Rick Rosner on His IQ Test Journey

Author(s): Scott Douglas Jacobsen

Publication (Outlet/Website): Phenomenon

Publication Date (yyyy/mm/dd): 2024/08/16

Scott Douglas Jacobsen: So this is for the World Intelligence Network audience. What was your first discovery of something like an IQ test? And then, what was your discovery of an actual IQ test?

Rick Rosner: Here is my history with IQ tests. I taught myself to read when I was three and three-quarters years old. I read whenever I could. Then, in kindergarten, we were given a test that included the Draw-A-Man test. It was an IQ test. I had no trouble since I’d already been reading a lot. I did well on the test. My mom had drawn with me. I did not think she would make me perform well on an IQ test, but it came in handy for the last page of the test, which was to draw a man. I didn’t draw a man. I drew a stoplight and put him in a shirt with a collar. Then the test results came back, and in a parent-teacher conference, the teacher told my parents I was a genius.

In 1965 or 66, this didn’t mean little in terms of specially added enrichment. In first grade, they tested me again. I scored about 140 on the WISC. But then they observed me on the playground and saw that I was terrible and had no friends. They decided against skipping a grade because that would make me even more socially isolated. My school, University Hill Elementary on Broadway in Boulder, Colorado, is across the street from the University of Colorado, where the major with the most students is psychology.

So, students at CU who were psych majors or grad students in psych and had developed their own IQ tests for a grad project needed kids to test them on. Occasionally, a couple of times, I would get pulled out. They pulled out the kids who liked and did well on IQ tests to take these experimental IQ tests. Also, in first grade, I started chanting to God and turning clockwise, spinning clockwise in a circle because my parents — my dad was in the ladies’ ready-to-wear business. In 1966, he had to fly to New York five times a year. He’d pick out clothing from the showrooms to sell in our store. Five times a year, he’d fly to New York. Once or twice a year, he’d take my mom with him. So they left me with this old, scary babysitter for a week, and I freaked out. When they came home, I was spinning and chanting, which got me sent to a child psychologist.

Who gave me another IQ test, the one with the red and white blocks? Is that the WAIS? No, that’s the adult version. It could be the Stanford-Binet. Anyway, the one with the blocks. By age eight, I’d already taken many IQ tests. Back then, people had a lot more invested in and respect for IQ tests. We probably got something equivalent to an IQ score once every three years. In my file, there were quite a few IQ test scores.

In ninth grade, I was in honours math, and we were each assigned a topic at Baseline Junior High. It turned out that they gave the kid with the highest IQ in the honours class statistics because that kid would be shown the IQs of every kid in the ninth grade, which is incredible. But that’s what they did back then. The kid that got to teach the class statistics would teach the class statistics based on doing a T-test to see whether there was a statistically significant difference between the IQs of the kids in honours math and the IQs of the entire ninth grade. So I saw my IQ, the highest in the ninth grade. It was 151, which, at the time, was pretty good.

Then, in high school, I underperformed from time to time because I was primarily concerned with how to get a girlfriend and maybe lose my virginity before I graduated high school. I could have done better at this. I started applying to Harvard and decided to do a statistical analysis. I learned what a 151 IQ corresponds to. That’s about three standard deviations above the mean, a little more than that. It equals about one kid in a thousand who should be able to score that high in practical terms.

It’s more than that because I’d taken over a dozen IQ tests by that point, mainly if you include the SAT and the PSAT, which can be converted into IQ scores. When you take a dozen IQ tests, you’re only going to tell people — if you are dumb enough to say to people your IQ — you’re only going to tell them your highest score. So, by luck, somebody with a 145 IQ taking a dozen IQ tests would probably score 151 on one of the tests. So, my 151, maybe one kid in 700 or 600, had a score like that. I realized that every kid at Harvard would be the smartest kid with the highest SAT scores, which I also had in their class back in high school. I’d be completely ordinary compared to everybody else, which is terrible because I wanted to get a girlfriend.

If I couldn’t get a girlfriend at Boulder High, how was I going to get a girlfriend in college at Harvard, where everybody’s exceptional, and I’m one more schmuck competing with the Kennedy family and people who went to prep schools and expensive private schools? So, I freaked out and went back to high school to try to redo my senior year correctly. Also, I wanted to change the world with a theory of the universe, and I didn’t think an IQ of 151, given its frequency among humans, which I calculated to be one in 600 or one in 1,000, was high enough. I thought I wasn’t smart enough based on my IQ to do what I wanted. So, I thought I would learn to live as a meathead, as a physical being instead of a mental being. I’d been lifting weights. By the time I got to college, I had blown out of Harvard. I never completed the application. I went to my hometown school, the University of Colorado, lifted many weights, and eventually became a bouncer and a stripper, and lost my virginity in the way people do, where being muscly didn’t hurt.

A couple of years into college, or a couple of semesters into college, somebody told me about the World’s Hardest IQ Test, which had been published in 1979 in either Omni or Games Magazine. This was a test, the Langdon Adult Intelligence Scale, written by Kevin Langdon. I tried taking that test, and I scored 170. I found out a couple of things. One is that I could break the 151 score. Two is that the tests I’d scored a 151 on only went up to 151. That made me think that I could be as bright as I wanted. Then, in 1985, Omni Magazine published the World’s Hardest IQ Test, the Mega Test, written by Ron Hoeflin. Four thousand people took the test published in Omni, and I tied for second with a couple of other guys among the 4,000 people. I liked it. I liked scoring that high. Whenever I could spare the time to find another test with a high ceiling, I would take a shot at the test. So, I took Ron Hoeflin’s Titan Test and got the only perfect score.

I took a bunch of tests by Paul Cooijmans. He has freaking challenging tests. He likes to bust people’s bubbles — people who think they’re smart based on their other test scores. Then they try his tests, and he says, nope, sorry, you’re not as bright as you thought. I’ve taken nearly 40 ultra-high IQ tests and gotten the highest score ever on maybe 28. I haven’t done it lately. I’m 64. So there’s a couple of things going on. One is that I shouldn’t waste my time because these ultra-high IQ tests take 150 to 180 hours to do a good job. With time ticking away, I shouldn’t waste my time on that. It’s tough to find a test with a high enough ceiling. There’s no point in me taking a test; my highest score on any test with a standard deviation of 16 is in the 190s. I haven’t looked at my scores lately. I need a test with a ceiling of at least 210 to break my record score, so I can miss one or two and maybe score in the 190s. On a test with a ceiling of 210, if it’s appropriately normed — and I don’t even know if you can adequately norm a test with that ceiling — you won’t be able to get a perfect score. The test is going to be full of idiosyncrasies. Taking an IQ test is, to a certain extent, profiling the test maker. And you won’t find your way through the labyrinth of that test maker’s mind and get a perfect score. For a few years, off and on, I was working on a Cooijmans test. I could break my record but didn’t crack enough of the problems. So, it remains a work in progress and a test I haven’t returned to in years. Also, like I said, I’m 64, and there’s a chance that my abilities have also declined. I’m unsure if that is even the ability for that test and the scores I want. That’s the thing — you need a decently high IQ, but you also need to be obsessed with it and willing to put in the time and explore all these different possible angles on the problems. It would help if you were willing to burn up much time. So it might be that if I were willing to burn up enough time, I could have a shot at breaking my record, but at this point, I don’t intend to do that.

Jacobsen: What do you consider the single most challenging test you’ve ever taken by a test maker?

Rosner: The Titan by Ron Hoeflin is the most significant, most challenging, ultra-high IQ test ever written. Though Cooijmans’s tests have ceilings in the Hoeflin range, Hoeflin’s test is crisper, and the answers snap more satisfyingly. But you can’t take the test anymore. Nobody accepts the test for admission to ultra-high IQ societies anymore. The reason is that these tests were written before the Internet, and they have become much easier with the Internet.

You can be told not to cheat but to plug in the test items. Each test, the Titan and the Mega, consists of 24 verbal problems, all analogies, and 24 math problems. These days, you can type in the three words given to you to solve the analogy, and the fourth word will probably pop up for half of the items. Also, people have been discussing Hoeflin tests on the Internet for as long as the Internet’s been around, which is, for most people, about 30 years. If you root around, you could find at least 30 answers out of the 48 test problems without exercising your thought. And I believe 30 out of 48 on that test gets you over 150. Also, taking the Mega has a practice effect for the Titan. So yes, the Titan is a stricter test than the Mega, but if you take it first, it gives you enough of a feel for Hoeflin’s test construction, effectively making the Titan easier. But if you’re going to take a test cold and without recourse to the Internet, the Titan is the most significant ultra-high IQ test ever, undoubtedly the greatest from the pre-internet era.

But I doubt it because the ways of looking at the way people think or look at their PET scans have the stink of phrenology of pseudoscience. So does IQ, by the way. But you can do all the Googling you want. You will still have to follow a twisted path of logic and intuition. He says one of the components of IQ, besides conscientiousness, is — that I always forget. It’s like associative breadth. What’s his term for it?

Jacobsen: Width of associative horizon.

Rosner: Okay, so the width of associative horizon means how much stuff you know: How much stuff can you learn? How much stuff can you pull in from different angles and disciplines? How many different angles can you come up with in your attempt to solve a problem? Will you stop with a half-assed answer after trying three different angles? Or are you going to take 40 hours or 60 hours thinking about the test and come up with 30 or 40 possible ways that you initially check to see if there’s a solution for that item in your new angle? Are your 40 different ways to solve the problem taken from all sorts of disciplines and areas of knowledge that could be anything that the test creator had access to, either in his mind or in his environment, when constructing the test? If you don’t come up with dozens of possible ways to solve Cooijmans’s problems, you probably won’t arrive at the correct answer. You probably won’t find the correct answer to his problems, even if you try dozens of angles.

What was the question I was trying to answer? Oh, Cooijmans’s tests are internet-resistant. He fully expects you to consult, and he doesn’t like when people talk about his tests. Because he considers it to be in the realm of giving hints. So I’m trying not to give hints, but look at the tests. Look at his tests; you can see that they will yield slowly. I don’t think I’ve given anything away by saying you’re going to have to bust your ass.

Jacobsen: Do you pay much attention to the high-IQ communities anymore?

Rosner: Not a whole lot. I’m friends with you, and you have a history of interviewing every high-IQ person who would say “Yes” to an interview. You’ve interviewed the heads of a ton of high-IQ societies. I’m friends with people from the Mega Society, particularly one guy from the Mega Society. I’m friendly with more people from there, but not a lot. I don’t have that much contact. I don’t have many friends in general. I tend not to cultivate friendships or maintain them sufficiently.

Jacobsen: What would you do if you had to make an IQ test practical now?

Rosner: It’s tough to say. It’s also a fool’s errand at this point, in that AI will, if it doesn’t make us smarter now — and it doesn’t because AI is only as intelligent as the material that went into it — but it certainly makes people more efficient by doing some of the people’s work. Eventually, it will be powerful enough to make people smarter. So why are we messing around with measuring people’s IQs at this point? Yes, you could give a kid an IQ test to see if they need specialized individual attention in school, but there’s something that will feel increasingly archaic. It seems like a sport that’s only followed and played by a few people, like the World’s Strongest Man — guys who can lift boulders two and a half feet in diameter or pull a truck with their teeth. People do that, and they compete in that, but the ultra-high IQ thing seems a little bit quaint.

But if I were to build a test now, it would focus more on how we live now. It would be about how well you use technology to solve challenging problems. I would design a test to do so, where all technical assists are permitted and encouraged. Google, AI, and other stuff are out there, but the items on the test are so brutal that even with tech assistance, most people will need help to solve them. That’s great in theory, but good luck coming up with those problems. There’s also the issue that if you make a test that’s at all popular, people will discuss it online and eventually contaminate the matters unless there’s a way to give people varied problems. But even then, you’re going to contaminate the test. Eventually, people will discuss the issues and get out there. So, you may come up with a test of intelligence that is not a traditional intelligence test.

When I was a kid, I remember they had a guy on either the Mike Douglas or the Merv Griffin show who said he had a machine that could measure your IQ based on how still you could keep your eyes. If you could stare at a point without your focus jiggling, or the more you could keep your eyes still, the higher he said your IQ was. And then he brought out a lady who he said was a super genius because she could keep her eyes still. That seems like A, primitive, and B; yes, maybe it’s correlated with intelligence, but it’s not what you’d want, especially 46 years later, 56 years later.

And yes, that’s a criticism of IQ: it isn’t a measure of intelligence going out and writing a great novel, a great screenplay, or starting Microsoft. You could give people the Turing test, where you sit and talk with them. Let’s say there’s a kid who people suspect might need individualized attention in school because they think he might be a genius. So, give the kid a regular IQ test; if the kid breaks 140, then yes, the kid, or 130 or whatever, could use enriched materials. If the kid seems bored, then the kid could use some enriching materials. And then, if you’re trying to figure out the difference between a kid with a 140 IQ and a 160 IQ for whatever reason you might have — maybe you run the Davidson Institute, and you’ve only got a certain number of slots for hyper-gifted kids — you could do a Turing test where you sit down and talk with the kid — three sessions of a half-hour each. You have somebody who is familiar with ultra-high IQ kids and what they can be like, and you have that person be the one who talks with the kid.

Then, the person estimates the kid’s IQ at the end of the sessions. If the kid can keep up the appearance of having an IQ of 165 when talking with somebody who knows what people with 165 are like, then it’s a Turing test. If you seem like a 165 IQ kid, then you are. Is that reasonable? Is that a decent strategy? Conversation can tell you a lot about a person. Yes.

Jacobsen: What about IQ in and of itself? This could be the last question or theme for this interview. What is the fundamental distinction between what IQ is measuring and what intelligence is?

Rosner: Okay, so intelligence — we talk about g, which is general intelligence. Is that right, G? Is it g, or do I have the initial wrong?

Jacobsen: Yes, g is right.

Rosner: Okay, so some assumptions go into the idea of g, and g is the idea that intelligence is a thing that people can have that lets them solve problems of any type that require thinking. Not the problem of how I am going to walk this boulder. Walking a 2.5-foot boulder across the room is a problem in strength rather than IQ. But the idea of g is that the world is full of challenges to thinking of different levels of difficulty and that somebody with higher g, higher intelligence, will be able to solve more challenging mental problems than somebody with lower g. I mentioned some of the assumptions that go into that.

Then, when you go to IQ, there’s the assumption that solving the problems presented to a test taker on an IQ test reflects their ability to solve problems in general. There are all sorts of ways for error to sneak into that. For one thing, you’re supposed to take the test cold like the Raven’s Progressive Matrices. You’re supposed to have never seen the Raven, the tic-tac-toe framework. There are nine squares, three by three. Eight of them are filled. You figure out what goes in the ninth square based on the logic of the relationship of the symbols in the other eight squares. That thing has a vast practice effect.

If you take a kid and have them do 10 Raven practice items before getting tested on the Raven, that kid will score at least one standard deviation higher than they would go in cold. And these days, there are plenty of opportunities on IQ-type tests to stay calm. The SAT is an IQ-like test. For most of the history of the SAT, you were supposed to take it cold, where your only prep for it was the PSAT. And the SAT people, ETS, Educational Testing Service, used to say, “Well, you can’t prep for the SAT. We’ve done studies of people who’ve prepped, and it doesn’t work.” That turned out to be bullshit. The early prep programs didn’t prep people enough.

My kid hired me to prep her for the PSAT. She thought it would be cool to be a national merit scholar, national merit semi-finalist, whatever it is. I dug up 80 real PSATs and SATs and made her take them. Then, we would go over what she got wrong. So before she fired me for working her too hard, she took 80 tests. That is how you prep. If you take 80 and 30 practice tests, you’ll see almost every problem they can throw at you. You’ll do fine. You’ll do better than satisfactory. My kid kicked ass. My brother did the same thing with the GMAT. To attend business school, he took 30 practice tests and raised his score by 150 points on a 200 to 800-point scale, which got him into Wharton. So, you have to prep extensively. So, there is a vast practice effect, and it’s unfair to the kids who don’t know about that.

There’s another way to game the SAT and the ACT: if you have the money, you go to an educational psychologist and pay him 500 bucks to be diagnosed with some learning difficulty that requires relaxed conditions for the SAT or the ACT. Extra time, somebody reading the items to you might be a thing. I don’t know. There are eight or ten different flavours of help on the SAT. What is the SAT? What are six or seven subtests? Depending on your diagnosis, you can not only get double time but only have to take two subtests a day. And so the kids who get more generous testing conditions often do better. So yeah, there are so many ways to game the system.

Then, there are historical problems with IQ, such as that it’s not culture-fair. Culture-fair means that if you took ten people from 10 different parts of the world and gave them the test cold, none of them would suffer on the test from stuff they didn’t know based on being from the southern tip of South America. The first IQ test, devised by Terman and given to a ton of American soldiers going off to fight World War I, had items like they would show you a drawing of a hood ornament and say, “What make of car has this hood ornament?” You can’t get any less culture-fair than that. And it took a long time for people to admit that stuff wasn’t fair.

So, many assumptions go into the idea of intelligence, the idea that intelligence can be measured and that IQ tests are the way to measure it. Then, you can end with a quote from Churchill. I’ll look it up, and then we’ll come back. Democracy is the worst form of government until you look at every other form. IQ is a lousy way to measure intelligence once you look at every other possible way.

The end.

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.