What is intelligence?

[K. R. Thórisson]

Please post two questions you would like to have answered or have the class discuss, based on the assigned reading en.wikipedia.org/wiki/Theory_of_multiple_intelligences

[Last chance to submit: August 30th @ 11:59]

You may also post answer to your own, or postings by your classmates.

[kristjan]

In the reading where it is mentioned that certain critics balk at Gardner's idea of widening the definition of intelligence by saying, that it ignores "the connotation of intelligence...[which] has always connoted the kind of thinking skills that makes one successful in school."
But does an IQ test even cover all the kinds of thinking skills that are needed for all classes and subjects in the school.
If intelligence represents thinking skills needed for school, how can the definition of intelligence stay still and not be broadened, when it is possible to teach new content in schools that utilize different kind of thinking skills than currently is used in under the term intelligence.

[fabrizio]

In the last class, we focused for a little bit on a question like this: If we take 2 people with the same experiences with the same instruction, grown in the same environment, and we give them a problem that they can solve but in different time, can we say that the fastest is more intelligent?
In the reading there is a sentence :”the theory predicts that a child who learns to multiply easily is not necessarily generally more intelligent than a child who has more difficulty on this task.” Is this a kind of answer for our question?

I want add another thing about the question that we posed during the last class: How can we talk about 2 living people that have the same experiences in all the fields? We are talking about 2 equal people and this isn’t reality, there aren’t two equal people so we can’t make that question but, if we are talking about artificial intelligence, then we can try to talk about two equal artificial intelligences, and reformulate the question. If we take 2 “equal” artificial intelligences and we give them a problem that they solve in different time, can we say that the fastest is more intelligent?

About the reading, I don’t think that the 8/9 areas analyzed are type of intelligences, they seem to me something like: “predispositions”. What do you think about it?

[paolo]

Answering first to Fabrizio, I agree with you. I do not think the areas suggested by Gardner are really types of intelligences.
Generally speaking, I agree with Gardner that the "intelligence" is not dominated by a single general ability. But, on the other hand, I do not agree that intelligence concerns the person's ability to learn (in the sense of assimilate new information). In my humble opinion, intelligence is something that concerns more reasoning than learning. As Pei Wang asserted in his article "Four Basic Questions about Artificial Intelligence" (www.iiim.is/2010/05/questions-about-artificial-intelligence/), "intelligence" refers to the ability of adapting to the environment while working with insufficient knowledge and resources.
But, to be honest, I do not even think that these areas can be grouped under the term "predispositions".
At the elementary school, I had a lot of problems with multiplications. Nevertheless, after, I actually had less problems than most of my ex-schoolmates of the elementary school to learn the "study of functions" or the integrals. So, what is happened? Have I become more intelligent? I do not think so. I actually think that it is just a matter of time, study and effort. Isn't it?

Coming back to the Wikipedia's article, it suggests that some teachers say that the theory of multiple intelligences validates that students learn in different ways. So, should not we guess that the real proof of intelligence is to find, for an educator, the best way to teach to every student? The less the student is "smart" (according to the definition of this article), the more the teacher is smart if the student get the point. Or not?

[sabrina12]

Which role plays curiosity in intelligence? There are not just instincts that trigger our intelligence, but also curiosity. If we watch non-human animals you can also see that animals are more often curious, when they have a bigger brain and a higher intelligence. Do living organisms need instincts AND curiosity to get intelligent? And is it theoretically possible to create an artificial intelligence with something like consciousness or curiosity in a few decades?
But an also interesting question that David Chalmers aks in his paper, is:
"We do not expect a computer model of a hurricane to be a real hurricane, so why should a computer model of mind be a real mind?"

[helgil]

1. According to the evidence, cognitive science seems to have the upper hand vs behaviorism and situated cognition, that is that cognition is indeed a collection of processes. If that is true, what's stopping people from creating an artificial intelligence that is just a collection of processes and the better the algorithms the better the intelligence. And is it possible that developing an artificial general intelligence isn't really possible yet because all the cognitive processes of the human mind for example aren't really understood yet and therefore we can't program the AI because we don't know how the human mind does it. Isn't intelligence just a product of processes and depending on the processes what kind of intelligence it is, and fast intelligence therefore the product of fast processes and not necessarily any more intelligence than one based on the same but slower processes.

2. Regarding salience, that is that a robot would not be considered intelligent if it wouldn't move if a tree were falling on it. What are these kind of algorithms called, ones that detect changes in a visual input for example, compare two frames and decide whether something is about to fall on it. Or are there maybe not any available and people have been making them ad hoc until now. Is there anything in the way of classifying those kinds of processes as a form of intelligence or at least as a part of intelligence.

[krummi]

Like many, I've got a mixed feeling towards the theory of multiple intelligences. However, I think that its existence is necessary to contrast the "single general ability"-approach. I think this is especially useful for educational systems all over the world, who have in my opinion been emphasizing the development of logical intelligence as if logical intelligence is something that is fundamental to every ability that we possess/want to possess. Which is exactly what the theory of multiple intelligences does not assume.

So my first question relates to this: How do you think that western primary-educational systems will evolve in the coming years. Towards a curriculum that emphasizes logical intelligence OR towards a curriculum that emphasizes something like the 8/9 multiple intelligences that Gardner talks about (even though its not necessarily based on his theory).

My second question is the following: The Wikipedia article tells us that the theory of multiple intelligences is not close to being "proven" since it is lacking in empirical evidence. But how would you go about proving the theory? What experiments could be conducted? This seems fuzzy to me.

[palli]

Where do I start? Why discuss a theory that lacks empirical evidence for humans? That is, that these are not intelligences but rather skills or abilities. But ok, let's not make a distinction – this is a good exercise in thinking. Are these all necessary abilities, to some degree, to be intelligent? Clearly not. Or so is assumed (by Turing). If we assume the Turing-test is a good test for intelligence then the “Linguistic” intelligence (excluding spoken) is the main one that matters. Let's say you were disabled from the neck down. Are you then intellectually challenged by Gardner's definition (“Bodily-kinesthetic”)? But maybe you can have the underlying intelligence (ability to learn if you tried/could try) but the test is flawed. But I doubt that. I'm not sure the brain will develop right without stimulation. I guess we need a body, that is the AI needs an embodiment (has to be a robot) for that skill and probably in general for learning/exploring.