Training my smartwatch to track intelligence | Hacker News Reader

Training my smartwatch to track intelligence

dmvaldman.github.io

156 pointsdmvaldman68 comments

I hope Garmin sees your passion project and greenlights it for inclusion. You have the right approach to ensuring folks are at their optimal health to grow intellectually as a person.

thank you! i think it's ridiculous how little they invest in their developer ecosystem. i have been thinking about jumping ship to oura or whoop simply because of this.

Yes, likewise! I've bounced off their developer portal several times, and I've colleagues who note they're extraordinarily difficult to integrate with.

I will gladly invest in a more accessible developer ecosystem.

> Often, it would also contradict how I was internally feeling. I’d wake up feeling rested, see my stats are low, and play a game of chess out of algorithmic rebellion, only to feel my mind up against a barrier and handedly lose.

It would be better to only look at the stats after playing if you want to verify it, this could easily be a self-fulfilling prophecy.

My thoughts exactly! The barrier could be mental, induced _by_ the app

The biggest thing for me is I don't understand how people can sleep with these watches on, it's so uncomfortable to me personally which is why the different ring technologies appeal to me more. I just wish either Garmin made one or that there was one I didn't have to buy a subscription to use.

People used to sleep with watches 50 years ago too. You just get used to it.

True. Though the difference with smart watches is that they tend to be significantly bigger in size than what common watches used to be like.

Smart "bands" tend to be less perceptible.

They’re not that much bigger than an automatic, though they are heavier than most.

Yeah, they get bigger if you spend more money on them.

I have a Garmin Forerunner 165, it is not very big, and not heavy compared to an old style watch made from metal.

I think it depends what kind of smartwatch you choose. It should fit your wrist. The amount of people I've seen with e.g. a Galaxy Ultra watch that stands over their wrist on both sides is staggering to me.

For me, the Galaxy Watch 7 in 44mm fits perfectly

Surprisingly, I find the Apple Watch significantly more comfortable and less perceptible than any other watch I've worn in the past. My theory is it's the contoured back for the heart rate monitor, etc. But maybe I'm just weird.

How didn't they sweat their watch to corrosion? Or is that just a me problem when wearing something on my wrist?

Some plastic silicone straps with the heavier overall weight make it a worse experience that traditional watches

I'm the opposite; I do not like wearing rings. I have no issue with my watch, however.

I don't wear one anymore, but I used to put in on an ankle overnight. I would wear a sock on that foot so that I didn't tear it off with my other foot during sleep.

I thought the same, I just got used to it after a few days (even if I wear a slimmer watch from Suunto)

The Garmin sleep tracker is reportedly pretty comfortable.

When I started wearing a watch, it constantly distracted me and cause minor annoyance all day while I got used to it. At some point it just started blending into the background. Now, when I wear my watch on my other hand, I still have the same problem, but I don't notice my watch normally anymore.

I've slept with my watch for a while (stopped because the battery is crap and I need to charge it every day or it won't last past noon the next day) and I've had the same adjustment period.

I also think some people are just more sensitive to these things. Some sleep with full-on CPAP machines hooked up to their faces like it's nothing, but others can barely stand wearing clothes when they're sleeping. Plus, some watches are more comfortable than others.

> I've slept with my watch for a while (stopped because the battery is crap and I need to charge it every day or it won't last past noon the next day) and I've had the same adjustment period.

I'm guessing this is an Apple Watch? Garmins typically last a week or more.

Samsung Galaxy Watch from years ago, still running Tizen. Got it as a company Christmas present, so I'm not too fussed about the mediocre battery life. From what I can tell the Garmin models in the same price range don't last much longer either.

If it makes you feel any better, I've got a perhaps Galaxy Watch 7 from almost a year ago or so running wearOS, and it's battery life is at best at the 36 hour mark regardless. Came from an older tizen-based watch with similar battery life, and at least wearOS is far more funtional.

I wear Huawei WatchFit during sleep, battery also lasts over a week. I didn't know Apple Watch batteries are that bad.

I dare you to sniff it

Nylon straps are pretty comfortable. The default strap with the metal buckle would dig in to my wrist, and also it was difficult to have a perfect fit since you had to pick a hole. The nylon strap allows for analogue refinement.

Not sure why Garmin or any of the exercise tracking watches are being used for sleep tracking. They're infamously bad at it from my experience.

The rings (notably Oura) are much better. I used to wear both and they gave completely different results, with the Oura being far more accurate to how I feel and the timings of going to sleep and waking up. Garmin almost always reckoned I woke up an hour earlier than I did and ended the tracking there.

It's honestly best not to get too involved in tracking sleep. The analysis does more to ruin your mood and give you nocebo effect than it really gives useful information.

I will confess, I do still wear my Garmin to bed because I like the vibration alarm over anything audible.

Golf watch is another beffudler to me. I don’t want anything on my wrist when I swing. I don’t even want anything in my pockets.

The Apple elastic bands are comfortable.

I have a light Huawei ("Honor") Magicwatch 2 for €45 with an old apk, and it's much better than my son's expensive Garmin. I got a good leather bracelet from a €15 Chinese smartwatch. Very comfortable. Battery lasts a week.

Much better than a smart ring.

I really wish Garmin had an official API that can be used by their users instead of the reverse-engineered solutions (although they're very good), but they're at the mercy of Garmin.

1000%

Lot of hate on Garmin sleep tracking in this thread, but I love it.

Maybe it is not super accurate, but it was eye-opening for me to see how the score changes for the worse even with a little bit of alcohol. I am way more careful with when and how much I drink since I've started wearing a Fenix 6 few years back

I didn't believe the stress numbers on my Garmin watch were very meaningful until I started taking Nebivolol (an atypical beta blocker) because there were so many gaps (even when I was sitting) that I didn't feel I could eyeball them or trust averages over time.

Taking that drug, however, it sees far fewer gaps and I show up in the blue "rest" zone most of the time.

I've been watching my heart rate a lot in the last month part because of health concerns and part because of a new stance I am practicing that has a physical component (e.g. adjusted gaits that are energy efficient) and a mental component, being an oceanic reservoir of calm with close mind-body-environment coupling 95% of the time but disconnecting that connection under peak stress -- like I am standing between two people who are screaming at each other and holding a barrier at my chest that I don't let my breathing cross and glance at my watch and my HR is 52 and it is not just the nebivolol talking because when I lose my shit it would be more like 70.

People taught me conventional Pranayama (diaphragmatic breathing) as a kid and it never helped me in "lose my shit" situations involving unstable environments and moral injury, with the intense practice I was doing recently it was clear to me that I was never going to do it better and I started researching emergency techniques for managing sympathetic overload and that one worked for me and now I feel like one of the people in [1] particularly when I show people my HRV web app [2] and demonstrate that I can turn my Mayer oscillation off

[1] https://en.wikipedia.org/wiki/Scanners_Live_in_Vain in the sense of ironclad autonomic control but with full sensory perception

[2] ... soon to be on Github

Which techniques did you try, and which worked for you?

(1) Putting cold water on my wrists. Maybe it would have worked if I tried it longer, but it didn't.

(2) Walking forward as if pushing through something with a forward lean. I was in a small building and this was awkward and my body searched through the crouching and hip swinging gaits as I tried this, I thought it might have been better in a bigger space and might give it a try sometime.

(3) What did work was the 'chest barrier' which is holding the chest absolutely rigid and NOT paying attention to whatever my body is doing to breathe. Since I have been practicing diaphragmatic breathing heavily I find that my body does that automatically and it just doesn't expand through or past my chest.

This works amazingly well. When I tried it I got flash of a scene from Neon Genesis Evangelion where the robot is raging out of control and they pull the power cable on it and it tries smashing it's way into the control room while an observer stands his or her [1] ground at the window before the battery runs out.

I added a "pull the cable" visualization and definitely sense that brave observer appear in my head, although I also realized that in this mind-body interaction it is the mind that is to blame because it brought in some trouble that it thought about from elsewhere. Trying to control the body consciously is a disaster in this situation because disturbances from the mind are coupled efficiently into the body and disturbances from the body are coupled into the mind. So I focus on holding the chest and only on holding the chest.

[1] it happens twice

Absolutely love this kind of project, combining different data sources to predict/model how you're doing. I also use chess as a proxy for my brain is working!

I've tracked sleep using a number of devices and algorithms and I haven't found a single one that regularly aligns with what and how I feel.

I know it's tracking real data, but the conclusions feel completely made up.

What are other people's experience -- especially from those who are more bullish about sleep tracking?

by tracking sleep, what exactly do you mean by that? sleep phases, sleep score, sleep duration?

Yes to all.

it's a lot of work, but something you could do is track how you feel (manually or some other way) and do a similar statistical analysis. chess elo was just convenient and aligned for me.

Yeah, I didn't mean to discount your article. Pretty clearly, it's working out for you.

I do have an iOS shortcut that tracks my mood with janky emojis. I use it at least once daily.

I track sleep and energy levels with a galaxy watch and there's a strong correlation with how I feel on a given day. Sometimes it surprises me, and day where I wake up thinking I slept well but after an hour or two I'll realize my energy level isn't great, and sure enough I'll check the app and see it's low, with a warning about my sleeping HRV or resting heart rate being away from my norm.

It's not perfect - there are definitely days when it's wrong. But overall I have a target of keeping my sleep and energy scores in the 90s and it's helpful. I think the most important thing is to keep in mind that it's an imperfect measurements but it's still the best one most of us have for now.

There are a number of studies comparing these devices with clinical sleep tracking. Just search for sleep tracking on google scholar. My take is that consumer devices have gotten pretty good at detecting when you are sleeping, but are not really good at detecting your sleep phases

Garmin’s body battery is very close to how I physically feel

Yep, can confirm. It's surprisingly accurate to the subjective feeling.

I find Garmin's sleep tracking "accurate" in that it typically matches how I feel the next day, but that also makes it not terribly useful because I don't really need a gadget to tell me when I feel tired. Mostly I wear the watch to sleep out of inertia and in case I need a flashlight in the middle of the night.

But there have been three aspects of sleep tracking that have been mildly useful:

1. A few times my heart rate variability went haywire and the sleep scores didn't match how I felt, and it turned out I was sick and had not yet noticed any symptoms. Since then it has been mildly useful to have a heads up when I'm probably coming down with something before symptoms show up.

2. You can use their Lifestyle Logging to track how things like caffeine, alcohol, and various nighttime routines affect your sleep. I mean, I haven't discovered anything that's not already common knowledge, but somehow having hard data makes it more compelling. I suppose if I was going to trial any sleep aids then Garmin's correlation would be convenient and save me from having to maintain my own spreadsheet.

3. It alters the suggested workouts if you haven't been sleeping well. Trivial to do manually, but it's a convenient reminder not to overextend.

Wrist based sleep tracking is worse than finger.

Oura for measuring sleep. Garmin for tracking physical exercise. Oura always reflected the timings and my feeling much closer than Garmin (Enduro), which basically always told me I had bad sleep (started late after my last woken moments and ended an hour early).

My own thought, it's honestly best not to track sleep unless you feel you have an underlying issue. It causes more anxiety than it solves. If you're tired, go to bed earlier, adjust tech use and food consumption before bed.

I use an Android app called "Sleep", it just works by accelerometer with your phone on the corner of your bed. I've found it's not really good with its automatic rating and suggestions, but the activity graph has an extremely reliable wave pattern on the mornings I am well rested. Like an hour and a half of no movement at all, then an hour of restlessness, which repeats through the night. Usually the hour of restlessness even has the same pattern each time it happens. But any other pattern and I generally feel tired the whole morning.

Originally I got this app for its alarm, it tries to go off in the half hour before the alarm time when you're already partially awake. The sleep tracking was just a bonus.

I’m surprised people need this data. Are they not aware if they had a good night sleep? Always had been obvious to me.

I think it’s more a way to explore the effects certain other behaviours can have on your sleep quality; for instance does a magnesium supplement help you get more deep sleep, what about workout timing does it effect how long it takes to fall asleep or interruptions and so on.

I don’t trust the raw numbers at all, really, but since i’m wearing the same sensor i can use it to spot trends at least.

Or that’s what i tell myself anyway…

How did you know it was magnesium supplement? There is no control and n=1.

Your devices are tracking sleep time, which makes about as much sense as measuring your diet based on how much time you spend chewing.

Time in deep sleep is not a direct marker of slow-wave delta power, which is just one of the measures of repair and glymphatic flush which occurs during deep sleep.

If slow-wave activity is slightly impaired, your brain can try to make up for that reduction by increasing time in deep sleep.

However, larger impairments mean your brain will struggle to stay in deep sleep.

This is why our work at https://affectablesleep.com focuses directly on enhancing the Neural Function of Sleep (and specifically deep sleep) rather than focusing on time.

> exercise-induced fatigue impairs complex cognitive tasks

Apparently your study is strictly about cardio.

Heavier compound lifts can surely knock me out for 30 minutes to 2 hours. I don't know how in the heck people train in the mornings. But a lot of this is because they are complex cognitive tasks.

Mornings (~6am), I often walk ~40min to my gym, do heavy compound lifts (~1hr total with ~2-3min rest between sets), walk ~40min back, and I tend to feel fine by the time I get back home. It's usually at the end of the day where the fatigue seems to catch up to me.

You’ve built up endurance already. The hard part of working out is getting to the point you got of not being completely sore for days after a single workout. I think that puts a lot of people off from consistent routines. I remember the first two weeks of track season no one could walk up stairs we were getting worked so hard, puking after ladder workouts 5 days a week. Eventually that petered out over the season but that initial hump from insufficient activity to being active is massive, and I don’t think a lot of us on the team would have surmounted it without essentially peer pressure and mutual support in suffering.

Oh, yea, for sure. One thing that keeps me consistent is that I know what happens when I take > 1 week off (which is sometimes unavoidable) and attempt to get back into it. DOMS⁰ for days.

---

0: https://en.wikipedia.org/wiki/Delayed_onset_muscle_soreness

Since this talks about sleep quality I want to add something that was almost unknown to me for over 30 years. Smartwatches AFAIK don't have CO2 sensors. It seems to me though, that that would be extremely useful for sleep quality tracking. I have just a couple of years ago found out that CO2 levels are having the highest impact on my sleep quality besides temperature. I can highly recommend getting a CO2 sensor to get a good feeling for that effect. Also understand how much air is available in your room and how much CO2 your body produces per hour. I totally underestimated this until I was able to measure it. In a small sleeping room with closed doors the amount of CO2 will reach unhealthy levels at the end of the night.

I'll also add in getting a standalone sensor (I went with Aranet) gave me insights into this was well, as a non smart watch wearer

In Soviet Russia, intelligence tracks your smartwatch.

In the USA, too. That's one of many methods they use to decide where to send ICE.

Isn't it a well-known fact that Garmin has terrible sleep tracking? The wearables can't handle deep sleep at all; even Muse with EEG can't reliably predict it, so I wouldn't be drawing conclusions here.

A small curiosity: I recently learned that sleep trackers in commercial wearables are terrible for people with sleep disorders like apneas, UARS, etc. It makes sense, as this isn't a typical dataset, but it's worth knowing.

https://www.youtube.com/watch?v=6FAz7QGmlBM

A petty correction to this excellent post: you say "Even a single serving of alcohol disproportionately impares REM sleep", but your source says "Reductions in REM sleep were observed starting at approximately 2 standard drinks".

thanks for catching this!

Has anyone looked into creatine? I've seen more mentions of creatine supplementation for cognition performance.

see https://fitnessvolt.com/rhonda-patrick-and-darren-candow-cre... and https://www.foundmyfitness.com/topics/creatine

Not much research in this area and I don't see anyway that a smartwatch can/could track this.

I work in neurotech/sleeptech and though I commend @dmvaldman on the project, it falls for some of the common flaws of current sleep medicine/research and some strange decisions in the program itself.

Measuring sleep by time is an antiquated idea, and the industry is just barely starting to move away from this metric. You wouldn't measure your diet based on how much time you spend chewing.

Decreased slow-wave activity, the hallmark of deep sleep, can increase time spent in deep sleep as your brain tries to compensate for the lack. This only works to a certain point, and with significant decrease, such as after drinking alcohol, your brain is unable to make up for the deficit and is actually unable to stay in deep sleep. The deep sleep length doesn't map directly to function.

Chess is an interesting metric because there is an opponent being played against, so what does that really say about the players mental clarity? There are too many factors.

I wonder if sleep regularity (consistent wake in particular) was a metric which was fed into the algorithm, and if it did not correlate?

Though many people say "garmin (smartwatch X) isn't good at tracking" that misses the point that tracking time isn't a valuable metric, and why so many people say "my sleep score doesn't match what my watch tells me".

Beyond just tracking, we have the ability to directly enhance the Neural Function of Sleep, and this is what we're working on at https://affectablesleep.com

However breaking people away from the "sleep time metric" is a challenging one.

I hear that the performance on Dual n-back is exceptionally sensitive to sleep quality. You might feel totally fine but notice that your cognition has actually tanked in measurable ways.

I love this, the interday variability in my intelligence is extremely annoying.

may the REM sleep gods light your path

Reply on news.ycombinator.com