All of these issues are resolved by having controls and by the variance within control.

Using different tests, given that the results don't correlate very well, would be a mistake.

Increasing blood flow to the large masseter muscles seems to have a broadly stimulating effect on blood flow to the head in general. You can buy unflavored gum for cheap, or xylitol gum which has a positive side effect on decreasing cavity formation.

Never thought about this one, quite interesting 🙃

So, if we are trying to prove something like:

• Plane A flies faster than plane B

We do indeed need to make sure plane A and plane B are the same in all instances, I'd say that engineering not science, but in the last 50 years little science has happened so people seem to confuse the two.

If we are trying to prove something like:

• Planes can fly

Then the specific plane design is less important.

My point here is something like "I did a thing, and people seem to have higher IQs upon retesting than control, and I'm controlling for things like motivation, memorization, exercise and diet" therefore, given that we don't expect anything (maybe sans stimulants?) to increase IQ, or at least anything besides going from slob to active lifestyle wise, this finding is interesting.

If you already believe that this sort of increase is possible and easy then without me comparing it with other such experiments and outlining my method precisely in a way that a 3rd party can replicate -- the claim would be useless. 

But that's not what I'm saying, I'm saying "Nobody has ever tried to take a cohort of adults at +1 to +3 STDs of IQ and have it improve", I did (well, people did it and I collected the data) and it seems possible, therefore something interesting is happening. Given that most people here would say this thing is not possible, this should be an update that it's possible.

I don't want to update on the precise step-by-step method by which it's possible, much like, if I had a crappy plane design and saw it fly, I wouldn't want people to replicate my crappy plane design, I'd want people to try and design better planes, after updating on the fact that building planes is, in principle, possible.

This goes doubly so if by publishing the exact plane design the FAA would sue me.

pinged you in DMs :) Happy to share, I don't need a liability waver just making sure people understand this is not medical advice, I am not a doctor+ not being assholes

That all sounds to me like increasing IQ ?

Like, if shape rotation is an underlying component of many valuable cognitive processes (e.g. math) and you get better at it in a generic way (not learning for the test)... that's getting smarter

memory tests are included within the FSIQ evaluation

That's learning effects (: The tests are the same because psychometry is BS and IQ tests aren't designed to be retaken (even though people, for some reason, make claims about IQ increases/decreases)

actually -- dmed you my signal, ping there

Can you send me your email and phone number? 

If you can get a motivated group together I might be able to fund you replicating it as long as you're ok being scrappy because I don't have that much money to throw at this

I agree, that's why I did that :)

Within the article you can find examples of that:

-> Control before vs control after p values are provided (not looking good, p value within group alone is insufficient, can capture learning, hence why I do a between group % change test)

-> Control before vs after for verbal IQ (significant -- learning effect / motivation / shorter time between tests ?)

-> Intervention before vs after for verbal IQ (no significant -- backs up hypothesis that this works for fluid IQ only, and the control effect is learning + the advantages on time & motivation)

No idea, I would re-do the tests on myself but I was semi-present for the replication so I'd rather wait more time.

All 3 of us might try to re-do the tests in a month and I can get 4-5 controls to re-do them too. Then I'd have numbers 1 month in.

This is also an important question for me.

With whom ?

Edit: I'll offer you a 50 USD even money bet that your method won't replicate when tested by a 3rd party with more subjects and a proper control group.

I doubt that will ever happen?

Like sure, I'll take that bet, but it will never be satisfied either which way.

The problem with dyi tests is that they have no external validation -- during my initial experiment I actually had a 5 min test I did 2x a day (generated so it was new problems each time) -- but the results from that don't really make sense to anyone but myself, hence why I've chosen to forgo doing it.

In terms of saturating the learning effect, that's a better approach, but getting people to put their time into doing that makes it even harder.

How is it contradicting ?

I'm saying "I don't think people that have a short enough attention span and level of interest to not even want to DM me will do a 4hr/day thing for 2 weeks"

But maybe some of them would be interested in DYI-ing it, because that's a different mindset.

So for those people I'd rather they DYI a thing.

You should try this and see, if noopept yield that much at doses where there's no CV side effects that'd be a great and novel finding

I would say this is not enough data to close the market, I'd need some 3rd party self-experimenters to replicate it.

I mean games as in "playing catch while blindfolded" physical group activities

As for calling meditation and journaling training, that just seems like motivated reasoning, under that definition anything is training.

If anything journaling would lead to better verbal results, and, well, read my analysis

one reported being significantly better with conversation afterwards, the other being able to focus much better

The reason I'm not mentioning a step-by-step guideline is as follows:

1. Giving people a step by step guide, if someone claims to follow it they could sue me (as an individual) - I don't want that. I will share it with people who seem to be good actors and actually interested - e.g. I shared it with the Lighthaven team, if they want to share it with people, I can't stop them - but the liability is on them. I obviously shared what I did with all the people that self-experimented as "intervention", if they want to plug other people in, fine by me.
2. My interest here is getting more information and aggregating it (which is public, I'm just measuring before/after for now) -- I'd like for people to reach out to me, for us to sync on data gathering, and for me to be updated on how they are doing whatever they are doing so I know if the before/after data is reliable.
3. It would take like 5 pages with if/else structures to get a perfect outline.
4. You need some dedication and common sense to do this, in that regard, it's easy enough to look at my hypothesis and run a quick literature review then reach a conclusion about what you should do -- and I'd actually like a "1000 ships" approach here where more people try to replicate in their own way -- maybe my way of putting it all together is wrong, I'm starting from out of distribution data and using theory + heuristics that I'm not an expert at to translate it into something relevant for normal people.

If what you're asking for is not "give me a step-by-step guide as to what to do" -- then I'd ask what is unclear from what I already wrote, like, the general model I have should be enough if your intention isn't to perfectly replicate this but just to extract "quick fix" style solutions for yourself (but those solution won't work)

My worry here is that I say something like "One of the things you can do is to experiment with racetams and through careful monitoring and mood-observation figure out which one to take, at what dose, through what method, so that it doesn't affect your BP or HR and feels good.  

And then someone goes ahead, overdoses on a bottle of X labeled "not for human consumption", and blames me. Or that someone does it badly (e.g. decoupled from the other interventions, without doing the careful experimenting required to figure out the correct substance/dosage/am) and then says "well, I did exactly what you said and it didn't work" -- Ignoring the fact that what I said involved 20 other things.

People tend to want easy solutions that they can apply in their sleep... and I actually want to build one, but I am like 2 years away in the best of cases, until then I'd rather recruit dedicated people to help me figure stuff out, as opposed to give advice about which I am uncertain, which people will follow poorly, and which makes me liable for spelling out those solutions

Sorry, the title was ironic, it seems that a lot of people got confused by that.

The point was more so that, indeed, there are no benefits to adding poison to your DMT...

See my correction, agree with both points, I don't think it changes the example, I did a quick google and I'm not into weightlifting/strongman stuff, so I didn't realize my misinformation was an order of magnitude off.

I still think it's essentially fair to say these dudes are "buffer" than historical dudes and seem to owe that to advances in training and (primarily) PEDs

Yeap, that was my impression. I will just confirm "no" and direct other people to confirm "yes" to you -- and, if you believe the trust adds up, you can resolve the market.

I mean if I write this it will sound very weird and not be followable because it includes things like:

Do this <weird practice> but find areas with low proprioception and do it there using something like <here's an odd sub technique I did -- but you kinda have to asses what works best for you>

I am trying to replicate this with more people right now so I'd rather not dilute the intervention specifically -- hence why this post was not about what I did as much as why one ought to expect increasing IQ, in general, works.

Somewhere in between actually, I tried to do something like (2) but in part I'm sure it's (1)

I avoided any conceptual/learning tasks and just did brain stimulation, non-stimulant drugs and various physical practices to avoid (1) as much as possible

You can toally do n-back training or take IQ tests to increase your IQ, and it's pretty boring.

(IMO: Small effects with cheap devices, unclear side effects; larger effects with medical-grade lasers, but easy to hurt yourself and also unclear side effects; having the sun shine red/IR light at you probably works better.

Define "small" ? I'm using ~100W of NIR + RED and my current EEG feedback + NIR stimulation prototype will be at 24W (but with clever use of interference, which no devices do at the moment afacit, if you know of one I'd love to buy it though)

Lovely, if I end up doing this with the LW people that data might be credible enough to close the market ?

Otherwise I can provide confirmation from the people I'm doing it with presently (all fairly respectable & with enough of a reputation in the bay tech scene)

I can ping you to resolve NO if the first run fails.

See my other replies:

Because it's an individualized approach that is a WIP and if I just write it down 99% of people will execute it badly.

If someone is smart enough to do this in a solo fashion they can literally google search for various techs used in various diseases, figure out what's easy and would fit a healthy person, then do it.

I posted a broad overview of what I did, I can't actually get it into a format where I could instruct someone to replicate everything well, that's practically my point... if this was pill-level difficulty I'd be on shelves by now, but it's not, it's easy but easy at a level that's hard to reach in current social structures.

Also I have no idea if overall +20 points is possible in healthy adults, as I push the limits playing around with this as a side project I'll figure it out (:

I believe you, but why do you only want to explain the exact stems in private messages? Are you uncomfortable giving away your work for free, or afraid that some of the methods will be ridiculed?

Because it's an individualized approach that is a WIP and if I just write it down 99% of people will execute it badly.

If someone is smart enough to do this in a solo fashion they can literally google search for various techs used in various diseases, figure out what's easy and would fit a healthy person, then do it.

I posted a broad overview of what I did, I can't actually get it into a format where I could instruct someone to replicate everything well, that's practically my point... if this was pill-level difficulty I'd be on shelves by now, but it's not, it's easy but easy at a level that's hard to reach in current social structures.

I'm perfectly fine giving this away for free, I am doing so as we speak with some people in SF :)

At this point though, I'm more interested in the trade-offs than in increasing my IQ. Education has made me more robotic, and it has made me think before I act (which does make me smarter), but this has made it harder to enjoy the moment and to "let go" and be myself.

Agree, social conditioning and amphetamins are a bad approach to increasing IQ.

Edit: Judging by the raw data, it seems like like your verbal IQ decreased? It's really important. I should know since my spatial IQ is about 50 points above my verbal IQ.

See note, I got interrupted during a time sensitive task and didn't care about it => so I moved forward, otherwise I'd have had to retake non-verbal components that I did care about.

I'm not a native speaker anyway (learned English in my teens), so the verabl IQ standalone is fairly meaningless.

Because it's an individualized approach that is a WIP and if I just write it down 99% of people will execute it badly.

If someone is smart enough to do this in a solo fashion they can literally google search for various techs used in various diseases, figure out what's easy and would fit a healthy person, then do it.

I posted a broad overview of what I did, I can't actually get it into a format where I could instruct someone to replicate everything well, that's practically my point... if this was pill-level difficulty I'd be on shelves by now, but it's not, it's easy but easy at a level that's hard to reach in current social structures.

yes

Is this in the bay?

I'm in SF right now and 6 friends (3 control, 3 intervention) are doing a self-experiment on a version of this stuff with me.

What I'd say is:

a) Wait until March 13/14th when I will have the data, that way it's not a waste

b) If you think it's a success (I can just give you the raw data and you can run your own analysis) and you have 6 or more people that want to part-take (split control/intervention, ideally -- the controls can just go after) -- I can come over for a few days and help you set up + lend some of my hardware (I'm currently working on a NIR stimulation system that uses an EEG for feedback) | If you're not in the bay I might be able to set up a remote version

I'm not making a hard commitment here since this whole experiment started as part of a broader neurotech I'm running (not for increasing IQ, but rather for preventing cognitive decline in young adults, however it's fun to play with stuff like IQ while I don't have a full neuroscience lab /w me) -- which is to say my schedule is what you'd expect from doing an early stage startup, minced -- but it would be awesome to try.

Feel free to email me (george3d6@gmail.com is what I use nowadays) and cc whoever may be interested, but I'd rather hype things up (or drop them) once I have the extra data.

Can you email me ? I'd love to set this up for you. If you're in the bay I might run a cohort for this, and if you aren't I can send you detailed instructions (but would love for you to have a control ... I am trying to get a standardized protocol so people doing things the same way ~ish would help me)

You would need ~300$ worth of equipment and substances (off amazon), and for the version I'm doing now an EEG but one may go without it I suppose.

Ideally you'd also want to throw money on some helpers along the way, I managed to incorporate most the "wowo" stuff myself but there are many ways to get "wowo" wrong, I've come to think, and I was lucky enough to filter out many wowo people in my life and build (what I think is) a pretty good framework for "how wowo applies to me" (in this case wowo standing in for stuff like "how to do Yin yoga with visualizations that enhance the poses & breath holds that help relax specific muscle groups" -- but like, 10x such things and to some extent individualized)

Still, I have little doubt that if this replicates in 6 more people it could replicate in a dedicated someone doing this remotely, even with cheap equipment & no on-site help.

Will retest in 2 weeks and probably in 6-12 months too. But some of the bits I did I quite like and I'll just keep doing

Will have an update on this in 2 weeks or so.

I realize that I was too vague with it, I think my main point is not so much:
 

This one intervention works

Because what I did was not that difficult, but rather "there's a lot of cases of IQ increases happening and people ignore them, here's why" -- hence why I lead with a study showcasing a much higher increase and advise people to do a search and see the hundreds (thousands) of studies attesting to such.

Are you interested in replicating the experiment ?

The former, there are no good tests designed to be retaken otherwise I'd use it with one.

Oh, ok, the mechanism is familiar to me and in hindsight this makes sense !

But then, my follow-up would be, if all you are doing is up/down-regulating certain proteins or regions encoding several proteins wouldn't you be able to more easily either get the proteins or plasmids or RNAviruses expressing the proteins into the brain ? Which would be temporary but could be long lasting (and cheap) and would not pose this risk

I don't particularly see why the same class of errors in regulatory regions couldn't cause a protein to stop being expressed entirely or accidentally up/down-regulate expression by quite a lot, having similar side effects. But it's getting into the practical details of gene editing implementation so no idea.

Quite confused about the non-coding region edit hypothesis.

Either you mean "non-coding" as in "regulatory" in which case... wouldn't off-target mutation be just as bad?

Or do you mean "non-coding" as in "areas with an undetermined role that we currently assume are likely vestigial" - in which case, wouldn't the therapy have no effect since the regions aren't causal to anything, just correlated? [Or, in the case where I'd have an effect, we ought to assume that those "non-coding" regions are quite causal for many things and thus just as dangerous to edit]

I don't think this would cover the entirety of science, it would just cover the bits that require statistical tests right now. I agree this is not a way to automate science, but statistical models are in themselves not expalinable beyond what a universal modeler is, they are less, since they introduce fake concepts that don't map onto reality, this paradigm doesn't.

As per reddit, I think the marginal donation doesn't affect anything, a trend might.

Here's a different counterfactual than what I gave on reddit:

If cca 80% of humans were lifelong sterile wuld you have an easier time getting orders of magnitude more funding for your research?

Can we force researchers to open data first ? and all medical institutions second ?

There are actually many potential data points here, they are maliciously hidden.

As an alternative, yes !

Or we could just spend more money and loosen regulations around artificial organs and potentially partially solve longevity while at it.

Second order effects rarely overrule first order effects

This is the type of argument that I expect from tall poppy syndrome afflicted Twitter users, not LessWrong.

This is a strewn about claim you can't possibly justify.

Sometimes this is true and sometimes it isn't, hence why I provide the example of artificial blood (both obviously doable with current tech and non-existent), it is the closest one I could find.

I'm of course not claiming this is a guarantee for kidney donation, but it is a possibility, and it's a possibility in which donation is destroying a lot of QALY compared to what it's saving. My point is not that it's certainly morally wrong, but that it isn't certainly morally right.

Yes, that was poetic license to drive the point, it is not the framing I'd used if this was a conceptual analysis paper.

It seems to me like claim (2) could easily make sense if you interpret it more charitably as "the mortality effects are too small for the studies to detect".

Bingo, partially, it's likely that at least in the Indian study the mortality was too low over that period to be accurately represented ... which is the same argument I'd have for 100% of the kidney donation studies, follow-up is not lengthy enough, and the longer you followup and the stronger your controls the worse things get.

Death is a bad endpoint for evaluating things and thus we should not be using it.

I would have a longer claim (in the linked article) that in some cases it is worth using, given that e.g. our views around why modern medicine is good and worthwhile ultimately root themselves in preventing mortality and such things are as of yet on shaky grounds.

But when doing risk estimates we should try looking at proxies for mortality and QAL downgrades as opposed to mortality, especially when we don't have life-long studies or studies following people into old age when most of them start dying.

corrected to aktually

Ok, I think I better see your point, one thing I'm not sure you are on board with is that unique language can come from:

• Direct experience of the world -- or linguistic thinking about a memory (sky here has green~ish tint before a storm)
• Linguistic thinking  -- with several nested layers of recursion (what is <abstract concept>)

Ultimately there's no hard line between the two but in practice, this seems quite relevant. Statements that come from 1) seem usually ok , and statements that come from 2) usually seem maladaptive.