Intelligence = IQ?

[Paragon Badger]

So an intelligence score of 10 represents complete average-ness, according to the PHB.

So is that 100 IQ? Is an intelligence score of 20 equal to 200 IQ?

Discuss. :smallbiggrin:

[Sir Jason]

What!? According to who? 'Gifted' is 130+. But there have been people with IQ close to 200.

Anyway, my way of seeing it is [10 + INT modifier] * 10, approx.

That gives INT 10 IQ 100 and INT 18 IQ 140, which is about right. It also give a good idea as to how much smarter a gold dragon is then, say, a commoner.

[Goff]

The *10 rule is normally how I run things, but this whole thing has been debated countless times on this board in my years here... Personally I don't think it's really neccesary to have more than a rough guide.

[martyboy74]

What!? According to who? 'Gifted' is 130+. But there have been people with IQ close to 200.

Anyway, my way of seeing it is [10 + INT modifier] * 10, approx.

That gives INT 10 IQ 100 and INT 18 IQ 140, which is about right. It also give a good idea as to how much smarter a gold dragon is then, say, a commoner.

On the other hand, this means that somewhat who has a 200 IQ (1:5,000,000, IIRC), would have a 30 INT. That seems kind of unreasonable. This also puts a 0 INT as a 50 IQ. :smallamused:

[sikyon]

The best way would probably be to do it with statistical distribution of 4d6 drop the lowest, correlating to the corresponding distribution of IQ's. I'm too lazy to do it, however, and I think someone's done it before.

[martyboy74]

The best way would probably be to do it with statistical distribution of 4d6 drop the lowest, correlating to the corresponding distribution of IQ's. I'm too lazy to do it, however, and I think someone's done it before.

That's the PC stat generation roll. It gives an average of somewhere around 12.445. NPCs (who are average) use 3d6, which gives an average of 10.5. That, using the *10 rule, creates a IQ of 105, which nicely accounts for the Flynn effect.

[Spiryt]

On the other hand, this means that somewhat who has a 200 IQ (which is actually about 1 in a million), would have a 30 INT. That seems kind of unreasonable. This also puts a 0 INT as a 50 IQ. :smallamused:

Yes, but 30 Int isn't easy too achieve, even with serious magic boosters.

And about 0= 50 IQ - this is a reason why we shouldn't take it too serious - what "knowing" the IQ of your character gives you anyway?

Besides 50IQ is still described as "severe Mental Retardation" if I remember.

[Tengu]

Intelligence improves over time (you can improve it by leveling and aging). Show me a case when IQ did that.

But if I really had to make a connection between those two, I'd go for IQ = 50 + Int*5. Certainly not IQ = Int*10.

[Shhalahr Windrider]

Besides 50IQ is still described as "severe Mental Retardation" if I remember.
And 0 Int is described as comatose due to complete lack of brain function. There's a difference.


But if I really had to make a connection between those two, I'd go for IQ = 50 + Int*5. Certainly not IQ = Int*10.
Do most nonhuman animals have the equivalent of a 60 IQ?

[Douglas]

If you must equate the two numbers, it should be done by matching their distributions. I have no idea how many times I've done the math or seen the math done, but comparing standard deviations gives an equivalency of 1 point of intelligence per 5 points of IQ, with the average at 10.5 int = 100 IQ.


Do most nonhuman animals have the equivalent of a 60 IQ?
It's probably not too far off, I'd think. That is well into the realms of mental retardation for a human, after all, and animals aren't really all that stupid about things that aren't simply beyond their comprehension.

[Tengu]

Do most nonhuman animals have the equivalent of a 60 IQ?

If they had language that humans can understand, we could probably perceive them as such.

[Querzis]

Intelligence improves over time (you can improve it by leveling and aging). Show me a case when IQ did that.

It doesnt improve by aging at all, it does improve by leveling but hey, in RL, when people become strongest warriors they dont have more 'hit points' either. A sword in your chest is still a sword in your chest so personnaly I consider that that D&D people are 'normal' when they are level 1 but after that things get really unlogical. Anyway, even if D&D intelligence can improve with levels and magic items, the definition of D&D intelligence match almost exactly our definition of IQ.

But even we dont see it often, people in RL can have an IQ of 200 or even more while nobody in D&D can have an intelligence of 30 without magic items lol. So as far as I'm concerned, just add a 0 at the end of your intelligence score and you have your IQ. Sure that means some magical creature could have an IQ of 350 but thats the point, its magic. I dont see the problem really.

[Droodle]

IQ would actually measure more than just one's intelligence score. Wisdom needs to be taken into account as well. On top of that, education is also a factor. Contrary to popular belief, studying for the IQ test will actually improve your score. Given that you can improve your IQ score by studying, the IQ test is clearly not representative of your actual intelligence score, but rather a composite of your reasoning skills, your education, and the level of experience you have dealing with the types of problems presented on an IQ test.

[Ceres]

Do most nonhuman animals have the equivalent of a 60 IQ?

I believe 20-30 or so is required for opening doors, and this level of understanding is only possessed by intelligent animals such as some dogs, cats and octopi. I really don't think you can translate intelligence into IQ with a simple formula. At least not on the very low levels.

[Querzis]

IQ would actually measure more than just one's intelligence score. Wisdom needs to be taken into account as well. On top of that, education is also a factor. Contrary to popular belief, studying for the IQ test will actually improve your score. Given that you can improve your IQ score by studying, the IQ test is clearly not representative of your actual intelligence score, but rather a composite of your reasoning skills, your education, and the level of experience you have dealing with the types of problems presented on an IQ test.

Dude D&D intelligence IS all about knowledge. Of course education is also a factor in IQ, just like it is in D&D intelligence and I really dont see how it measure Wisdom. You dont have to be cunning or crafty to take an IQ test and someone who is incredibly naive could very well have an IQ of 200. Experience has nothing to do with it either since they never tell you which problems you had wrong and the test are almost always different. So even if you actually take the same test 5 times and it will give about the same result (except if you fails some questions on purpose).

[Krellen]

But if I really had to make a connection between those two, I'd go for IQ = 50 + Int*5. Certainly not IQ = Int*10.
I have a problem with any correlation that gives me a 20+ Intelligence. A +5 bonus would make me far more skilled than I am, despite my genius-level IQ.

To be truthful, INT*10 doesn't really work either. In reality, it's far more complex than that.

[Joracy]

I believe 20-30 or so is required for opening doors, and this level of understanding is only possessed by intelligent animals such as some dogs, cats and octopi. I really don't think you can translate intelligence into IQ with a simple formula. At least not on the very low levels.

Birds can open doors. Also some animals may not be physically able to do certain things. Fish for for one example are much much much smarter then most people seem to think they are, heck I'm out smarted by fish weekly but I don't think they could easily open doors. There is nothing funnier then a perfect cast to a carp, only to have it look at the fly, Turn and look at you then slowly keep going. or having a red fish spin around the dock supports to knot line up. or being taunted by trout.

[Droodle]

Dude D&D intelligence IS all about knowledge. Sure. But knowledge != IQ.
Experience has nothing to do with it either since they never tell you which problems you had wrong and the test are almost always different. So even if you actually take the same test 5 times and it will give about the same result (except if you fails some questions on purpose).Simply untrue. The IQ test is not infallible. You actually can study for it. Sure, you won't know what questions they are going to ask, but that doesn't mean you can't practice with problems similar to the ones presented on the test....you know, the same way you study for any other test. The IQ test can give you an idea of how smart someone is. But it isn't perfect.

[mroozee]

Intelligence improves over time (you can improve it by leveling and aging). Show me a case when IQ did that.

But if I really had to make a connection between those two, I'd go for IQ = 50 + Int*5. Certainly not IQ = Int*10.

It happens if you take levels in Scholar :smalltongue:

As far as IQ scoring goes, an anecdote:
I used to teach mathematics at the university level. My combinatorics students would get mathematically based IQ problems as a pre-test and a post-test for the class. Since the methods covered in the class correspond to the types of questions asked on IQ tests, their scores naturally jumped. This is more of a commentary on the inadequacies of IQ testing than actual change in ability, of course.

As to this problem, I would use:
IQ mean = 105, IQ standard deviation = 15
3d6 Int mean = 10.5, 3d6 Int standard deviation = 2.96

Simplifying for game purposes gives: IQ = 5xInt+50 or Int = (IQ-50)/5
So if your Wizard has a 22 Int, this is a 160 IQ. If you have a 130 IQ, you would have an Int of 16. This is simple, and reasonably accurate when you are away from the tails.

[horseboy]

I think this comes from how they actually used to do that in prior editions. (Int 15=150 IQ) This edition, with as easy as it is to get stat mods, it really can't hold up any more.

[bugsysservant]

Um, just as a point of interest Da Vinci had an estimated IQ of 220. How does one achieve an Int. score of 34-35 without magic?

[Arceliar]

Averages:
Int: 10.5 (for an adult human, using 3d6 stat generation)
IQ: 100 (by definition, relative to persons one's own age, we'll assume adult)

Lack of Brain Function
Int: 0
IQ: 0 ...I think?

Maximums:
Int: 18, at level 1, without race modifiers
IQ: 180 reaches a point of near redundancy where it's essentially impossible to accurately and consistently measure.

It is also roughly the point where a person will, in at least a few fields of study, likely be more than capable of comprehending any available information in said fields. To my understanding it is more likely a person simply lacks access to new information (or a desire to learn it) than lacks the ability to comprehend something. That's not to say a higher IQ isn't significant, but the significance of such quickly becomes less obvious and, relative to a 100 IQ, almost arbitrary.

I don't know if 180 IQ and 18 Int are the same, but given that the 0-0 and 10.5-100 relationships are so close, and that 18-180 are at least both significant numbers, I'd just assume call them the same thing and be done with it, despite the fact that this causes too shallow of a bell curve distribution wise in D&D. That's more of a problem with the 3d6 method of stat generation. More importantly, I don't care. It's just a game.

EDIT: With regard to bugsysservant post, Marilyn vos Savant has an IQ of, if I recall correctly, 228. More hilariously, Stephen Hawking's is I believe somewhere just over 200. Can you say Min-Max?

[Ceres]

Stephen Hawking's is I believe somewhere just over 200. Can you say Min-Max?

Heh, yeah. God must have let him use GURPS-rules for character creation :smalltongue:

[bugsysservant]

EDIT: With regard to bugsysservant post, Marilyn vos Savant has an IQ of, if I recall correctly, 228. More hilariously, Stephen Hawking's is I believe somewhere just over 200. Can you say Min-Max?

Just a nitpick, but not according to this http://www.aceviper.net/estimated_iq_of_famous_people.php

Da Vinci tops the scales of "Estimated I.Q. of Famous People" with 220, whereas Vos Savant has a "paltry" 186. Hawking, on the other hand, has 160, for an int. of what, 22?

[Fenix_of_Doom]

Since results of IQ test can vary so stating someone's IQ is if it is a set number is rather pointless. This, at the same time, makes an int to IQ debate irrelavent because an IQ test isn't that accurate let stand a IQ estimate.

[Logic]

...when people become strongest warriors they dont have more 'hit points' either. A sword in your chest is still a sword in your chest so personnaly I consider that that D&D people are 'normal' when they are level 1 but after that things get really unlogical.
The best way to explain "Hit points" is the ability to roll with the damage, or failing that, fatigue. You can only dodge so many blows before you get tired enough that the next one becomes fatal.

[MrNexx]

Guys, there's a difference between "Rule of Thumb" and "thick, bright line."

As a rule of thumb, I think 1st level Intelligence *10 (180 for the smartest humans) is a good rule of thumb. Increases after 1st level are a matter of knowledge and technique, not raw IQ.

[Ozymandias]

I say that the two systems are both really nebulous and fairly arbitrary. Intelligence is far too complex and multifaceted to pin down to a single number; most IQ tests (and the tests are in no way uniform, again limiting their usefulness) predominantly measure symbolic logic, leaving other facets, such as creativity and mathematics, unexplored. D&D's system is similar, but it has a good excuse in that it is a game.

Trying to draw a correlation is, in my opinion, a poor idea.

[Malic]

This is what I go by.

Average= 111 I.Q.
Average=10 Int.

11=116
12=121
13=126
14=131
15=136
16=141
17=146
18=151
19=156
20=161

According to current standards an I/Q. of 160+ is a super-genuis. (ie. skip to collage when you start 8th grade)

[Spiryt]

Average in D&D is statistically 10.5 - or rather more, since statistically when someone rolls 4 for one ability (even when other are good) there is

"What?! You will allow my to reroll this, or I'm not talking to you :smallbiggrin: "

And seriously, I though that 111 is already slighty superior inteligence, at least Wiki says so (though I know that Wiki is not always best source...)

[Fenix_of_Doom]

I say that the two systems are both really nebulous and fairly arbitrary. Intelligence is far too complex and multifaceted to pin down to a single number; most IQ tests (and the tests are in no way uniform, again limiting their usefulness) predominantly measure symbolic logic, leaving other facets, such as creativity and mathematics, unexplored. D&D's system is similar, but it has a good excuse in that it is a game.

Trying to draw a correlation is, in my opinion, a poor idea.

My sentiments exactly.
Edit: Also QFT and What he said.

[bugsysservant]

And seriously, I though that 111 is already slighty superior inteligence, at least Wiki says so (though I know that Wiki is not always best source...)

You Lie! Bow before the knowlege of the Wiki God!:smallfurious:

[Jimmy Discordia]

I tend to stick with the Int * 10 = IQ rule of thumb, but for large populations it's obvious that this starts to fall apart. If we assume everyone other than PCs, who are already above-average by virtue of their being PCs, rolls 3d6 for all their ability scores, then 1 in 216 people have an IQ of 180 (or higher, if they've put bonuses into Int or have a racial Int bonus). I'm going to discount the effects of magic, because your average person doesn't have access to stat-boosting magic. The actual proportion of real-life people who have an IQ of 180 or more is much less than one in a million... I think the one in a million point is somewhere in the low 170s, and at levels that high the proportion drops off sharply with each added point. I would be surprised if the number of people in the US with IQ >= 180 is more than two hundred. Not to toot my own horn, but I have a super-genius-level intellect, and I've never tested that high. The rule of thumb with Gaussian distributions like IQ is that 99.7% of people will test between three standard deviations below the mean and three standard deviations above. No more than 0.15% of people should have IQs over 145 (since half of that 0.3% is on the low end). Needless to say, this doesn't map itself well to a distribution of rolls on 3d6.

This is all assuming a mean of 100 and a standard deviation of 15, by the way. On different scales, the proportions will obviously be different.

There's plenty of debate in real life over what IQ scores actually measure anyway... as has been stated a couple of times in this thread already, studying the types of questions you're likely to be asked on an IQ test can lead to a marked improvement in your test score. I've taken maybe a half-dozen (different) IQ tests in my life, and I've found that the mood I'm in when I'm taking the test can lead to variations of as much as 10 points from my "average" score. This isn't even getting into cultural factors... I'm not entirely up on the debate (I could call my ex-roommate, who studies that sort of thing, but that seems like a lot of work for a simple forum post) but I understand that one's cultural background can also have a significant effect on IQ scores. IQ, it seems, measures a combination of raw intelligence, general knowledge, skill at test-taking, and cultural background in common with the person who designed the test.

Now, what does the Intelligence score in D&D actually measure? Is it pure, raw intellect, or some combination of factors? The fact that one can improve it suggests that there's a knowledge component to it, but of course D&D isn't supposed to model real life - with a few months of weight training, one could conceivably add a few points to one's Strength score, especially if one started out of shape; but in D&D you can only raise your Strength score once every four levels without the aid of magic (unless your DM works out some kind of training system, but that has "rife for abuse" written all over it). Ability scores in D&D are abstractions, and you can't read too much real-life context into them.

Regarding comments that it's possible to have an IQ over 200 IRL: yes, theoretically real-life IQ has no upper limit, since it's a statistical distribution. Some higher scores will be unlikely enough that they're practically impossible, but there's never a point where the probability drops to zero on the upper end.

Wow, that was a lot of writing. Point is, feel free to stick with IQ = Int * 10 as an abstraction, but realize that this is just an abstraction, and that D&D isn't meant to model real life. Hell, a fighter with average rolls on his hit dice and no Con bonus is more than 50% likely to survive a hundred-foot fall at 6th level. One with a good Con bonus, better-than-average rolls, or a higher level could be relatively certain to survive a fall from any height. No matter how much you work out, in real life you're extremely unlikely to survive being run through with a sword. In D&D, with enough training you're almost certain to. Abstractions are fun, especially when they produce ludicrously unrealistic results.

[Aquillion]

Let's not forget... 150 IQ is not that rare (2%, I think? It's the requirement to join MENSA.) 150 is a level that normal humans can have without any sort of special adjustment, so it shouldn't be above 18. Honestly, given that 18 is the absolute max that a human can start out with with no special bonuses, I'd say that IQ = Int * 10 is actually pretty good, better than any of the others people have suggested. It gets animal intelligence about right, has MENSA-intelligence as 'exceptionally high, but not superhuman', and hits the absolute natural-human limit (180) pretty accurately.

Remember that most characters in the game world do not have stats very far from 10. In the archtypes and lower arrays, 15 is a PC's exceptional stat--saying that 15 int is MENSA-level allows for that, while still allowing a bit more for genuinely heroic individuals.

[Jimmy Discordia]

Let's not forget... 150 IQ is not that rare (2%, I think? It's the requirement to join MENSA.)

It's actually rarer than that, assuming the test is calibrated to a mean of 100 with a standard deviation of 15. The rule of thumb I learned in statistics is that in a perfect Gaussian distribution, 68% of results will be within one standard deviation of the mean, 97% will be within two, and 99.7% will be within three, so only 0.3% of humans will have IQ lower than 55 or above 145 (hence, only 0.15% will be over 145, since half of the 0.3% will be on the low end). I agree with the rest of your post, though - the best stat in the elite array being equivalent to minimum Mensa-level intelligence makes sense to me. It's kind of a big assumption to think that your 'average' person is a set of unmodified 3d6 rolls for stats anyway, since in a large population this makes exceptionally high and exceptionally low stats much more common.

[NullAshton]

The average commoner does not roll 3d6 for stats. They are given 10 10 10 11 11 11 base stats, just like every other creature in the SRD. Exceptional creatures may be given the elite array or alternate stats.

[martyboy74]

The average commoner does not roll 3d6 for stats. They are given 10 10 10 11 11 11 base stats, just like every other creature in the SRD. Exceptional creatures may be given the elite array or alternate stats.

Technically you're right, but we're not discussing the average commoner. We're using the commoners as the average for humans.

[Shhalahr Windrider]

The average commoner does not roll 3d6 for stats. They are given 10 10 10 11 11 11 base stats, just like every other creature in the SRD. Exceptional creatures may be given the elite array or alternate stats.
"Average" in this case meaning "nonunique, reusable stat block?"

The three 10s and 11s are just based on further simplifying the 3d6 roll every common NPC is assumed to have. It's a way of saying, "Yeah, they're supposed to be 3d6 each, but do you really want to roll 40 sets of scores fore all these villagers. No individual roll is really gonna make a difference anyway, mechanically speaking." If you were to use a unique but unexceptional NPC, you'd base his stats off of 3d6 or the nonelite array (or even 15 point buy).

[StickMan]

Well wizards said awhile ago that (And no its not raw) that every point of INT is about 10 points of IQ give or take a bit.