Whats your I.Q.

[MottMan]

chrisname wrote:

Also, clearly you can't test motor skills very well on paper (I guess you could test dexterousness by asking people to draw or assessing handwriting) but that wouldn't be conclusive.

I believe truly professionally recorded IQ test do look at motor skills. They often contain physical objects you must manipulate, locate patterns amongst, and other common questions that you would normally do on paper.

This is what I've heard/seen anyway.

[iw0XoG1T]

IQ test are rarely administered correctly; I have never heard of anyone who gives IQ tests being subject to interrater reliability analysis. You really cannot compare two people tested by different individuals; there is way to much variance. I have seen school districts in my area where the mean IQ is 130--anyone who stayed awake in statistics would know that whoever was giving these tests was incompetent.
In the US standardize test like the SATs are much more useful--the only problem is they really only accurately evaluate US high school students. I remember when I tried to get into a University I considered taking the SATs and was told that at my age any score would be meaningless.

[chrisname]

My maths teacher went to America for a year to teach and he often complains about it standardised testing, and based on what he said, it seems completely ridiculous.

[helios]

I just don't believe it truly quantifies intelligence

Was it ever intended to?

motor skills

That's ridiculous. Motor skills are not an aspect of intelligence.

problem solving

Computers can solve problems.

Every IQ test I've ever seen involved questions with only one answer, and they were all multiple choice, There's no space for creativity, you're either right or wrong

I remember seeing one question in a test where there were two equally valid answers, but one was harder to find. By "harder" I mean you had to look deeper into the patterns to see it.

IQ tests test, I think, pattern matching skills, which makes sense if you think about it. Someone good at finding patterns will be good at solving problems because they'll be able to quickly apply previous experience to new situations, thus they'll also be quick learners. For example, they'll need less practice to realize when a solution in a group of solutions is more appropriate to solve a given problem.

[iw0XoG1T]

The SATs are a accurate predictor of how well a US high school student will do in academia--and that is all they were ever designed to do. The problem is most people think these test measure intelligence which of course they do not.

[chrisname]

helios wrote:

Was it ever intended to?

Given that it is called "intelligence quotient", I would hazard a guess at "yes."

helios wrote:

That's ridiculous. Motor skills are not an aspect of intelligence.

Arguable. You could argue that someone whose CNS (central nervous system) is developed to allow greater control of muscle fibre activation (and hence greater strength, muscular endurance or dexterity) is in some ways more intelligent than other people. Some people are able to perform mental arithmetic very easily (I mean, it comes naturally to them, like walking or breathing) due to the way their brains work. Other people have a naturally well-developed CNS. In this way, they are more physically capable, more agile or more dexterous than the average person. I think this should be considered a form of intelligence.

Further, having superior motor neurons would allow greater control over motion, and hence better dexterity. I don't know if such a thing exists, but if some parts of the brain can be enlarged in some people (leading to different people being good at different things), then why not parts that deal with motor functions?

helios wrote:

Computers can solve problems.

And? Humans are usually better at it, unless it requires heavy number crunching and such like. Computers are bad at solving problems creatively. Creativity is of huge importance when considering intelligence, and something that IQ tests do not

helios wrote:

I remember seeing one question in a test where there were two equally valid answers, but one was harder to find. By "harder" I mean you had to look deeper into the patterns to see it.

That's one example, it hardly defeats my whole argument. Just out of interest, did you get equal points either way?

helios wrote:

IQ tests test, I think, pattern matching skills, which makes sense if you think about it

So then it should be PMAQ (Pattern Matching Ability Quotient). Pattern matching is not the only aspect of intelligence.

helios wrote:

Someone good at finding patterns will be good at solving problems because they'll be able to quickly apply previous experience to new situations, thus they'll also be quick learners. For example, they'll need less practice to realize when a solution in a group of solutions is more appropriate to solve a given problem.

All true, but again, it is only one aspect of intelligence. Something which claims to quantify intelligence (hence "intelligence quotient") should consider all manifestations of intelligence.

[helios]

Arguable.

No, it's not arguable. Where are you going to find a definition of intelligence that includes motor skills?

Other people have a naturally well-developed CNS. In this way, they are more physically capable, more agile or more dexterous than the average person. I think this should be considered a form of intelligence.

Dogs have larger portions of their brain dedicated to processing scent. So then sensory processing capabilities is an aspect of intelligence, as well? Then it's possible to say "Jimmy is really intelligent! He has great eyesight!"? Likewise, it's possible to say "George isn't very smart. He's crippled from the neck down."

Humans are usually better at it

Computers are more thorough and exact, while humans make mistakes and forget about things, so I don't know what you mean by "better". Humans are, however, capable of finding novel solutions. That doesn't make problem solving an aspect of intelligence. It does make it an activity that is helped by intelligence.

Something which claims to quantify intelligence (hence "intelligence quotient") should consider all manifestations of intelligence.

"IQ" is an old term, and what it supposedly measures has changed over time. At one point it was mental age / physical age * 100. Just because it has "intelligence" in its name doesn't mean that it tries to measure all aspects of intelligence.

[chrisname]

helios wrote:

Dogs have larger portions of their brain dedicated to processing scent. So then sensory processing capabilities is an aspect of intelligence, as well? Then it's possible to say "Jimmy is really intelligent! He has great eyesight!"? Likewise, it's possible to say "George isn't very smart. He's crippled from the neck down."

My whole argument is founded on the idea that intelligence is measured in more than one way, and you're telling me that I implied someone can be intelligent purely because they have good eyesight? What I said was, those abilities should be considered aspects of intelligence. Not the whole thing.

helios wrote:

Computers are more thorough and exact, while humans make mistakes and forget about things, so I don't know what you mean by "better".

By better, I mean more adept, or more able. Computers don't come up with solutions, they just perform them. Actually, that's not strictly true, some programs do come up with solutions to problems (parser generators for example), but in this case they have been programmed to do so; the code they generate is a direct result of the problem solving done by a human, i.e., the person that wrote the parser generator.

helios wrote:

Humans are, however, capable of finding novel solutions. That doesn't make problem solving an aspect of intelligence. It does make it an activity that is helped by intelligence.

The capability to find novel solutions is the problem solving intelligence. Problem solving is not just having a method and using it; the most important part (and the part humans excel at) is figuring out the method. That is what I'm referring to by problem solving.

helios wrote:

"IQ" is an old term, and what it supposedly measures has changed over time. At one point it was mental age / physical age * 100. Just because it has "intelligence" in its name doesn't mean that it tries to measure all aspects of intelligence.

But I'm saying it should, not that it does.

PS: Excuse me if I miss out any "d"s, my d key doesn't seem to be working properly at the moment.

[helios]

you're telling me that I implied someone can be intelligent purely because they have good eyesight?

No, that's not what I'm saying. I'm saying that considering something as intelligence (or part of it) merely because it happens in the brain is ludicrous. "Intelligence" as is generally understood refers to a quality that a vaguely defined set of processes in the brain have. For instance, if you could make two identical humans, but give one better dexterity or eyesight or sense of smell, one normally wouldn't say that that one is smarter.

The capability to find novel solutions is the problem solving intelligence.

And the capability to make delicious dishes is the cooking intelligence?
Evidently we're never going to agree on this, so never mind.

But I'm saying it should, not that it does.

Just because it has "intelligence" in the name? It also has "quotient" in the name, but it's not the result of dividing anything by anything.

[chrisname]

helios wrote:

Evidently we're never going to agree on this, so never mind.

You're right, though I do see where you're coming from.

helios wrote:

I'm saying that considering something as intelligence (or part of it) merely because it happens in the brain is ludicrous.

Point taken.

helios wrote:

And the capability to make delicious dishes is the cooking intelligence?

No, but both require creativity.

helios wrote:

Just because it has "intelligence" in the name? It also has "quotient" in the name, but it's not the result of dividing anything by anything.

Google wrote:

quo·tient/ˈkwōSHənt/Noun
1. A result obtained by dividing one quantity by another.
2. A degree or amount of a specified quality or characteristic: "an increase in our cynicism quotient".

Using the second definition (since the first is irrelevant because as you said, nothing is necessarily being divided), Intelligence Quotient implies a quantity of intelligence.

[Audie]

I did one of those searches "Inside This Book" at Amazon to get as many pages displayed out of it as I could, for the book Artificial General Intelligence. One query was "define intelligence" in consonance with an article's subsection with that name. This is an excerpt:

A working definition is a definition concrete enough that you can directly work with it. By accepting a working definition of intelligence, it does not mean that you really believe that it fully captures the concept "intelligence", but that you will take it as a goal for your current research project.

Therefore, the lack of a consensus on what intelligence is does not prevent each researcher from picking (consciously or not) a working definition of intelligence. Actually, unless you keep one (or more than one) definition, you cannot claim that you are working on artificial [general] intelligence.

And I'd go on to add that, without a good working definition, one not hung up on a religious "true intelligence," a person could not suggest he or she is relevantly interested in intelligence. . . .

By accepting a working definition of intelligence, the most important commitments a researcher makes are on the acceptable assumptions and desired results, which bind all the concrete work that follows. [. . .] For our current purposes, there is no "right" or "wrong" working definition for intelligence, but there are "better" and "not-so-good" ones.

My own working definition would involve the performance on a set of tests, like a set where a g factor could be extracted (Wikipedia has an entry on it), and I wouldn't assume the performing cognitive system was of any predefined range of shapes. Some tests would not require much mobility and others would require some mobility (the cognitive system would be no use if an optimal solution required complex fluid tasks). Some tests would have a time limit of a few minutes or hours and others would have a time limit of a few months or years. And so on.

[wtf]

@chrisname I agree with almost everything you said.

but there are usually lots of different methods for finding the right answer

I would also like to point out that with mathematics, as a general rule, its not testing intelligence period. Its testing knowledge and capability in applying that knowledge. A 'true' test of intelligence might test for something like speed in computing arithmetic but leave all other advanced mathematical subjects off limits, as this is almost always a subject not studied by the average person, and would thereby produce a strong bias towards mathematicians and math majors. Want to prove your worth to a graduate school, take the GMAT, which is fine, but its not an IQ test. A perfect IQ test, if one existed would test for only routine tasks (such as arithmatic etc), and speed, and test for other abstract reasoning skills like logic and problem solving and 'being able to think outside the box' (otherwise known as creativity), and as noted by other posters other types of intelligence such as motor skills etc to give an INDICATION of how well one MIGHT be able to succeed in an advanced college math course for example. To sum it up, IQ tests test for aptitude, hence the synonym "aptitude test".

[AngelHoof]

IQ != intelligence;

Case closed.

[jsmith]

I think intelligence == IQ is a tautology since they are one in the same (that is the I in IQ after all), however knowledge != intelligence. That's why IQ tests don't ask trivia questions. (Though why some ask for definitions of words is beyond me).

[chrisname]

But... but... I could make a test called Speed Quotient, which implies measurement of maximum instantaneous speed, when it in fact measures average speed and say SQ == speed. Someone with an SQ of 100 might be a better sprinter than someone with an SQ of 140, but the latter is a better distance runner.

[helios]

The value of the maximum instantaneous speed could get skewed if someone happened to, for some reason, get really fast for just a few moments. Measuring the mean speed irons over such irregularities, giving back a statistically more meaningful value.