My 16-month theanine self-experiment

This is actually a thing I wish existed, but I don't have time and energy to make it right now. I'd pay $25 a month or more.

Basically it's an application that lets you do self-experiments like this, properly blinded and with good statistics. A challenge-dechallenge-rechallenge study is one of the ones I like, but if you want to do one you essentially have to design the study anew each time, and it would be convenient to run multiple at once if that's possible.

I'm not interested in generalizing, I just want to know if (for example) taking Vitamin D every day at 1000 iu is enough, or whether I should be taking more or less. I can get labs done on this, of course, but again I'm more interested in subjective wellbeing than blood levels beyond avoiding deficiency or hypervitaminosis.

Maybe such an app exists and I simply don't know about it.

While not exactly the same https://examine.com/ is doing this sort business and seems to be popular. They aggregate and meta analyze research for pills like L-Theanine and people pay $30/month to access it

https://examine.com/supplements/theanine/?show_conditions=tr...

you want a service to: mail out a 30-pack of daily pills, there's an app where you record your mood/whatever, and then at the end of it, reveal which days were VitD and which was collagen?

how much would you pay for this?

I use Bevel for this. https://apps.apple.com/us/app/bevel-health-performance/id645...

Skip the pills. Make the app.

how do you blind the contents of the pills unless someone else does it?

I've been thinking the same thing! The app should have protocol templates for whatever intervention I'm testing, what outcomes I should track, and what confounds might be. And if it isn't part of the templates, help me develop the protocol, including being able to set what level of confidence I want. It should tell me what actions I should take to improve confidence (e.g. bloodwork). It should hide the results from me until the overall intervention is complete.

All the other health apps and tracking systems I've seen are operating at one level of abstraction too low (just the data itself and its direct insights), or try to find patterns from p-hacking passive signals (it looks like when you work out you sleep better, did you know your HRV is 1.1 higher when you do [unrelated thing]). There's no buy-in or sense of direction in these products, no pushing me to do more to acquire data, no laser focus on a targeted test / intervention.

Skip the app. Make the pills.

Put them in numbered containers at the start of the experiment.

>how do you blind the contents of the pills unless someone else does it?

My idea ( you need to adapt the numbers)

1 you build say 30 paper small bags

2 you then get 15 pills of vitamins/drug you test and 15 of placebo , you need them to be similar in shape/color

3 you put the pills in bags and make two piles, say on left side you have the vitamin and right side you have the placebo

4 you make a script to generate 30 long code numbers , you print the codes and stick them on the bags(or use a pen), my idea is that even if you now see that the code you are sticking on the bag, since it will be some long code you should forget it, or have someone else stick the codes . You save the code nubmers in a file, the first 15 codes are the medicine the last 15 the placebo

5 each day you take a bag, open, take the pill and record the code on the bag

That's not blinded though right, you'd know if you were taking Vit D or not and that would corrupt the experiment

I think this is an awesome idea. You get two boxes of daily pills labelled A and B, you take A for a month then take B for a month, and log your feelings throughout. At the end the service reveals which month was Vit D (or whatever) and which wasn't.

I guess you could get your partner or a friend or family member to do it for you?

I’m actually working on this now, starting with sleep quality and cognitive performance (memory/attention/fluency) as dependent variables.

The vision is to have an index of protocols that people can try for themselves and see whether and how broader claims apply to their own minds and bodies.

If you or anyone else is interested, please send me an email at camhashemi (at) gmail.com. I’m looking for early adopters!

A device that randomly distributes pills into two chutes.

While this is a wonderful idea in theory, a key reason it's often not used for experiments is that there are just too many variables to control for (including, for example, the ongoing effects of spending a month wondering if you're taking something that's affecting you... which affects you in and of itself).

This would work! QR codes would be even more opaque.

Don't those effects cancel with the effects on those participants who take the same 2 treatments but in the reverse order?

I agree it's not perfect, you're right to cite the Hawthorne effect. The vitamin D one for example would be affected by weather and seasonality. But I still think it would be really cool and a good product.

That’s why you need multiple participants and why this idea for doing experiments on just yourself doesn’t work well

I made this, it’s a lot less than $25/mo. Happy to take suggestions. It’s called Reflect.

https://apps.apple.com/us/app/reflect-track-anything/id64638...

I made an app for this called Reflect [1]. There aren’t templates yet but you can run self guided experiments for anything you can model as a metric in the app. I just wrote about an experiment I did with nootropic coffee [2]. I think you have a point about sense of direction and premade templates being targeted and useful. Reflect is very much generic and untargeted.

[1] https://apps.apple.com/us/app/reflect-track-anything/id64638...

[2] https://open.substack.com/pub/reflectapp/p/my-experience-wit...

Yeah, even watching the news can make your stress levels go up.