I’m working on a self-blinded experimentation platform to make self-blinded experiments easy. We are starting small and aiming to have a self-blinded L-Theanine experiment ready to ship by Dec 1st. We will provide a combination of placebo and real capsules to you (50/50, but you don’t know which are which) along with a daily self-surveying framework to track results.
I’d love any feedback you have.
What nootropics/supplements do you want to try this with?
Would you use this? Why or why not?
How long of an experiment would you want to do on yourself before being satisfied with the results? (1 week? 1 month? 6 months?)
Would you be interested in tracking other things like exercise, diet, or meditation (obviously they would not be blinded, just tracked)?
As opposed to a randomized ratio? Since the days that you are getting L-Theanine vs the days that you are getting placebo are still random, it is still solid - however, I do see your point. Is there some writing on the subject?
Great idea! I’d suggest putting up a privacy policy and some kind of “about us” page, or else you’ll scare off those of us who don’t like to order pills from random Internet people
One month seems sufficient to establish if something that is supposed to have a more or less immediate, noticeable effect works (maybe even less for people who have a very regular schedule). If you wanted to know if (and how fast) you build up tolerance to a substance, several months.
@ejain I agree about the about page - putting one together this weekend.
Also agree about the month thing. I’m thinking we’d like to offer longer-term experiments as well for people that want to see how things play out over a longer period (eg tolerance, like you mentioned). And it’s going to be interesting because each substance will have different requirements based on their half life and mechanisms of action. Eg, SSRIs take time to build up. Caffeine has a fairly short half life. Etc (Not that we are going to use either of those, but as examples hah).
Cannot find any reference to something similar, the design is really new. By the end of the experiment subjects will have a reasonable feeling of what they got, as they know the 50/50 ratio. Maybe the results will reveal a slight bias at the end.
You can reduce it by hiding the previously entered results before the end of the experiment.
You can strongly reduce it with blind ratios in the range of 25-75 for the next experiments.
@Skav I don’t think it’ll be too biased by previous results because the expected effects of the drug are already readily available to anyone that cares to research them (in theory). Or another way to put it, it’s no more biased than any existing trial would have been.
I do think the 25-75% ratio of placebo could be interesting, although I worry that people might feel a bit slighted if they receive more placebo than actual substance, given that part of what they are paying for us the supplement. What are your thoughts on this?
At the end of each day you will fill out a short survey with a few questions evaluating the effects of the pills.
Do you expect the survey questions to be sufficient, vs., say, a Quantified Mind cognitive test? The site is well-suited for this type of experimentation.
What nootropics/supplements do you want to try this with?
Modafinil, as long as you manage to guarantee users will stay legally safe.
How long of an experiment would you want to do on yourself before being satisfied with the results?
Do you expect the survey questions to be sufficient, vs., say, a Quantified Mind cognitive test? The site is well-suited for this type of experimentation.
I’m in contact with Quantified Mind and considering using them in addition to surveys.
Modafinil, as long as you manage to guarantee users will stay legally safe.
Modafinil is probably not going to make sense for our model because it requires a prescription in the US, so users would have to supply their own. How would you organize this experiment if it were up to you?
It depends on the supplement, but I think at least month, and at most six months (see the Gotu Kola supplementation, which allegedly kicks in only after 6 months).
Cool, thanks for the feedback. I agree that it’ll depend on the drug and how long it takes to take effect. Bacopa is another one that doesn’t seem to work until a few months in.
Related: a Duke University study showed that Costly Placebo Works Better Than Cheap One.
I’m speaking in front of 500 supplement company CEOs this June. Would love to have a call with you to discuss your platform. I personally think it would be ideal for you to provide supplements for free and have the companies pay for them.
@Toby_Fox We started in early December but had a lot of people that requested to wait until after the holidays so there are still people that haven’t completed it yet. I’ll follow up with more info.
