Ask A Genius (or Two): Conversation with Erik Haereid and Rick Rosner on Intelligence (Part Six)

Author(s): Scott Douglas Jacobsen

Publication (Outlet/Website): In-Sight: Independent Interview-Based Journal

Publication Date (yyyy/mm/dd): 2020/03/01

Abstract

Rick Rosner and I conduct a conversational series entitled Ask A Genius on a variety of subjects through In-Sight Publishing on the personal and professional website for Rick. According to some semi-reputable sources gathered in a listing here, Rick G. Rosner may have among America’s, North America’s, and the world’s highest measured IQs at or above 190 (S.D. 15)/196 (S.D. 16) based on several high range test performances created by Christopher Harding, Jason Betts, Paul Cooijmans, and Ronald Hoeflin. He earned 12 years of college credit in less than a year and graduated with the equivalent of 8 majors. Erik Haereid earned a score at 185, on the N-VRA80. Both scores on a standard deviation of 15. A sigma of ~6.13 for Rick – a general intelligence rarity of 1 in 2,314,980,850 – and ~5.67 for Erik – a general intelligence rarity of 1 in 136,975,305. Of course, if a higher general intelligence score, then the greater the variability in, and margin of error in, the general intelligence scores because of the greater rarity in the population. This amounts to a joint interview or conversation with Erik Haereid, Rick Rosner, and myself.

Keywords: America, Erik Haereid, intelligence, Norway, Rick Rosner, Scott Douglas Jacobsen, standard deviation.

Ask A Genius (or Two): Conversation with Erik Haereid and Rick Rosner on Intelligence (Part Six)[1]^,[2]*

*Please see the footnotes, bibliography, and citation style listing after the interview.*

Scott Douglas Jacobsen: Few people, statistically speaking, qualify for general intelligence quotients above 4 standard deviations. One reason remains the ceiling set on mainstream standardized intelligence tests. Another is the rarity of the population. A small number of people, internationally, developed some tests for above 4 standard deviations. 

What is intelligence? What is IQ? Why the limits on the mainstream standardized tests? What is the reliability and validity of the alternative tests for above 4 standard deviations above the norm? How many have each of you done? What is the range of earned scores? What do this score or these scores indicate about the alternative tests, the mental abilities tapped, and the conceptualization of general intelligence?

Erik Haereid: Intelligence is, strictly, about the ability to think abstract and learn new stuff. It’s about the g factor; if you are good/bad at one thing you are probably good/bad at another thing too. Since there are a lot of opinions among scientists, psychologists (psychometricians) and laymen, I conclude that there is not one single definition. We don’t know what intelligence exactly is, but that it has to do with how we learn, adapt, solve problems and understand. It’s more about how we process knowledge than knowing per se.

IQ is a measure of intelligence. One of the main difficulties by making tests that are supposed to measure intelligence is that they can’t capture the culture’s knowledge; they discriminate because some know things other don’t and score higher (gaining higher IQ) without having a higher intelligence. Most people in the world would score poorly if the test was in the Chinese, Norwegian or Swahili language or using ancient Egyptian hieroglyphs.

Other disturbing factors are such as bad preparation and performance anxiety on proctored timed tests, which seem to be the only accepted psychometric IQ-tests today. If you have 20 minutes once in a lifetime to show your level of intelligence, those sort of tests obviously will discriminate on how good you are at dealing with that kind of pressure. As far as I know, that kind of nerves are not correlated with intelligence. Then ordinary timed IQ-tests are not measuring pure intelligence. But which tests are? No tests are, if you ask me. The experts, the psychometricians claim that tests like WAIS is optimal concerning measuring intelligence. So let’s stick with that.

Why the limits on the standardized IQ-tests? They have those limits, “low” ceiling, because they are supposed to measure correct IQs. To draw the right statistical distribution, e.g. the normal distribution which one has measured and calculated as the best statistical distribution on measuring IQs, you need plenty of data, scores, testees, in each area of IQs. But since there are fewer and fewer scores at the bottom and top, you can’t say for sure that it is the normal distribution, or any other distribution, that fits that area of IQs. The higher (and lower) you get on the IQ-scale, the more uncertain is the calculation of the IQs. We know that between let’s say 50 and 150 (standard deviation 15, which I consequently mean) in IQs, we have enough data (scorers and scores) to measure a quite correct statistical distribution; the normal one fits almost perfect. That’s not the case deciding IQs under 50 and over 150. That’s why you get on most (?) proctored, psychometric accepted IQ-tests have ceilings, and that it’s more right to achieve “Your IQ is 150+” than “Your IQ is 172”.

When you increase the ceiling (more difficult items, less time), you will also get more uncertainty connected to the highest IQs. So when you gain a 145 IQ on a WAIS-test, this is probably close to your g. If you gain >155 (or wherever that limited limit on that standardized test is) you probably have an IQ over that level (155+), but without knowing what it is. If you gain 175 on a WAIS-test, you still probably have an IQ>155 and also >160, but with increasing uncertainty. Maybe it’s 165, 170 or even 180; it is uncertainty connected to those high IQ-scores compared to those <155 on standardized tests.

The Mensa entrance test (e.g. FRT) is constructed for the purpose of dividing the top 2 percent from those below, so if you gain a >130 IQ on that short 20 minutes test with limited ceiling, you probably are at the top 2 percent of the population because of the many testees in that area, and the fact that the test is highly correlated with other psychometrical tools like WAIS.

Let’s say we have developed a perfect measure of intelligence, for every level (as many say the WAIS-test approximately is today for levels < 155 or so); we have a valid IQ-measure. Because of the huge amount (I guess) of testees in the area below 145-155 (let’s say on WAIS worldwide), we can be pretty sure about the coherence between the measured data (the scores) and the distribution; they follow the normal distribution let’s say up to 150-160 (that is 155). And since we here assume (for simplicity) that the test also measures exact intelligence from IQ-levels of 155 and infinitely (that means that no one will get the full score, never) we only have to destine the IQs related to the scores. But since there are few testees among those who scores over 155, we can’t for sure say that the normal distribution counts for these scorers. So far, based on their WAIS (or whatever “perfect” test we use)-scores, we can tell that they are above a certain IQ-level, even though the test measures IQ-levels all the way (has an infinite ceiling), but we can’t say that their scores follow a normal distribution over 155 because of the lack of data to confirm that. It becomes just a qualified guess.

So, when we talk about other HR-tests (e.g. high range untimed tests) with high ceilings (>160-170), and compare these scores with standardized tests like WAIS, we can’t say for sure that the HR-tests are reliable even though they are correlated with WAIS on the high levels. If you gain IQ 170 on WAIS, LS24 (spatial HR-test by Robert Lato) and SLSE1 (numerical HR-test by Jonathan Wai), you can’t claim a reliability on the 170-level even though it seems to be, because also WAIS-scores on that level are unsure. But there is, certainly, a correlation here. Maybe the theoretical true IQ is 180 on the WAIS-test. Then your real IQ (g) maybe is more like 175 than 170. Even though the tests are correlated (you gain the same score on several, different tests) it’s not sure that (on these high levels) the statistical distribution (the formula that calculates the IQ) is correct because of the few data to predict that distribution.

When we have few data we can use non-parametrical methods, which is the second best choice. If every person in the world took this test, we would have plenty of data to measure levels up to 180; deciding a parametrical statistical distribution, not necessarily normal, from the population with IQs over let’s say 155. Now we really don’t know if this distribution (IQ>155) is normal, but we use the normal distribution to decide IQs also in this area, because the normal distribution is right to use for everyone else (IQ<155). We presume that IQs follow the normal distribution also for people >155. But this is a vague anticipation.

I don’t have any insight to data from proctored, psychometrical accepted IQ-tests in this area, so I just speak theoretically.

I presumed that the best tests, like WAIS, measures intelligence. But this is also a definition of intelligence, and I mean that the psychologists (psychometricians) are the experts. One thing is that you have a nice test and plenty of observations (scores), and therefore can predict a solid distribution (like the normal). Another is if that the test and scores really measures intelligence.

As said, let’s say you force people to take one 20 minute IQ-test once in their lives, that is said to measure something that important as intelligence, you certainly measure much more than intelligence (nervousness, performance anxiety, your relation to authorities, the culture’s weight on such tests…). Then you at least not only discriminate on the mental capacities like intelligence. It’s a lot of statistical disturbance that is difficult or impossible to measure.

There are for example some, many, that believe that untimed high range IQ-tests measure something more than intelligence; perseverance, stamina, patience and so on, in addition to intelligence. I am one of those. So when you gain a 180 IQ-score on an untimed credible HRT, you probably do have a very high intelligence, but also a high degree of stamina.

Assume that a person A (preferably a future super AI-agent) take Lato’s LS24- and Hoeflin’s Titan-test in one hour with all items right. The second best achiever is, let’s say a person B that scored 20/24 on LS24 and 48/48 on Titan, but used totally one month. B is said to have 200 in IQ, based on a dozen or two of other testees in the same range (170-200). But how should we calculate A’s IQ?

In a parametrical distribution, like the normal one, we would calculate it directly. But is this normally distributed in the end of the tail? Maybe not. The problem is how we should decide, calculate, if A’s IQ is 300, 700 or 1,000. We have a lot of statistical methods to measure uncertainty, predicting something that makes it a qualified guess, and the common factor of those methods is that when the data become fewer the guess-factor becomes larger.

This illustrates the problem with little data at the end of the tail. We can say that extraordinary achievement, considering the short time used, deserves an extreme high IQ, but we can’t know how high. The distribution is unknown in this area.

I mention two factors that influence intelligence: 1) the ability to solve abstract problems, including different degrees of complexity and a diversity of cognitive problems like in a WAIS-test, and 2) the time used to do so. I can’t see any major obstacles creating extreme difficult IQ-tests, because it’s about combining degrees of difficulty and available time or time used. It’s not any problem creating an IQ-test that measures (theoretically) IQ’s at 900 or 1,000-level.  If a person solves the Titan-test in six hours, with all items right, she/he/it would obviously have an IQ superior to the most intelligent person we know of today. The problem is to decide the IQ-level, not proclaiming that persons superior level of intelligence.

I have taken something like 30+ HR-tests since 2013, and one proctored standardized test in 2013 to get into Mensa (FRT). I am among the top scorers on several HR-tests with high credibility and ceiling. Before 2013 I was not concerned with IQ-tests. I am more interested in how we humans can use our intelligence than measuring it, but it’s a lot of fun doing these HR-tests; you sort of get addicted.

On the tests I have taken seriously I have scores in the range 145 to 185. I have a quite good assembling on some of the most accepted, respected and oldest HR-tests, like LS24, Algebrica, SLSE1, SLSE2, LSHR, some of T. Prousalis’ tests and some more. My IQ on these tests with high credibility is in the range 166-176, as I remember it, and maybe my g is about 170-171, maybe a couple of points higher since I score high on different type of tests (numerical, spatial, verbal); I don’t know, and I don’t care. I am pretty sure that my IQ (g) is in the range 0 to 200! And I am pretty sure that I am 56 years old.

Rick Rosner: Intelligence is generally finding consistencies in the world, consistencies and relationships. If you want to be slightly grandiose about it, then it is what separates human beings as generalists from other species whose search for exploitable consistencies don’t have as much fluidity as humans.

That’s it. It is figuring stuff out about the world. You do not know if something is inconsistent; until, you are aware of things. Chaos is just chaos when you haven’t pulled anything out of it. Until, you’ve pulled some things, some consistencies, out of it. Then you can find out what is consistent or not. 

If you do not know anything about anything, then that means not knowing anything about what is inconsistent. IQ is an attempt to measure intelligence via testing, standardized testing. I don’t want to go into the whole history of IQ. You can look it up. 

Basically, it started with – intelligence testing that leads to IQ – Binet in France who had a 5-point scale designed to help kids be designed appropriate educational resources. If you are a 1 or a 2, then you need extra help because you’re not that smart. If you get a 4 or a 5, then you get extra help because you’re smarter than average.

Terman put this on a 100-point scale where 100 is average. He probably is the one who came up with the ratio IQ. If the kid is 10, but scores on an IQ test like the average 12-year-old, then it is 12/10 for an IQ of 120 for the kid.  Then largely in America, you had a small intelligence testing industry grow from there.

With the heyday of IQ testing probably being in the 50s and the 60s, people really believed in it. Kids get tested now, as part of school. In the 50s and 60s, kids reallygot tested. People entirely believed in the results of those tests. Now, they seem old-fashioned and superfluous. People have the same objections to IQ as aptitude and achievement testing, which is part of college admissions in America, e.g., SAT and ACT. 

People are skeptical of those. They should be. They say, “This doesn’t help us differentiate between the rest of the student’s application. It doesn’t add anything to an application. A kid who scores 1420 on the SAT is no more likely to be a successful and good addition to your student body, then a kid who scores 1120, 1390, or 1510. SAT scores are not predictive. If you want more on this, I say this all the time with Lance [Ed. Richlin from “Lance vs. Rick“], “Just Google it! Read about it.” 

Limits on mainstream standardized tests are for efficiency in two ways. One is on group-administered IQ tests. You don’t go below 50 or above 150. It goes to Binet’s original point of IQ tests. At institutions, people will have to address the kid’s needs and behaviours, regardless. It doesn’t necessarily help to know whether the kid has an IQ of 55 or 45. 

At the extreme limits, or even within the normal range, there may not be differences that can be pinpointed within 5 or 10 points. I used to work with developmentally disabled people. Every kid with a low IQ is unique. You have to treat every kid as a kid, not as an IQ score. It takes work to differentiate between a 140 and 170 IQ. 

The IQ testing industry was intended to differentiate at ultra-high levels. Because if you have a kid with an IQ of 140, then you have a smart kid. You have enough information to give this kid enough study materials outside of this kid’s study level. You see how the kid does on the study materials. It is a waste of effort to turn this into a sport, where people are competing to be Mr. 180.

The best test constructors – Hoeflin, anybody who tries to norm the tests. That is, figure out where the test performance stacks up to test-takers’ other self-reported IQ scores with a fairly large sample. Those tests are, I think, no worse, no less accurate, in their ranges, as long as you limit the ranges to below what a perfect score will get you. 

Because every test blows up with 0 wrong or 1 wrong. It is hard to tell where you are at that point. If you are wondering what a score of 37 or 41 on the original Mega Test might equal in terms of IQ, those scores are no less accurate than a score from taking the group-administered test in a classroom in 3rd grade. They’re fine. They have a plus or minus of 8 points. 

As long as people put adequate effort into those tests, which, in itself, is hard to put the adequate effort in because adequate effort on tests like the Mega is dozens of hours, if people put adequate effort in from test to test, the scores are bound to be consistent. 

When I started taking the tests, I racked up scores from the 160s to the 190s, which is a big range. Also, some of the tests were sloppily normed. I was always looking for tests that were slutty to give me the highest possible scores. I didn’t put in the effort on some of the tests. If you look at the range of some of my scores, I have a range of 25 or 30 points.

Part of this is me. Part of this is the kinkiness of various tests. You might see a smaller range if you see someone who averages 120 and then give them a dozen different tests. they may show scores from 105 to 135, across the various tests. I don’t know if it was determined whether the Mega or the Titan had a higher ceiling. 

As I said, it is hard to determine if it is possible to determine. I think Hoeflin, himself, would say, ‘The Titan is harder.’ I would say, “It is harder.” It is paradoxical. If you have taken the Mega and done a really good job, and worked the problems, it gives a skill-set that makes the Titan easier. Because you have already done the Mega and know how Ron thinks. 

If you took a bunch of really smart people and gave half of them the Mega and half of them the Titan, people would probably find the Mega easier. The Titan has been called the hardest test ever. I would argue it is the highest rigorous test ever made. Cooijmans has come up with a bunch of really good, really challenging tests. 

I would say that his problems need more leaps of faith. When you’ve got the correct answer on a Hoeflin problem, you know it. It is still pretty true about Cooijmans’s problems. But they are more idiosyncratic, have more personality. You may not be as confident in your answers. It makes them somewhat harder.

The hardness comes from a not exactly poetic and not exactly not poetic kind of freehandedness in the associations, patterns that you’re trying to find. That mirrors the world, though, where one indicator of intelligence is picking out the faint signal, the nebulous relationships. They are so faint among the noise. 

You could call Cooijmans’s problems noisier. The signal that you’re trying to pull out will not provide as spiky a spike as a Hoeflin signal. 

Jacobsen: As an interjection for the record, did you get a perfect score on the Titan Test on the first attempt? 

Rosner: I got in an article in the Wall Street Journal for the perfect score on the Titan. Nothing really about the Titan, specifically. I almost got on T.V. because I scored really high on the Mega, but I fucked it up. I made the guest booker nervous. She cancelled me because I sounded like a lunatic. I thought you supposed to be a lunatic.

I thought you were supposed to be interesting. It was supposed to be a news show. I worked in bars. It was in the morning. I didn’t wake up in the fucking morning. I didn’t know it was supposed to be happy and soothing, and not some fucking lunatic in the morning. 

I have done like 40 tests. None lately, I don’t even know if my brain works anymore. I have been sedated, general anesthetic, like two and a half times in the past year. All of the way out of propofol, which killed Michael Jackson. I was in a twilight sleep when they gave me the once in five years colonoscopy. Unless, you elect to be put all the way out.

Anyhow, I was in twilight sleep, which is sedated and still conscious. You are sedated and not supposed to still remember it. The last time, it was fine to be not asleep. It is watching a camera go up your butt. I was proud of myself. I did not see a bunch of flakes of poop floating around.

The tests have personal meaning to me. I felt like a loser until I started getting kickass scores on these tests. It is not justified because I am not an idiot. I know the tests still don’t mean that much. I couldn’t get a girlfriend. I was bad at P.E. My orientation was: if I couldn’t get a girlfriend, then I was shitty at stuff. A girlfriend was what I really wanted. I couldn’t get a girlfriend.

To me, it was a general indicator of my suckiness. I was proud of some of the stuff that I had done. I felt this overarching suckiness because I couldn’t hook up.

Appendix I: Footnotes

[1] Erik Haereid: “About my writing: Most of my journalistic work I did in the pre-Internet-period (80s, 90s), and the articles I have saved are, at best, aged in a box somewhere in the cellar. Maybe I can find some of it, but I don’t think that’s that interesting.

Most of my written work, including crime short stories in A-Magasinet (Aftenposten (one of the main newspapers in Norway, as Nettavisen is)), a second place (runner up) in a nationwide writing contest in 1985 arranged by Aftenposten, and several articles in different newspapers, magazines and so on in the 1980s and early 1990s, is not published online, as far as I can see. This was a decade and less before the Internet, so a lot of this is only on paper.

From the last decade, where I used more time doing other stuff than writing, for instance work, to mention is my book from 2011, the IQ-blog and some other stuff I don’t think is interesting here.

I keep my personal interests quite private. To you, I can mention that I play golf, read a lot, like debating, and 30-40 years and even more kilos ago I was quite sporty, and competed in cross country skiing among other things (I did my military duty in His Majesty The King’s Guard (Drilltroppen)). I have been asked from a couple in the high IQ societies, if I know Magnus Carlsen. The answer is no, I don’t :)”

Haereid has interviewed In-Sight: Independent Interview-Based Journal Advisory Board Member Dr. Evangelos Katsioulis, some select articles include topics on AI in What will happen when the ASI (Artificial superintelligence) evolves; Utopia or Dystopia? (Norwegian), on IQ-measures in 180 i IQ kan være det samme som 150, and on the Norwegian pension system (Norwegian). His book on the winner/loser-society model based on social psychology published in 2011 (Nasjonalbiblioteket), which does have a summary review here.

Erik lives in Larkollen, Norway. He was born in Oslo, Norway, in 1963. He speaks Danish, English, and Norwegian. He is Actuary, Author, Consultant, Entrepreneur, and Statistician. He is the owner of, chairman of, and consultant at Nordic Insurance Administration.

He was the Academic Director (1998-2000) of insurance at the BI Norwegian Business School (1998-2000) in Sandvika, Baerum, Manager (1997-1998) of business insurance, life insurance, and pensions and formerly Actuary (1996-1997) at Nordea in Oslo Area, Norway, a self-employed Actuary Consultant (1996-1997), an Insurance Broker (1995-1996) at Assurance Centeret, Actuary (1991-1995) at Alfa Livsforsikring, novice Actuary (1987-1990) at UNI Forsikring, and a Journalist at Norsk Pressedivisjon.

He earned an M.Sc. in Statistics and Actuarial Sciences from 1990-1991 and a Bachelor’s degree from 1984 to 1986/87 from the University of Oslo. He did some environmental volunteerism with Norges Naturvernforbund (Norwegian Society for the Conservation of Nature), where he was an activist, freelance journalist and arranged ‘Sykkeldagen i Oslo’ twice (1989 and 1990) as well as environmental issues lectures.

He has industry experience in accounting, insurance, and insurance as a broker. He writes in his IQ-blog the online newspaper Nettavisen. He has personal interests in history, philosophy, reading, social psychology, and writing.

He is a member of many high-IQ societies including 4G, Catholiq, Civiq, ELITE, GenerIQ, Glia, Grand, HELLIQ, HRIQ, Intruellect, ISI-S, ISPE, KSTHIQ, MENSA, MilenijaNOUS, OLYMPIQ, Real, sPIqr, STHIQ, Tetra, This, Ultima, VeNuS, and WGD.

Rick G. Rosner: “According to some semi-reputable sources gathered in a listing here, Rick G. Rosner may have among America’s, North America’s, and the world’s highest measured IQs at or above 190 (S.D. 15)/196 (S.D. 16) based on several high range test performances created by Christopher Harding, Jason Betts, Paul Cooijmans, and Ronald Hoeflin. He earned 12 years of college credit in less than a year and graduated with the equivalent of 8 majors. He has received 8 Writers Guild Awards and Emmy nominations, and was titled 2013 North American Genius of the Year by The World Genius Directory with the main “Genius” listing here.

He has written for Remote Control, Crank Yankers, The Man Show, The Emmys, The Grammys, and Jimmy Kimmel Live!. He worked as a bouncer, a nude art model, a roller-skating waiter, and a stripper. In a television commercial, Domino’s Pizza named him the “World’s Smartest Man.” The commercial was taken off the air after Subway sandwiches issued a cease-and-desist. He was named “Best Bouncer” in the Denver Area, Colorado, by Westwood Magazine.

Rosner spent much of the late Disco Era as an undercover high school student. In addition, he spent 25 years as a bar bouncer and American fake ID-catcher, and 25+ years as a stripper, and nearly 30 years as a writer for more than 2,500 hours of network television. Errol Morris featured Rosner in the interview series entitled First Person, where some of this history was covered by Morris. He came in second, or lost, on Jeopardy!, sued Who Wants to Be a Millionaire? over a flawed question and lost the lawsuit. He won one game and lost one game on Are You Smarter Than a Drunk Person? (He was drunk). Finally, he spent 37+ years working on a time-invariant variation of the Big Bang Theory.

Currently, Rosner sits tweeting in a bathrobe (winter) or a towel (summer). He lives in Los Angeles, California with his wife, dog, and goldfish. He and his wife have a daughter. You can send him money or questions at LanceVersusRick@Gmail.Com, or a direct message via Twitter, or find him on LinkedIn, or see him on YouTube.”

[2] Individual Publication Date: March 1, 2020: http://www.in-sightjournal.com/haereid-rosner-six; Full Issue Publication Date: May 1, 2020: https://in-sightjournal.com/insight-issues/.

*High range testing (HRT) should be taken with honest skepticism grounded in the limited empirical development of the field at present, even in spite of honest and sincere efforts. If a higher general intelligence score, then the greater the variability in, and margin of error in, the general intelligence scores because of the greater rarity in the population.

License

In-Sight Publishing by Scott Douglas Jacobsen is licensed under a Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International License. Based on a work at www.in-sightpublishing.com.

Copyright

© Scott Douglas Jacobsen and In-Sight Publishing 2012-Present. Unauthorized use and/or duplication of this material without express and written permission from this site’s author and/or owner is strictly prohibited. Excerpts and links may be used, provided that full and clear credit is given to Scott Douglas Jacobsen and In-Sight Publishing with appropriate and specific direction to the original content. All interviewees and authors co-copyright their material and may disseminate for their independent purposes.