• Select Your Topic Then Scroll Down
    Alcohol Bupe Benzos
    Cocaine Heroin Opioids
    RCs Stimulants Misc
    Harm Reduction All Topics Gabapentinoids
    Tired of your habit? Struggling to cope?
    Want to regain control or get sober?
    Visit our Recovery Support Forums

Misc My Paper: "Noopept & The Placebo Effect"

Mihkal

Greenlighter
Joined
Nov 17, 2011
Messages
23
I couldn't figure out a way to post the table without losing the formatting so I just put it up in image form. Apologies for that...
Anyway, I started this experiment almost a year ago and only finally got around to typing it all up. So let me know what you think!

Noopept & the Placebo Effect: Searching For Objectivity With A 20-Person Double-Blind Study


∙∙∙∙∙∙PREFACE∙∙∙∙∙∙

Please note that even though is formatted like a scientific research paper, it is more of an anecdotal story. My limited access to funding and other resources, combined with the highly subjective nature of the properties analyzed, means that this study is far from scientifically sound. I am not a doctor, psychologist, neuroscientist, or biochemist, but I believe this little home-brewed pharmacology experiment still adds to the limited existing literature on Noopept's effects, and can help others develop a methodology for performing such trials themselves in the future.
And who knows? Maybe someone will be inspired by this to attempt to repeat my experiment.


∙∙∙∙∙∙ABSTRACT∙∙∙∙∙∙

The paper compares the reported psychological effects of controlled doses of noopept and placebos as self-administered by 20 healthy human subjects in an attempt to prove nootropics have objective effects which can be distinguished from those of a placebo. In a double-blind study, 16 individuals were given 48 20mg doses of noopept combined with 250mg of choline bitartrate with a recommended regimen of two doses per day or as desired, and an addditional 4 participants were given 48 doses containing only choline. They were asked to record any perceived effects over the course of the study, and asked about their experience at the conclusion. The percentage of subjects who reported experiencing "nootropic effects" was 68.75% among those given active doses, compared to 25% in those given placebos. The correlation is examined between consistent dosing and reports of effects, an analysis is performed on the reported effects shared by both active and placebo groups, A final review examines the experiments strengths, flaws, and lessons


∙∙∙∙∙∙INTRODUCTION∙∙∙∙∙∙

-The Elusive Quantification of Nootropic Effects-
Nootropics are somewhat loosely defined as any substance which when ingested results in improved mental functions such as cognition, memory, intelligence, motivation, attention, and concentration.[1] These are very subjective effects, which makes them hard to measure accurately and quantify. After all, if you're concentrating on how much you're concentrating, aren't you then distracted? How does one know that the reason they improved on a memory test is solely because of a substance they ingested and not the result of some other possible cause, like a better night's sleep or the development of a system to better recall sets of words/pictures/objects over the course of an experiment? Numerous unofficial reports have been written by individuals describing their experiences with nootropics, but they often lack important information like height/weight and dosage regimen. The conclusions span the spectrum from glowing positive reviews to dismissive accounts of inactivity. Few official studies existed that attempted to determine if a substance was more effective as a nootropic in healthy people than a placebo, so I decided to pool what resources I had and conduct my own little experiment.

-"The Original Novel Nootropic and Neuroprotective Agent Noopept"-
After months of searching out information regarding various nootropics, I chose to study N-phenylacetyl-L-prolylglycine ethyl ester, a Russian pharmaceutical compound of the racetam family used for a variety of neurological ailments and better known by the brand name Noopept. There are four reasons for this. First, it's very potent compared to many nootropics, which would simplify some logistics of the study (such as capping doses). Second, it has an established safety profile and history of use in humans. Third, there was a lack of reliable information available at the time regarding its effectiveness as a nootropic, so the study had more potential to be useful to others. Finally, noopept's reported effects are uniquely intense compared to descriptions of other nootropics. Most racetams require regular administration over long periods of time before any results appear, while noopept supposedly took effect within an hour of ingestion. This property would hopefully minimize the number of vague or questionable effects reported. 30 days was estimated to be more than long enough to allow effects to manifest, and 20 volunteers were recruited to see which effects were consistent between individuals and which were merely individual reactions wrongly attributed to Noopept. A dosage regimen was to be strictly-followed so as to control as many objective parameters as possible. Ultimately however, an effective nootropic is defined by its impact on subjective parameters, so aside from answering some simple questions I left it to the participants to keep note of their experiences.

-Double Blind Studies-
Although gathering personal reports from 20 people in a controlled study would still have helped shed some light on noopept's effects, I decided to increase the potential objectivity of the experiment by making it a "double-blind study," one of the most effective methods for distinguishing the effects of a substance from other influencing factor. A double-blind study is the "comparison of the effect of a drug or other intervention in a group of subjects with that of a placebo in a second similar group, when neither those taking part as subjects nor the investigators observing the effects are aware of the group to which any subject has been allocated, until after completion of the study." The idea is to eliminate bias - even subconscious bias - towards a specific outcome by those involved in the experiment. By making sure neither I nor the person taking the noopept were aware of which doses were active and which were placebos, any results could later be attributed to noopept, observable conditions, or "the Placebo Effect". I hypothesized that the active group would have clearer descriptions of effects than the placebo group, but the most intriguing conclusions would come from analyzing the list of effects reported by both. It was also possible that the most positive reviews of "enhanced cognitive function" would come from those who had been given placebos, and this would cause the entire idea of nootropics as a class of supplements to be called into question.


∙∙∙∙∙∙MATERIALS & METHODS∙∙∙∙∙∙
-Assembling a Research Group-
In pharmacological trials, a large number of participants is desired to help minimize the influence of individual anomalies that can skew results. I didn't have a university paying to place an ad in the paper for volunteers, so I was lucky to locate 25 friends and acquaintances willing to entrust their fragile nervous systems to my very limited clinical experience. Each potential subject was given a detailed summary of noopept's pharmacological profile, including method of action and known side-effects, in addition to links to other resources if they wished to investigate for themselves before confirming their participation. It was important that anyone who got involved be aware of the possible risks beforehand and not drop out of the experiment, as this would mean the time spent preparing their doses would be wasted. A list of reported drug interactions was provided (namely, alcohol and stimulants) to each potential subject, as was a link to drug interaction checkers at two different websites. They had to use different members of the racetam family as rough approximations since the drug checkers did not list noopept among their references. As a result of this, 5 of the initial 25 volunteers opted not to participate in the experiment, most of them concerned noopept could interact negatively with a drug they were currently prescribed. The remaining 20 did not exhibit a great deal of diversity, being composed entirely of white Mid-western Americans aged 22-30, but it was discovered that they had valuable differences between when they answered questions regarding their physiologies, drug histories, and general lifestyles. Table 1 provides a short summary of this information. A few participants were less cavalier than the others about putting an obscure Russian Alzheimer's medication into their body, asking to review all the noopept-related literature I had, but they were all strongly unified by one particular characteristic: curiosity.

-Determining A Dosage Regimen-
Perhaps I've become too used to seeing long lists of hazards and disclaimers on consumer products here in America, but the official directions printed on boxes of noopept from manufacturer JSC LEKKO Pharmaceuticals seem very sparse to me. It becomes very apparent that Russia does not believe in legally-required idiot-proofing when you read the only directions provided with the drug: "Recommended Dosage: 10-40mg". Thankfully, noopept has been established as very safe, with an impossibly-high lethal dosage and only a few mild negative side-effects[3], but I still read as many accounts from others on numerous online communities to come up with a dosage regimen for all participants that would maximize the potential positives while avoiding negative ones. A dose of 20mg twice a day, once in the morning after breakfast and one in the afternoon, seemed the most conservative amount that would likely generate results, skipping every seventh day to avoid an undesirable increase in tolerance. Since an increase in the body's ability to process choline is one noopept's mechanisms of action, it was determined that a supplement of 250mg of choline bitartrate taken with each dose would improve the likelihood of results while simultaneously avoiding headaches induced by choline deficiency. The volunteers were asked to record any thoughts, sensations, and events which they felt might be a result of the noopept, as well as any personal observations or feelings regarding the substance's activity or inactivity.

-Sourcing-
It was several months after my initial conception of the experiment that an opportunity to affordably purchase a sufficient quantity of noopept presented itself. Noopept is not currently a controlled substance in the United States, but at the time this experiment was conducted (Sept. 2012) it was not easy for the average American to obtain in significant amounts at reliable purity for a reasonable price. Most of the available supply was either low-grade bulk knock-offs manufactured in China or expensive boxes of capsules smuggled through the Russian black market. Thankfully, I met an individual through a chemical research forum with an interest in nootropics who had connections in Russia. He planned to capatalize on the weak US supply by importing a large amount of noopept to resell domestically, and offered to have 20 grams seperated out and shipped to me from his importer. He sent me an NMR from the lab and even gave me the international tracking number, which was as much of a proof of quality as I was likely to get without paying for an independent assay. When I received the package from his reshipper, my doubts ceased when I read the Russian name on the return address. As a personal experiment, I decided not to test the substance on myself so that I could enter myself as one of the subjects in the study. Without foreknowledge of what noopept felt like, I could still potentially receive a placebo and not realize it.

-Metrics-
Much of the objectivity of the experiment depended on the doses being consistently accurate, so a great deal of time was spent preparing them. Each subject needed 2 doses per day, 6 days a week, for 4 weeks - 48 total (2 *6*4=48) - so I decided to just add in 2 extras to make a gram's worth for each of them (50 doses * 20mg/dose = 1000mg). Both the noopept and choline bitartrate were thoroughly and evenly powdered before any measurement took place. Then, using a well-calibrated milligram scale, 100mg of noopept were weighed out at a time and poured onto a smooth surface where they were split into 5 equal parts. Each of those parts was then scooped into a size 0 capsule and placed into a capsuling machine. I knew based on prior measurement that 250mg of choline bitartrate filled approximately half a capsule, so I simply poured the choline over the machine and leveled all of the capsules off before closing them. Because choline left such a strong odor to the capsules, it was felt that filling the placebos with choline would make them more convincing. This meant there were no totally inactive doses in the experiment, some just contained no noopept. The supply of choline bitartrate ran out after 700 doses, so to prevent delaying the start of the experiment, subjects were given half of their noopept supply with the other half to be distributed after two weeks. The second batch of choline bitartrate was less finely-powdered and had a more crystalline appearance, so the last 300 doses made were slightly less uniform in weight and looked different. At both dose-preperation session, a sample of 50 capsules (10%) were randomly chosen to be weighed to ensure consistency.

-Seeding the Placebos-
The subjective nature of the effects being tested meant that no method for determing the statistical power of the sample group could be devised. This made it difficult to decide on the number of subjects who would receive placebos. Too many placebos would dilute the sample pool so much that the genuine effects of the active doses might be hard to confirm, while having too few placebos risked thinning the experiment's control group too much to act as a reliable reference point. Eventually, it was decided that 4 subjects would receive placebos since 20% seemed an appropriate proportion, and the number remained at four even though my decision to join the experiment changed that fraction to 23.8%. A third-party with no interest in the outcome of the experiment was contacted to objectively select the placebo group's members. To ensure complete objectivity, the third party was given only numbers to represent the subjects. They then rolled a pair of dice 6 times for each number, recording the outcome of each roll and adding them at the end to get a determining value. The four subjects with the lowest values were assigned to the placebo group.

-Critical Questions-
At the end of the 30 days, all participants were asked four simple yes or no questions. Did they experience any effects from the noopept? Were any of those effects nootropic in nature (improved mental functions such as cognition, memory, intelligence, motivation, attention, and concentration)? Do you think you could tell the difference between a dose of noopept and a placebo? Will you continue taking noopept in the future?


∙∙∙∙∙∙RESULTS∙∙∙∙∙∙
-Immediate Deviations-
Within the first week of the experiment, many participants tried to adhere to the dosage regimen, but found it difficult. By the end of the second week, many had adopted their own patterns or abandoned regular intake altogether. Some felt the twice-daily regimen would build an unwanted tolerance or dependency, while others wanted to limit their use of noopept to situations where they felt their memory or focus would be most challenged, hoping this would help them report more accurately on its possible nootropic effects. These changes in individual dosage patterns meant that some subjects did not consume the last of their doses until 90 days after the start of the experiment. The subjects were still encouraged to record their dosage regimen along with their experiences, but the quality and consistency of this documentation decreased significantly as time wore on, and some did not submit detailed notes at the study's conclusion.

-Reported Activity-
Thirteen of the participants (65%) claimed to have experienced effects caused by noopept, 20% said tbey did not experience any effects, and 15% weren't sure. That number rose to 75% When examining only those who received active doses, and dropped to 25% in those given placebos. Among those who claimed to have experienced effects, eleven (77%) said they would describe them as nootropic, examples given of these effects include "completing tasks quicker due to better planning", "ability to focus on one thing for longer", and "feeling like the right words just spring to mind". The remaining two (23%) felt the effects were not nootropic, with one describing noopept as a "social lubricant" that "relieved anxiety" and the other stating, "I would describe noopept as a stimulant, and I'm not sure if that really makes it a nootropic." The three subjects who answered the first question with "maybe" experienced physical or psychological changes which they felt were either too minor or too ambiguous to be attributed to noopept with certainty.

-Negative Side-Effects-
Among the accounts given, relatively few mentioned negative side-effects. By far the most common was "headache", and two-thirds of participants attributed at least one to noopept over the course of the study. Other common complaints included "irritability", "insomnia", and "upset stomach". One individual ceased taking his doses for a week after feeling "detached and analytical, like I'd become a robot with no emotion or passion". Another described an experience where "colors became more vivid, almost overwhelming" which culminated in "dizziness and shortness of breath". These reports indicate that noopept alone can create unpleasant reactions, but the most serious negative episodes were the consequence of combining it with alcohol. Within the span of three days, two subjects independently chose to take noopept after drinking 5-6 beers, despite the fact they had been warned against this at the start of the trial. The subject in the first case could not recall most of what transpired after he took the dose, piecing together his recollections from the stories of others days afterward. He was described as uncharacteristically "belligerent, confrontational, and sassy" even by drunk standards, and he was forced to compensate his roommate for things he had hurled onto the lawn from the back of their SUV during a rainstorm. In the second case, an "unremarkable night at the bar" involving "maybe 8 drinks post-pept" resulted in the subject being too disoriented to get out of a bath tub where had regained consciousness with no pants on, covered by the shower curtain, vomit, and the remnants of a potted plant he'd knocked off a stand. Following these incidents, all subjects became much more cautious about combining noopept with other substances.

-In Their Opinion-
When asked whether they thought they could distinguish between noopept and a placebo, 70% of the subjects said yes, 10% weren't sure, and 20% said no. These numbers were similar in the active group, while the placebo group was split evenly between yes and no. All subjects gave the same answer to this question as they had to the first one, with the exception of one person who said they could differentiate between a placebo and noopept despite having been uncertain about noopept's effects. This person was in the placebo group. 40% of the participants said they would try noopept again in the future, although among those who reported nootropic effects that number reached 80%. The one member of the placebo group who reported experiencing effects also stated they would continue experimenting with noopept.


∙∙∙∙∙∙DISCUSSION∙∙∙∙∙∙

-Subject 17 Corrupts Results-
The percentage of subjects who reported effects was significantly higher in the group who received active doses than it was in the placebo group, and this lends support to the idea that noopept has an objective nootropic action. However, it is worth exploring why half of the placebo group said they did or might have felt effects from their doses. It could be that subjects were so consumed with analyzing their cognitive functions that they perceived changes where there none, attributed unrelated changes to noopept, or subconsciously manifested changes themselves. Another possible explanation stems from the fact that choline bitartrate was used as the placebo instead of a totally inactive substance, and this is a possibility since the effects of choline supplementation are similar to what one might expect from a nootropic. But a simpler explanation was given by the only member of the placebo group who had reported feeling nootropic effects from his doses, Subject 17. He revealed that he had shared his capsules with one of the other subject's after noticing they differed in physical appearance. Although the experiment was double-blind and all of the doses were supposed to be equally active, Subject 17 was one of the few participants who received doses made from a different, more crystalline, batch of choline bitartrate, and thus suspected he might have been given placebos. Not too concerned with the possible impact this would have on the results, Subject 17 started taking one of the other subject's capsules about as often as he took one of his own, and by his own admission can't say for certain which capsules caused what effects. If Subject 17's responses are subtracted from the placebo group results, the percentage reporting false positives shrinks to 0%. The one "maybe" in the placebo group claimed that after taking his doses he'd been "doing a lot more reading" and "spending longer stretches [of time] researching things on the internet", but admitted his increased workload at school had also compelled him to do this.

-Mind Over Matter-
The subjects in the active group who reported no effects also had the least consistent dosage regimen, with all four summarizing their intake as "irregular". Correspondingly, all of those who stuck to the original plan (2x20mg/day) experienced very prominent effects. One could logically deduce that these subjects reduced their consumption as a result of a disappointing lack of effects. However, if the consistency of dosage were determined by the presence of effects, then one would expect to see very irregular regimens in the placebo group as a reflection of their inactive doses. Instead, the trend is present in both groups, indicating that following a consistent dosing regimen increases noopept's perceived effects. This is not as ridiculous as it may initially appear. It could be noopept gradually increases in strength as it accumulates in the body, or acts more effectively when a certain threshold is maintained. Alternatively, adhering to a predetermined schedule may create a reliable background from which to discern anything noteworthy. Perhaps confirmation bias is to blame, as the mind attempts to prove to itself that discipline pays off in the end. Given the universality of this trend across both active and placebo groups, this last explanation seems the most likely. These questions would not have arisen had all subjects followed directions, but they are a welcome avenue for further exploration.

-Sharing & Comparing-
As expected, the active group's descriptions of effects were consistently more well-defined and specific compared to the placebo group's. A report from the active group might describe "strange pressure in the left temple, been growing for 15+ mins", while one from the placebo group might say "slightly restless". What was less expected, however, was the large number of effects shared by both groups. Often these were a self-evaluation made by a subject of a specific mental faculty, usually with the only point of reference being a prior state or desired level. Examples include "feel[ing] more alert than usual", "having more motivation compared to last time", or "struggling less to recall memories". In other instances, members from either group would credit a capsule with engendering a particular emotion: "strangely depressed", "upbeat", "dealing with inferior humans makes me so angry". What these accounts describe are not the effects of a particular drug, but rather the effects of a particular kind of experiment, one in which the participant is asked to study themselves. With no criteria set in advance, subjects were left to determine the relative significance of their observations and report them haphazardly. Future nootropics investigators could prevent this by mandating that subjects attach quantitative measurements when possible ("I feel more social than usual, struck up 5 conversations with strangers") as well as provide evidence for everything they attribute to the nootropic ("The noopept was making me numb, this was the first time I'd ever watched Space Jam without laughing"). In addition, the practice of examining placebo and non-placebo groups for common elements can serve as a litmus test for objectivity, as has been shown here.


∙∙∙∙∙∙CONCLUSIONS∙∙∙∙∙∙

Despite my best efforts to devise and execute a legitimate scientific experiment according to the strict guidelines of modern pharmaceutical science, my aims proved too lofty for my means. The experiment played out very differently than was planned, and these changes weakened the data and results, There are undoubtedly still insights to be gleaned from the information, although most would consider the results of this experiment unsurprising and move on. There still has not been a human clinical noopept trial in the US, however, so with the proper resources this experiment could still answer many fundamental questions about nootropics. I may not have been able to answer them with this small study, but I personally have come up with some better questions. And in the end, I feel the nature of nootropics may just transcend ideas of subjective and objective, shrinking and shape-shifting in as attempts to elude definition. Subject 17, who stole another subject's doses because he suspected his were placebos, put it to me this way: "If all nootropics turn out to just be placebos, then they're still the best placebos ever."


∙∙∙∙∙∙ACKNOWLEDGEMENTS∙∙∙∙∙∙

Infinite thanks are due to all the great people who volunteered for this experiment. It wouldn't have been possible without them.
I'm also grateful for those who assisted my efforts in other ways, from helping find volunteers to recommending a choline supplier to staying up late with me while I capped 1,000 doses.
And of course much thanks also to the communities who inspire me, and the members they consist of. I'm so blessed to have you around, even if you're actually half a world away.


∙∙∙∙∙∙WORKS CITED∙∙∙∙∙∙

[1} Lanni C, Lenzken SC, Pascale A, et al. (March 2008). "Cognition enhancers between treating and doping the mind". Pharmacol. Res. 57 (3): 196–213. doi:10.1016/j.phrs.2008.02.004. PMID 18353672.
[2] definition of double-blind study, Dictionary of Sport and Exercise Science and Medicine by Churchill Livingstone, Elsevier Limited. 2008.
[3] [Preclinical study of noopept toxicity]. L P Kovalenko, N M Smol'nikova, S V Alekseeva, E P Nemova, A V Sorokina, M G Miramedova, S P Kurapova, E I Sidorina, A V Kulakova, N O Daugel'-Dauge
Laboratory of Drug Toxicology, Institute of Pharmacology, Russian Academy of Medical Sciences, Baltiiskaya ul. 8, Moscow, 125315 Russia. Printed in Eksp Klin Farmakol. ;65 (1):62-4 12025790


∙∙∙∙∙∙DATA∙∙∙∙∙∙
table111.png
 
What are you suggesting? Comparing noopept+choline versus active placebo is sound enough, in theory.

It would be nice to have some assurance the drug given was actually Noopept, rather than blind faith. And the lack of discipline amongst the subjects is kind of to be expected. Though it does dilute the results, it also provides some new interesting data points: with lower frequency of use, noopept is less effective.
 
Yeah, I wish had been able to do some kind of NMR or something but I just didn't have the funds. I felt it was important to mention that in the study, often we forget that there's always a slim possibility that the white powder we get isn't what we were told.

And yes, there are pros and cons to the varying dosage regimens. It would be nice to see a study where that was actually controlled though ha.
Thanks for reading!


Thanks for that. Truly, it's incredible how much we can deceive ourselves.
 
The paper compares the reported psychological effects of controlled doses of noopept and placebos as self-administered by 20 healthy human subjects in an attempt to prove nootropics have objective effects which can be distinguished from those of a placebo. In a double-blind study, 16 individuals were given 48 20mg doses of noopept combined with 250mg of choline bitartrate with a recommended regimen of two doses per day or as desired, and an addditional 4 participants were given 48 doses containing only choline...The percentage of subjects who reported experiencing "nootropic effects" was 68.75% among those given active doses, compared to 25% in those given placebos.

This is a strange design. So the blinding & randomization was only whether a subject was in the n=16 Noopept group, or the n=4 placebo group? If you were trying to test whether self-reports would differ between groups, this is wildly inefficient as a between-group design: you should have had 10 Noopept and 10 placebos (diminishing returns mean that you shrink error more by balanced groups). Even better, you should've done a within-subject repeated-measures design (https://en.wikipedia.org/wiki/Repeated_measures_design), where subjects were given both placebo & Noopept doses, because then you could compare each subject's placebo ratings to their Noopept ratings, and control for people's variations in how suggestible they are.

The remaining 20 did not exhibit a great deal of diversity, being composed entirely of white Mid-western Americans aged 22-30, but it was discovered that they had valuable differences between when they answered questions regarding their physiologies, drug histories, and general lifestyles.

No no, this is awful! You're trying for external validity by pointing out how diverse your sample is, but you're eliminaating what little internal validity you had by increasing the variance of your two populations! There was no way you were going to be able to detect any effect with a 16/4 split, and if they were all sorts of life styles and histories, that makes it even more impossible. Ideally, you'd experiment on a population of identical twins...

The subjective nature of the effects being tested meant that no method for determing the statistical power of the sample group could be devised.

You could've estimated it from doing a few blind trials yourself. You could've fed it to two or three friends as a pilot study. You could've read some existing literature on placebo effects to get an idea.

Too many placebos would dilute the sample pool so much that the genuine effects of the active doses might be hard to confirm, while having too few placebos risked thinning the experiment's control group too much to act as a reliable reference point. Eventually, it was decided that 4 subjects would receive placebos since 20% seemed an appropriate proportion, and the number remained at four even though my decision to join the experiment changed that fraction to 23.8%.

Huh? Perhaps I'm misunderstanding something here, but 'dilution' is not a relevant concept. You screwed up the power of your experiment by grossly inflating the error of your control group in exchange for a relatively smaller improvement in the estimate of your active group. 'Seemed appropriate'?

Here is an example of what I mean. Suppose you were going to do a _t_-test, and you were trying to decide whether to allocate your 20 subjects in 16/4 (as you did) or 10/10, and you know the hypothetical effect is d=0.5 (arbitrary value). Does that change your statistical power (chance your data will throw up a statistically-significant result if the effect does exist)? Yes, by quite a lot! In fact, you go from a 13% chance of successfully detecting the effect to a 19% chance, an increase of 37% (`(0.1851 - 0.1355) / 0.1355`) just from balancing your sample sizes! As I said, diminishing returns (this logically has to be true - imagine doing it as 16/4, then 17/3, then 18/2, then 19/1, then 20/0... Clearly 20/0 is useless, so if you started at 10/10, at what point did the imbalancing start to go bad?)

Code:
R> library(pwr)
R> pwr.t2n.test(n1=16, n2=4, d=0.5)

     t test power calculation 

             n1 = 16
             n2 = 4
              d = 0.5
      sig.level = 0.05
          power = 0.1355
    alternative = two.sided

R> pwr.t2n.test(n1=10, n2=10, d=0.5)

     t test power calculation 

             n1 = 10
             n2 = 10
              d = 0.5
      sig.level = 0.05
          power = 0.1851
    alternative = two.sided

(At 18/2, the power has fallen to 9%, and this library won't even calcuate 19/1.)

Despite my best efforts to devise and execute a legitimate scientific experiment according to the strict guidelines of modern pharmaceutical science, my aims proved too lofty for my means. The experiment played out very differently than was planned, and these changes weakened the data and results, There are undoubtedly still insights to be gleaned from the information, although most would consider the results of this experiment unsurprising and move on.

Still a good try. You made some big mistakes in the setup, but you learned a lot about designing & running an experiment. Maybe you can try again.
 
My head can't take all this. Can someone just tell me whether noopept works or not?
 
Hey gwern! I'm a big fan of your site and am greatly honored that you've taken the time you review my work!
I did a fair amount of research on placebo trials before I started, but the mathematical analysis aspect was definitely hard for me to grasp. I tried to explain my thinking process as clearly as possible in lieu of a solid mathematical approach, and looking back now I realize I could have devised a better method.
The reason I chose the 16/4 arrangement was because initially I was more interested in getting a lot of reports about noopept from people. It was only after the start of the experiment that I realized the most insightful aspect of the study was seeing whether subjects could distinguish between an active dose and a placebo. Really, I just didn't plan it out as well as I could have.
I have actually been planning to run a second, larger, more controlled study once I have the funds. I live in a big city now and can find more cooperative participants who'd be willing to help pay for a GC/MS. I really want to thank you for walking me through those numbers, it's helped me to gain a better conception of how I want to execute this next run. I'll be trying the repeated measures design, splitting the volunteers into equal parts placebo and active doses, as well as having everyone run through a period of taking JUST choline bitartrate and recording their reactions.
Do you have any suggestions as to tests I could put the subjects through or some other way to help quantify the data? The subjective nature of a nootropic's "recorded effects" makes them difficult evidence to deal with scientifically, and I haven't really figured out a simple, easy way to try attaching numbers to people's responses.

My head can't take all this. Can someone just tell me whether noopept works or not?
Haha well the honest answer is that this paper really can't solidly prove it one way or the other. But for those who want to investigate a little further, they can look at the data and decide for themselves.
 
Hey gwern! I'm a big fan of your site and am greatly honored that you've taken the time you review my work!

I always have time for reviewing a sincere seeker's work.

I have actually been planning to run a second, larger, more controlled study once I have the funds. I live in a big city now and can find more cooperative participants who'd be willing to help pay for a GC/MS. I really want to thank you for walking me through those numbers, it's helped me to gain a better conception of how I want to execute this next run. I'll be trying the repeated measures design, splitting the volunteers into equal parts placebo and active doses, as well as having everyone run through a period of taking JUST choline bitartrate and recording their reactions.
Do you have any suggestions as to tests I could put the subjects through or some other way to help quantify the data? The subjective nature of a nootropic's "recorded effects" makes them difficult evidence to deal with scientifically, and I haven't really figured out a simple, easy way to try attaching numbers to people's responses.

Ah, well, as always, it depends on what you're interested in. I'm guessing it wouldn't be feasible to ask them to do something like dual n-back training, as that is pretty demanding. But you could do some self-reports. For example, you could have them fill out a simple 1/2/3 item on whether they think they got the active agent (1=no, 2=uncertain, 3=yes), and that would help replicate your first experiment. While you're at it, you could do the same Likert scale for benefit/harm (1=I feel harmed / 2=I don't feel any benefit or harm / 3=I feel benefit). Hm... what else? Maybe sleep quality? Well, probably a placebo-test & benefit question is enough, combined with some background questionnaires to collect demographic info. Actually, it might be worthwhile to try a short IQ test to see if it predicts responses: given the apparent Yerkes-Dodson/u-curve effects of some nootropics, there might be an inverse relation between reports of Noopept benefit and IQ.

You don't want to ask *too* much because that probably will increase attrition, and attrition will hurt your results a lot and introduce systematic bias.
 
I would like this thread to not be purged along with the others. I feel like this experiment, though somewhat flawed, still provided some good data.
 
Top