constantly roll my shoulders anymore, nor do I constantly feel like my lower back is in a mildly annoying level of pain. I can lift slightly heavier weights than I could before. I can work out for longer. Muscle soreness doesn’t bother me as much as it used to. I feel more in control of my body, no longer are my body movements labored. My body feels more like a taut, perfectly tuned guitar string, unlike the loose, out of tune mess it was before.

The numbers, the scores, and the records are fine, but looking at them alone left me dissatisfied, because I knew not only how much wasn’t recorded, but also because I knew how much it couldn’t tell me about myself, even if it worked perfectly 100% of the time. This “qualified self” as I like to call it, is what brought me back to being happy with my progress and how far I have come.

Routines and the Quantified Self

What is the “Quantified Self” ?

These days some of us or quite a few of us try to capture certain minute details about our daily lives in a digital format. We keep a track of the amount of steps we have taken, the amount of calories in the meals of the day, and so on. The aim here is to keep a track of these things so that we may reflect, analyze and learn about what is going on with ourselves, to eventually improve ourselves over time. This has become much easier due to smart devices and wearable technology. Each and every one of us is generating tremendous amounts of data about ourselves every single day. Systems like the Nike Fuel band, the FitBit and even Apple and Google’s fitness oriented application suites want to take advantage of this current trend.

At the heart of this new “Quantified Self” movement are tiny, inconspicuous sensors embedded in various devices, that help record and log surprisingly accurate and incredibly detailed information. These sensors, combined with ubiquitous computing that allows these numbers to be crunched and presented to the users in an easy to understand format, and social networks that allow the users to share and collaborate, form the core of the new “revolution” in health and wellness oriented experiences.

Although all of this is a great example of how the latest technology can be used for our benefit, the idea of the Quantified Self is not as completely new as one might think. We have been keeping track of ourselves in various ways long before the advent of miniaturized biometric sensors and portable smart devices. Certain things like keeping a track of spending, or stepping on a scale every morning, have been a part of our lives for quite a while now. What’s new is this increased need for self-knowledge, helped by the rich and detailed information that can be recorded about ourselves.

Of course, there are still a few issues with the whole Quantified Self movement. One of them is keeping the user engaged. These systems currently require the user to constantly monitor or observe the information daily or over time. This may lead to information overload, or confusing the user because of too much information. Another is keeping the user motivated and interested in the system. It is observed that after a while a lot of people tend to revert back to their old ways because they get bored or lose motivation, and their fitness trackers end up in a desk drawer.

Routines

One of the things I realized as I read and researched about human factors, is the importance of routines in our daily lives. Certain things we do, certain actions that we perform, are so familiar to us that we do not spend too many attentional resources to complete those actions. They become “routines”. We continue to follow those routines until something unusual happens.

To understand how we can make the above mentioned Quantified Self systems better, we need to understand how to design them better. That’s where the understanding of routines comes into the picture. If the systems become a part of our routine, completely non-intrusive without too many requirements on our attention, they might just become better experiences.

Today’s solutions

Designers have tried to work around the issue of keeping users motivated in the case of fitness tracking. Gamification, or adding game-like interactive elements such as competition with others in your social network, trophies or achievements for achieving goals, or Role-Playing Game like elements such as character creation and progression, have all been tried out. The problem here is that it lacks a universal appeal to people. Some people really like Gamification, and others can’t be bothered with it.

Other attempts at helping users maintain motivation have been actual monetary incentives, such as the “Pact” app that allows you to bet money on whether or not someone will complete their fitness goals, or the “PavLok”, a wearable device named after the Pavlov experiment, which literally gives the wearer an electric shock if he/she does not complete the pre-decided goal.

I believe that the solution lies in understanding how routines are created, maintained and modified. Creating a new routine or modifying an existing one is difficult compared to maintaining an existing one, because changing certain habits takes conscious effort and attention. It takes a few cycles of the routine to fully internalize the changes. If it is too difficult, the individual may revert back to old habits. Superficial motivation like Gamification may not provide enough incentive to the user, to completely change their routine.

What I feel would be the ideal experience:

One of the key aspects of the quantified self is the focus on the individual. Self improvement, and detailed information that is specific to the individual are the key points of this whole experience. Using pre-set goals like “10,000 steps a day” thus seems counter-intuitive to this point. If every person is different, then every person should have goals as per their requirement, or their capacity. That is where biometric sensors fall short, and human intervention provides a more suitable solution. Sometimes it’s better to jog or run until you can feel your legs tiring out, for example, rather than just stopping after 10,000 steps every time.

That is where I feel this system needs to improve not only simply recording detailed information, but also to help create routines, and help you find your own way of making the best use of the sensor data. Information that can help you improve upon your fitness by showing you how much you can do, and what you should do to push your limits. The user would know when they have done enough, when they can feel it in their own bodies, without the need of a 3D avatar of themselves telling them they did a good job.