Int = IQ?

[Jolly Giant]

A 3 int creature is able to speak, and I think you won't find speaking human below 60 IQ. I think a dog may have 25-30 IQ.

http://upload.wikimedia.org/wikipedia/commons/thumb/f/f7/IQ_curve.svg/600px-IQ_curve.svg.png

An average dog has roughly the same IQ as a 3 year old human child. If we assume Int 3 is the minimum needed for speech, dogs should have Int 3 too; since 3 year olds can usually speak more or less coherently (if not perfectly). Assuming a dog has an IQ of 25-30 (as you're guessing, and it sounds about right to me), the whole Int*3 = IQ thing works out pretty good, actually.

On a side note, they say an IQ of 20 is just enough to figure out how to open a door...

 

[S'mon]

On a side note, they say an IQ of 20 is just enough to figure out how to open a door...

Hm, my cats have no trouble figuring that out!

 

[Galethorn]

An average dog has roughly the same IQ as a 3 year old human child. If we assume Int 3 is the minimum needed for speech, dogs should have Int 3 too; since 3 year olds can usually speak more or less coherently (if not perfectly). Assuming a dog has an IQ of 25-30 (as you're guessing, and it sounds about right to me), the whole Int*3 = IQ thing works out pretty good, actually.

On a side note, they say an IQ of 20 is just enough to figure out how to open a door...

One problem I should note is that IQ is adjusted by age; an average three-year-old therefore has a 100 IQ, even though they're sure not as smart as a twenty-year-old with even a 90 IQ.

 

[Delta]

I posit the following: The D&D fantasy world has a greater variation in the population's intelligence than in our world. Therefore, matching bell-curve areas is unnecessary for the Int --> IQ conversion.

Alternatively, you could take into account the fact that most NPC's don't have 3d6 rolled for them, they're just assigned Int 10.

D&D FAQ, first question:

Ten points of IQ per point of Intelligence is a good rule of thumb...

An essay on the subject: http://www.superdan.net/dndmisc/int_iq.html

 

[orsal]

I think that was me; roughly mapping a 3d6 bell curve INT to IQ distribution.

I use that standard too -- leaving aside that Int, as defined for D&D purposes, is not exactly the same as IQ, defined as "what an IQ test measures". For present purposes, just considering what the appropriate scale for conversion is:

IQ is (theoretically) normally distributed with mean 100 and standard deviation 15.
3d6 has a distribution with mean 10.5 and standard deviation 1.7. So:

IQ scale    3d6 scale    percentile
55        5.4        0.15
70        7.1        2.2
85        8.8        16
100        10.5        50
115        12.2        84    (aside: this is about average for PCs)
130        13.9        97.8
145        15.6        99.85

In other words, assuming that 3d6 describes the distribution in the general population (and I like to stick with that benchmark), an ability score of 16+ is about as unusual as an IQ of 145+

 

[mara]

While there are some political attempts to claim there's no such thing as intelligence, general intelligence or 'g' seems to be mostly about brain processing power, much as strength is a factor of muscle power. The simplest measure of your g is how fast you respond to a stimulus, eg in a "light goes off, press buzzer" test, how fast you begin to move hand towards the buzzer after a light goes off correlates closely with IQ as measured by other tests.

At which Stephen Hawking would fail miserably. There's a difference between claiming there's no such thing as intelligence and claiming there are multiple forms of intelligence that produce competence in different contexts that can't be measured with a single (biased) test.

I don't use any INT = NxIQ formula because the two measurements are abstract and unrelated. If you want to represent differences in INT, I'd look at INT's curve and list cognitive tasks that the majority of the population should be able to accomplish without problem and put those at 10, then work your way outward. INT by the rules just equates to more skill points and spell potential, which represents differences in learning abilty enough for me.

 

[Thunderfoot]

Too true, mara, too true. Einstein never did figure out how to tie his shoes, he always wore slippers. The man could unravel the fabric of existence but couldn't tie two pieces of string together.

 

[Warren Okuma]

Too true, mara, too true. Einstein never did figure out how to tie his shoes, he always wore slippers. The man could unravel the fabric of existence but couldn't tie two pieces of string together.

Yeah, well he had lots of skill levels in physics. I statted him up as com 20 (you know, so he can beat up those lethal housecats). Not like Dnd is supposed to even resemble reality.

My formula: INT x 5 + 50 = IQ

 

[Tonguez]

While there are some political attempts to claim there's no such thing as intelligence, general intelligence or 'g' seems to be mostly about brain processing power, much as strength is a factor of muscle power. The simplest measure of your g is how fast you respond to a stimulus, eg in a "light goes off, press buzzer" test, how fast you begin to move hand towards the buzzer after a light goes off correlates closely with IQ as measured by other tests.

Actually response to stimulus probably isn't a good model of intelligence since even the most basic organisms will respond to said stimulus and often at a much faster rate than any human.

What is far more important to human intelligence is the ability to relate and synthesis disparate information and recall it in a organised form. Now whilst brain processing power is a factor, so is the density of neural connections and efficiency of acquired heurustic structures, plus a whole lot of other factors which are still not fully understood. In other words we start with x, as we develop we acquire b and our various experiences impose c, f, y, z and r so g = x * b + c(fyz)/r~2 (or so claim I now)

Or more simple Different people have different ways of recalling information and this process is not fully understood, which is why IQ tests are biased

 

[S'mon]

At which Stephen Hawking would fail miserably. There's a difference between claiming there's no such thing as intelligence and claiming there are multiple forms of intelligence that produce competence in different contexts that can't be measured with a single (biased) test.

There are all sorts of different IQ tests, of course.

'no such thing as intelligence' - I meant no such thing as a single intelligence, in the way there is strength or dexterity. Claims that there are multiple intelligences which do not in any way correlate fit this. Where intelligence-related abilities do correlate, that indicates a single underlying intelligence, or 'g'.
Of course you can argue there are multiple different sorts of strength or dexterity likewise, but they do correlate - with muscle mass, with hand-eye co-ordination.