19% of my readers self-reported Human Benchmark sequence memory highest scores of level 21+ (after 10 attempts).
3% of my readers self-reported highest scores of 6 or less.
Evidence continues to accumulate showing that on a scale where Americans average IQ 100 (SD = 15), my global readership towers with an average IQ of 129 (SD = 19). Thus assuming a normal curve, the top 19% and bottom 3% should have IQs of 147+ and sub-98 respectively.
Assuming the sequence memory test is sufficiently g loaded, this implies level 21 = IQ 147 and level 6 = IQ 97.
Thus I would predict that a random sample of American youngish adults would average 6.84 (SD = 4.5).
Put simply:
IQ = 77 + (highest level obtained in first 10 tries)(3.33)
However one oddity about the self-reported data is that all of the people scoring 21+ score 24+. Nobody reported a score of 21 to 23. This suggest inaccuracy of self-reported data but it may also suggest that above level 21, the test starts measuring certain cognitive strategies and stops measuring g.
Pumpkin Person 
“Nobody reported a score of 21 to 23. This suggest inaccuracy of self-reported data…”
More than anything it suggests a small sample size.
Speaking of cognitive strategies, I scored 29 AND i still remember the sequence two days later. Not only that, today i scored 44 after one try. Now, this “suggests” that i a crazy high IQ, but in reality I converted the tiles into musical notes and imagined the sequence as music—something I’m pretty well trained in.
I think the technical term for this is “cheating” lol.
How is using strategies to remember the sequence cheating?
I was wondering which challenge/game on that site improves fluid intelligence the most, if any.
Also, would my 16 score imply a 110-120 IQ sd 15?
As far as I’m aware, no game or exercise has been shown to noticeably improve fluid intelligence (at least not in the long term). Scores on the task itself increase, but those improvements don’t generalize. Although that is somewhat depressing, the positive take away is that you can improve scores on specific tasks with practice.
That’s also the gimmick with all those “brain training” apps/games. You can improve your score on the game itself, but that’s about it. Total waste of time.
Yeah, I’m aware of that “task-specific” phenomenon, but there have been brain exercises that have been shown to improve Gf long term.
Example: https://www.ncbi.nlm.nih.gov/pmc/articles/PMC7709590/
Dual-N-back effects on Gf are iffy, as far as I know, though.
‘Cognitive strategy’? What, like noting down the sequence on a grid? Or just outright lying? Yeah, not exactly measuring g at that point.
My best was 11, but this is a clear outlier explainable by the repetitiveness of the pattern (avg. only 7.4 over 10 trials). Even 113 would be quite a discrepancy, assuming an overall I.Q. around three sigma as the IHIQS admission test suggests. My personality doesn’t entirely fit ADD either (e.g. forward digit span 13-14), but I’ve noted similar deficits in reaction time and general processing speed.
“Thus assuming a normal curve”
Quite the assumption—evidence?
there is none. and harvard phd in physics and goldman sachs partner london young made the same comment at steve shoe’s blog.
the bottom line is: psychology PhDs have low IQs.
So piers Morgan was fired because he claimed meghan markle was making up stuff. I find the claim that the royals wanted a white baby really dubious. Even I wouldn’t say something like that out loud to anyone. It’ seems clearer every day the Oprah planned the interview with meghan and that meghan was acting.. wouldn’t surprise me if at least 50% of Oprah most famous interviews were a work.
You’re too naive. Of course they said it
you mean the queen or who?
In a sane world you’d be allowed to say that you don’t want mystery meat in your royal family.
Thanks, Denmark!
illuminaticatblog – Harry specified it was not the queen or philip. Probably charles. Really, kind of unimportant for me. Harry possibly implying they might want to kill meghan is more important as well as the issue (reminded to us by meghan) of prince andrew and the royal family’s wanting to protect him (logical, but still mafia-like tactic).
God, I got like 10. I’m really bad at all these remembering sequences, patterns, etc. I tend to do average or worse! How psychometric are these exercises? What does that say about me intellectually?
Where are my comments addressing the article??? So my on topic opinions are now censored too?
dangerous comments are censored
How are my comments ‘dangerous’??
plant seeds
I have strep thought. I don’t have the energy to do 10 trials. The first time I tried I got 13 which is 120 IQ. But I can’t do anymore.
I don’t know whether “strep thought” is an intentional pun, Autocorrect fail, or mistake caused by strep thought, but I definitely have that too
i aint scared of the dark indeed i am the darkest
What is it about libertarianism that you like so much anyway?
The fact that their politics is so different from mine. Opposites attract.
How would you label yourself, politically, pumpkin? And political compass position? Auth right?
marxist
ahahaha at least you have more self-respect than the regular libshits and admit to being a full on leftie
Wtf loool that was unexpected, you defy all the stereotypes, except the high-IQ part, I guess. Marxism could just be the highest IQ ideology there is. Out of the ones with a large following, anyhow.
You could do a post on the sorts of people who are Marxists.
What’s doubly weird is that you’re an HBDer. I understand there’s no contradiction, but BlankSlateism and Marxism often go hand in hand
marx was wrong about a lot. so it’s better to say, “i’m a socialist who agrees with marx on a lot of things but not all.”
I would describe a lot of people on the right as being socialist. I would consider Steve Bannon to be socialist. I would consider the 2016 iteration of Donald Trump the candidate to be a socialist, based on what he was saying. I would consider Tucker Carlson to be a socialist. — glen greenwald
Marxism = an oligarchy of misguided narcissists wielding an inordinate amount of resources
Capitalism = an oligarchy of misguided narcissists wielding an inordinate amount of resources but people produce things
feudal governments were protection rackets. the lords protected the peasants from other lords. some of the peasants were happy to pay, knowing the alternative was their despoliation. that is, the government’s job was primarily “defense”.
libertarianism only works when there’s a yuge frontier, low population density, 80% of people are farmers, no large organizations, etc…
there’s nothing natural about government. there were no governments prior to the neolithic. government should be seen as just another technology which can increase the median material level of human existence.
good or bad government is mostly an empirical matter not a theoretical one. yet it’s treated as 100% theoretical. rawls’s theory is the only theory required.
This is the first and probably last time when I’ve wholeheartedly agreed with Mug of Pee about anything
But governments dont come from protection rackets. Anthropology from around the world shows that it comes from tribal affiliations and clan-grouping. I think saying its all a racket is hyper-marxist and factually incorrect.
Interesting. This formula predicts an IQ of 167.
Both my HRT scores and composite of verbal and nonverbal childhood scores are in that approximate range but I don’t think my memory is.
does the “h” in “hrt” stand for “heoflin”? sad.
the glia society is for people with low IQs.
Mug of Pee could you please watch this documentary of yourself and tell me how you feel about being a public figure?
Interesting. It gave me a 300 IQ.
very strange. if true, something must be off with the norms and/or deviation IQ is not an interval scale
My score on this test is 311 according to pumpkin person’s conversion formula.
Link to the screenshot:
https://postimg.cc/dhV9qs3J
This Matthew commenter is among the top most 10 intelligent commenters of all time on this blog. That much is certain.
It is amusing that his name is reminiscent of a fictional character who is a centibilliionaire with a perfect mega test score and maximum raw scores on every WAIS subtest called AdamK/Matthew. A reader of pumpkin person’s blog propagated that absurd story on quora.
Nice score. I recently got a score of 85 on the sequence memory test which gives me a 360 IQ. It took me around an hour and I got pretty tired of it.
“Matthewsaid:May 11, 2023 at 5:14 am
Nice score. I recently got a score of 85 on the sequence memory test which gives me a 360 IQ. It took me around an hour and I got pretty tired of it.”
What were your scores on number memory and visual memory?
Today I learned America has 17 spy agencies. As people who have read the Devils Chessboard will know, they exist to control you, not foreign governments.
I bet red china calls their secret police ‘intelligence agencies’ too.
Im going to google what kim jong un calls his secret police.
Ministry of State Security of the Democratic People’s Republic of Korea
Doesnt sound too orwellian. Boo.
pp, I wanted to make sure you received my payment?
No I have not but maybe it’s a sign that I shouldn’t take on anymore requests. I underestimated how much work it is.
My toshiba laptop has such a shitty trackpad that I simply can’t do this. Keep clicking the wrong one with my sausage fingers. I used to play the simon game growing up, kids these days will never understand.
I have a toshiba laptop. I think its great. It was one of the first machines with sold state hard drive. Was a revelation at the time. My old laptop took a fortnight to boot up.
I think it’s interesting how many tracking companies would love to gain access to user information like big 5 traits and IQ to send people targeted adverts based upon that. Data is $ whether u like it or not.
I am sexually attracted to the WAIS-IV.
Do we have a WAIS-chan more anthropomorphism yet?
Moe *
Trump said his favourite book was the bible. Only autistic people say he is lying. Why? Because anyone with any social intelligence knows it is hyperbole.
The idea of him reading a book is hyperbole.
How can I email you a picture of my human benchmark scores?
Sequence Memory: 50
Verbal Memory: 187
Reaction Time: 112ms
Typing: 130 WPM
Chimp Test: 18
Number Memory: 15
Visual Memory: 16
Aim Trainer: 379
my email’s on the upper left side of the blog.
Sequence Memory: 50, predicted IQ: 243
My other scores:
Reaction Time: 112ms
Verbal Memory: 183
Typing: 130
Number Memory: 15
Visual Memory: 16
I can email proof as well
^ I recognize this guy’s name. He couldn’t understand basic statistical principles when I tried to explain them to him, so take his supposed scores with a grain of salt.
Who are you? If you could point out what I don’t understand so I can try to correct it, I would appreciate that.
If I recall correctly, our one and only conversation proceeded as follows:
1. You claimed to have an IQ of 183 based on your score on Sigma Test, a test which is suspect even by the already questionable standards of high-range testing because it contains many weird items, some of which can’t even be objectively graded
2. I expressed skepticism at your score because an IQ of 183 is, by definition, only found in about 1 in 200,000,000 people
3. You replied that, no, an IQ of 183 is orders of magnitude more common than that because IQ deviates from normality at the high end
4. I reiterated that this is impossible by definition because IQs are forced on to a theoretical normal distribution, and if it were true, then it would undermine your claim of having a 183 IQ anyway
5. You nonetheless maintained that you had an IQ of 183, while also apparently not realizing that your own logic severely devalued this supposed IQ
I don’t recall 183 specifically, but I understand and don’t disagree with you. I was probably making the case that fitting IQ to the normal curve by definition might not be advised in the extremes, because intellectual ability probably has a fat tail, like many other products of assortative mating. I do understand that by definition IQ is fit to the normal curve, but wasn’t always the case, ratio scores, for example. My rarity IQ on the Sigma Test is 174 SD 16, and on their pIQ scale that’s around 195 (used to be 183 on an unfinished form I submitted). pIQ is like the scale used on the Mega Test. My WAIS FSIQ is 160, and based off of an extended-range SAT designed by a former psychometrician on reddit, my SAT-derived IQ would be about 166, but this might change because the Quant section is deflated according to her. So I’d say high 160s
is roundabouts where I’d show up on the normal curve SD 15. Obviously by pure rarity there’s no way in hell I’m at 183, but on a ratio test or if I had been given the chance to take the WISC-extended at the right age I think I could have scored in that range. This is of course because theses scores don’t tend to be normally distributed. From what I’ve read 3SD WISC scores often turn into 4SD WISC extended scores, etc. Thanks for clarifying! I have no qualms with what you’ve said, and believe I likely wasn’t being very clear, especially if it was in a discord group.
Thanks for the response.
I still contend that scores on Sigma Test mean nothing beyond ~140 but it’s good to know that you understood the point I was trying to make, lol. I agree that “raw,” “underlying” intellectual ability probably has a fat tail, especially at the extremes, like beyond four-sigma, although on the other hand it also seems to have diminishing returns at those levels.
I don’t know what pIQ is. I would normally presume “performance IQ,” but that interpretation doesn’t fit the context of Mega Test since it had a verbal section.
Basically, scores on the Mega Test were inflated, and the actual cutoff was around 4SD rather than the 5SD Mega Society claimed. pIQ was an attempt at a ratio scale on the Sigma Test. rIQ stood for rarity IQ. If you’ve seen the website which applies the log-normal distribution to ratio scores, pIQ is sort of like taking a rarity IQ and mapping it to what would be expected on a ratio scale, but Melao wasn’t using mental age, and had his own construction.
If I recall the conversion from rIQ to pIQ, it was something like rIQ + 2.023*e^((rIQ-100)/32). Comparing it to the website I linked below, it seems to undershoot ratio scores a bit, but it’s reasonably close.
As to how this relates to the Mega Test: basically, Mega Test scores were like the reverse of this. If you wanted to convert a Mega score to rarity IQ, you’d look at log-normal ratio scores and convert them to deviation scores. E.G. a 200 on the Mega Test is actually around 175 SD 16. I don’t think the conversion is exact, but it’s similar — I’ll leave a link to the site comparing ratio scores and their actual deviation counterparts.
The one thing I’m not sure of is if the log-normal distribution of ratio scores presented on this website is true to the actual distribution of ratio scores, but I think at least at the time it was made it was.
http://miyaguchi.4sigma.org/BloodyHistory/ratioiq.html
Oh yeah, and the p stood for “potential”
I guess I should also add that the central limit theorem describes the behavior of random variables, and the genes for ability really shouldn’t be expected to be random, again, due to assortative mating. This likely doesn’t matter for the middle of the bell curve, but I kind of feel like this could causes issues in the high range. Of course you can always include the normal distribution as a part of your definition for IQ, which is the case currently, but I am apprehensive if that’s wise for extremes, although it’s certainly possible!
Assortative mating would increase the variability. No reason to assume it would make the curve less normal.
If the effect of assortative mating wasn’t symmetrical, would that matter? e.g. I’d wager that 2SD+ people have an easier time finding each other and mating than -(2SD+) people. Assuming norms are static I feel l like this would have an effect. On the other hand, I think mutations which lower IQ are more common than mutations which raise IQ, especially considering that high IQ often seems to correlate with a lower mutational load. It’s possible that these effects would counter each other and simply increase the variance, as you say. The more powerful the former effect is, the more log-normal I feel the distribution would be. In that case, if log-normal is subsumed by normal, I’d agree.
In the case where the two effects balance out, wouldn’t you get fat tails on both sides? A curve extrapolated on a sample of mostly +-2SD people would produce more 4-5SD scorers than would be predicted by a purely random sample. I read for instance that in the SB5 norming there was a person with a 592 CSS, or roughly 6SDs. I don’t know how accurate that information is.
I guess I’m kind of saying that a sample of people would be closer to purely random selection in the middle 4SDs, and hence a standard normal curve would work, but that such a model would fare more poorly the further you get from that range.
I’ve not really taken statistics, and my understanding of it is dilettantish, so I’d be interested in your response, as you seem to have a strong command of the subject.
Oh yeah, and my second example feels like it could be described as increasing the variance, I’m not sure if that’s accurate though. Would the S.D change then? Sorry I can’t put too much thought into this right now, I have a test in less than an hour that I have to prepare for, so excuse any poorly-thought-out comments.
Your scores seem too high for you to need to prepare for anything tbh
I pretty much agree with most standard coursework. Throughout k-12 I never really prepared or even paid attention in particular and was able to manage basically all As, and I could intuit math especially.
In college, If I read the material once and complete my homework I get an A on the exam with no further review. In many courses I don’t even have to do the homework, but I’m also not taking a particularly challenging course load.
For the test I just took, I still hadn’t covered one third of the material, and so my prep was literally just getting exposure passively before the exam. Also, in this class (Game Theory) I have been deducted points for not following notation in the past. I’ll get the right answer, but because I haven’t seen the standard presentation, I get points knocked off for how I put it. So basically by prep I mean exposure, rather than review, and especially rather than memorization, at least in this instance.
But in general, I’m trying to get in the habit of preparing and being more conscientious even in instances where it doesn’t matter because I know those skills are important. It doesn’t really matter how high your g is, at some point you run into concepts which challenge you, and maybe those concepts aren’t accessible at all to a vast majority of people, but that’s besides the point. At any rate, you need the emotional tools to deal with difficulty. I haven’t really built those because I was kept at grade-level, so as long as I was conscious I succeeded in my coursework.
That sounds like my experience in basically every class I’ve ever been in except for Calculus 3, which strained my relatively poor spatial visualization abilities. I steamrolled most classes through purely by intuition, like Discrete Mathematics, which many people told me that they struggled in despite extensive study.
I never knew you two went to my high school! Neat!
What are you studying? What do you like to do?
I’m doing four minors at my university, three together is considered an interdisciplinary studies major. Philosophy, economics, psychology, and religious philosophy. I didn’t really pick those, and I had planned to drop out of college previously due to poor fit, but the dean suggested I try my school’s online program because it’s self-paced. The other plus side is that the online version of my school is a bit cheaper due to the subtraction of the student activity fee, so my scholarship covers all costs and then some, which means I get paid around $3k per semester to go to school. The down-side is that the major options are severely limited in the 100% online version of my state school, so I sorta had to cobble together random subjects I don’t necessarily find particularly valuable to the advancement of my career, but I don’t think I could follow a very traditional path regardless.
There are a lot of things I enjoy: Making beats for various genres in Ableton, soccer, Starcraft II (almost got Grandmaster, but I haven’t been playing recently), writing, hanging out with friends, problem-solving. I could go on for a long time but I think what I’ve put so far is what people generally find relevant when asking such a question. Currently, I want to get in the habit of reading more.
Cool, Steven, I also play SC2. I never got above Plat though, and I always preferred the Arcade custom maps anyway. I’ll post my battle.net friend code here tomorrow when I can get it
Awesome, I’ll fetch mine as well after I finish some work!
CosmicAC#11505
i would do terribly on this test