The Naked Scientists

The Naked Scientists Forum

Author Topic: Intelligence as an Illusion?  (Read 15932 times)

Offline paros

  • Jr. Member
  • **
  • Posts: 24
    • View Profile
Intelligence as an Illusion?
« on: 22/02/2008 00:29:07 »
Quote
Originally posted by Hadrian
This all shows us just how we and the planet are miss served by our illusion of human inelegance.

While I agree that humans can be very inelegant, I assume the word you wanted was intelligence.

Making this substitution, I would agree that much of our intelligence is illusory (this is part of the trouble in trying to define artificial intelligence, that we can never really get to grips with how illusory our own intelligence is).  The problem is that simply removing that illusion does not give us the intelligence we lack.

I'd love to talk more about this topic. 
Intelligence is an illusion? Are you taking a behaviorist stance here?  Something else?  Please expand!

MOD EDIT - See below
« Last Edit: 22/02/2008 10:23:06 by BenV »


 

another_someone

  • Guest
Re: Intelligence as an Illusion?
« Reply #1 on: 22/02/2008 00:42:14 »
Yes, I do regard myself as primarily an empiricist, and thus take to take a behaviourist perspective.

If one looks at human behaviour (not the technology) as an outsider (the only way one can look at the behaviour of any other species), then how are we any qualitatively different from many other species (at least to the extent that we can draw a line and say that humans act 'intelligently' while other animals do not).

We like to think of ourselves as unique (no doubt elephants think they are unique also), and hence we invent this notion that we are uniquely intelligent, and that other species do not think and behave as we do.  The fact is that far more of our behaviour is predicated on instinct than we would like to admit, and intelligence is only a thin layer over and above that, and that thin layer is only marginally thicker than the thin layer of intelligence that our close animal relatives use.
 

Offline paros

  • Jr. Member
  • **
  • Posts: 24
    • View Profile
Re: Intelligence as an Illusion?
« Reply #2 on: 22/02/2008 00:56:52 »
. . . then how are we any qualitatively different from many other species . . .

The question is well stated.
Define intelligence as "adaptive behavior" and you have all you need to form a complete spectrum of all animals.

Let the scientific/biological definition of the word "intelligence" stand for adaptive behavior.  You can then form a complete totem pole of intelligence with insects at the bottom and primates and humans at the top.   Insects' behavior can be changed, but only very slightly, and only through repeated trials with an aversive stimulus.   Mammals such as cats and dogs have a much wider range of adaptation (the ability to adapt).   

Mirror, mirror, on the wall, who is the Most Adaptive of all?   Yeah, that would be homo sapiens.

Melvin Konner authored a book (that is actually about anthropology) called The Tangled Wing.   I beleive chapter 2 of that book is called Adaptation.    You might want to peruse those pages to get an intuitive feel of the perspective I am using to deliver this post.
 

Offline paros

  • Jr. Member
  • **
  • Posts: 24
    • View Profile
Re: Intelligence as an Illusion?
« Reply #3 on: 22/02/2008 01:04:44 »
Oh second thought, it looks like I gave you a quantitative difference rather than a qualitative one.
Please disregard!    ;)
 

Offline BenV

  • Neilep Level Member
  • ******
  • Posts: 1503
    • View Profile
Intelligence as an Illusion?
« Reply #4 on: 22/02/2008 10:22:23 »
Re MOD EDIT: Please excuse my editing, I've re-titled the thread to make it easier to find again, and taken the html out of the quote to make it easier to read.
 

lyner

  • Guest
Intelligence as an Illusion?
« Reply #5 on: 22/02/2008 22:07:32 »
Jeez - that took some real intelligence to do that!!
You would have to be a human, both to do it and to appreciate it.
 

another_someone

  • Guest
Intelligence as an Illusion?
« Reply #6 on: 23/02/2008 03:48:11 »
Jeez - that took some real intelligence to do that!!
You would have to be a human, both to do it and to appreciate it.

Actually, he did neither - he gave instructions to his computer to do it - but we do not recognise the intelligence of his computer (and the servers) in following his instructions, but we do recognise his intelligence for issuing the instructions.
 

lyner

  • Guest
Intelligence as an Illusion?
« Reply #7 on: 26/02/2008 19:23:56 »
Quote
Actually, he did neither - he gave instructions to his computer to do it . . . . .
By that argument what does any of us actually do?
IK Brunell only told other people to do things.
 

another_someone

  • Guest
Intelligence as an Illusion?
« Reply #8 on: 26/02/2008 23:16:16 »
Quote
Actually, he did neither - he gave instructions to his computer to do it . . . . .
By that argument what does any of us actually do?
IK Brunell only told other people to do things.

To some extent, that is true - but there are also differences.

I would suspect that many of the things IK Brunell told other people are things he himself was competent to do, and in any case it was done by humans, and so provably capable of being done by humans.

But, in the wider sense, it is still perfectly valid to say the intelligence (ability to solve problems) of the human collective is greater than the intelligence of the sole human individual.

The point is that the collective intelligence that allowed the edit of title of this document included both human and non-human elements, and the work that each party undertook in the process was a role that could not have practically been undertaken by the other party.  The human could be unlikely to even be able to calculate all of the electronic signals to send down the wire, and certainly could not do so in an acceptable time, while the machine requires a human to give it an starting point to be able to deliver a result desired by another human.
 

Offline rhade

  • Sr. Member
  • ****
  • Posts: 142
  • Do what you will, but first be such as can will!
    • View Profile
Intelligence as an Illusion?
« Reply #9 on: 28/02/2008 10:57:20 »
Humans have hands to build things. Other creatures, which may have similar intelligence, don't have hands. I think I'm agreeing with another someone about taking human technology out of the equation here.
« Last Edit: 12/09/2008 17:22:59 by rhade »
 

Offline paros

  • Jr. Member
  • **
  • Posts: 24
    • View Profile
Intelligence as an Illusion?
« Reply #10 on: 28/02/2008 20:56:56 »
Humans have hands to build things. Other creatures, which may have similar intelligence, don't have hands. I think I'm agreeing with another someone about taking human technology out of the equation here.

Okay well.  Is another_someone pointing out an absurdity in order to make a point without saying it directly, or is he actually stating what he means to say?   

I know that question sounds confusing. Let me write down the two possibilities so you can better understand what I am asking.
1) another_someone is showing that "intelligence" is a meaningless word because look at how stupid and contrarian people are when they forget to attribute any intelligence to the hardware.
2) another_someone is demanding that we should equally attribute intelligence to the hardware in order to be consistent.

Seriously, I don't know what he is 'getting at' here!
 

another_someone

  • Guest
Intelligence as an Illusion?
« Reply #11 on: 28/02/2008 22:24:45 »
OK, partly I suppose I did get sidetracked, but also I suppose there were a number of issues that I had not properly formulated in my own mind, so I knew what I was not happy with in terms of the way we used concepts of intelligence, but it was more difficult to say what would be more meaningful to me.

I don't think we can really talk about something is intelligent or is not intelligent until we have a clear (even if arbitrary) definition of what we mean by intelligence (for many people, they become unstuck with this because they seem to start with the notion that humans are intelligent, and to a greater or lesser degree, non-humans are not intelligent, and then try and form a definition of intelligence that fits this prejudice - and it was that prejudice I was arguing against - which is itself probably a rather negative position to be starting from).

I would regard intelligence to be the ability to solve the widest range of problems in the shortest time.  The problem here is that we already have a two dimensional definition, in that we have both the scope of the problems, and the speed of coming to a solution.  But I cannot see that we can reduce it to a one dimensional measure, since nature itself can solve all problems (problems far more complex than humans have yet been able to tackle) if given enough time, and it can solve some problems exceedingly fast.

But then we come to the issue of what are we measuring the intelligence of?

Are we measuring the intelligence of a single organism (e.g. an isolated human being), or the collective intelligence of the human species (where each human being may be good at solving some problems, but collectively humanity can solve far more problems - this goes back to the issue of the scope of the problems that may be solved), or the intelligence of human society (in which case, are the machines themselves part of that society, since the machines can accelerate the speed at which we can solve problems)?

Let us take as a deliberately extreme example, the comparative intelligence of humans and bacteria.  Ignoring prejudice, and just looking at the ability of bacteria to solve problems.  If one looks at the collective intelligence of humanity and the collective intelligence of the bacteria that causes TB, humanity has created antibiotics to attack TB, yet TB has come back with antibiotic resistance, so each has met the challenges provided by the other.  Clearly we are talking here about the collective intelligence of the group (antibiotics were developed using the intelligence of many humans, not the intelligence of one; and the resistance to antibiotics was created using the 'intelligence' of many bacteria, not the intelligence of one).  It is true that if you look at the intelligence of a single human and a single bacteria, the human is more intelligent, but are you really comparing like with like.  A bacteria is a single cell, the human is already a colony of billions of cells; so should you not really compare the intelligence of a single bacteria to the intelligence of a single neuron?

The point is that in my view, intelligence is not so much an intrinsic measure of a thing, but a measure of an interaction between a system (whether the system is a machine, a human being, or a collective of many humans and machines combined) and the environment it is functioning in.  When you talk about intelligence, you have to make sure you understand what you include within the system and what is without it; and to make sure that what you measure is the intelligence of only that which is a part of your system and not a component of the wider environment (i.e. if you are looking at the measure of intelligence of a single human being, make sure you are really measuring the intelligence of the isolated human being, and not how that single human being reflects the intelligence of the wider human society he is a part of).
 

Offline john ford

  • Jr. Member
  • **
  • Posts: 13
    • View Profile
Intelligence as an Illusion?
« Reply #12 on: 28/02/2008 22:32:07 »
The only good thing about intelligence is the perpensity to act stupid - and the only thing that saves us all is the fact that we aren't all stupid on the same day.

Sorry - just my ignorance showing.

 

lyner

  • Guest
Intelligence as an Illusion?
« Reply #13 on: 28/02/2008 22:47:22 »
You will get used to the quirks of A-S's arguments if you visit these forums.
He enjoys being controversial and adds spice to many a thread.
Don't get wound up - just enjoy it!
 

lyner

  • Guest
Intelligence as an Illusion?
« Reply #14 on: 28/02/2008 22:54:35 »
What about the Turing test?
That works if 'intelligent' means 'like us'.
 

another_someone

  • Guest
Intelligence as an Illusion?
« Reply #15 on: 29/02/2008 00:24:16 »
What about the Turing test?
That works if 'intelligent' means 'like us'.

But what does the Turing test actually test?  It rather depends on the details of how the Turing test is set up, and what you mean by 'like us'.

If you compared a computer to some Amazonian native, which would be more 'like us'?  It rather depends on what aspects of personal experience you are dealing with.  A Turing test that removes the need to equality of experience, and equality of sensory input, then I don't think the test is that difficult to pass (even the ELIZA program in 1966 did a fairly good job of it).  Aside from experience (which is as much cultural as anything else), the thing the Turing test tests most is natural language processing.

Ofcourse, the real irony about using the Turing test to test for intelligence is that it cannot test for an intelligence greater than human intelligence, for that would be as non-human as an intelligence less than human intelligence.
 

another_someone

  • Guest
Intelligence as an Illusion?
« Reply #16 on: 29/02/2008 00:29:42 »
The only good thing about intelligence is the perpensity to act stupid - and the only thing that saves us all is the fact that we aren't all stupid on the same day.

Sorry - just my ignorance showing.

Your ignorance, or mine?

It could be argued that human intelligence (in crude colloquial terms - not in highly specific definitions we are trying to develop here) is about explaining something in such terms as to make is so obvious that everybody else is kicking themselves for not having thought of it first.  If I have not achieved that (and I can imagine, since I am still trying to work somethings out in my own head, that I probably have not achieved that), then I would suggests the fault lies as much with me as with you.

If you want to point out what does not make sense, I will think about how to put it another way.
 

lyner

  • Guest
Intelligence as an Illusion?
« Reply #17 on: 29/02/2008 07:56:32 »
This thread is on to a loser.
The word intelligence was coined long before anyone tried to quantify it. It is not a Scientific term and it is not definable. The problem is that it is a very woolly concept and a multidemensional 'quantity'.
There are occasions where one can say "A is more intelligent than B" and for it to have some sort of vague meaning but that's about as far as it goes.

This discussion is just going to get people cross with each other to no purpose. We're talking faith not fact here.
 

another_someone

  • Guest
Intelligence as an Illusion?
« Reply #18 on: 29/02/2008 13:10:02 »
This thread is on to a loser.
The word intelligence was coined long before anyone tried to quantify it. It is not a Scientific term and it is not definable. The problem is that it is a very woolly concept and a multidemensional 'quantity'.
There are occasions where one can say "A is more intelligent than B" and for it to have some sort of vague meaning but that's about as far as it goes.

This discussion is just going to get people cross with each other to no purpose. We're talking faith not fact here.

This is substantially true, and is why I am generally wary about such discussions about 'intelligence' - but one can also say that the fact that a word is colloquially used rather loosely does not prohibit a discussion of the ideas if one applies a more rigid definition of the word (words such as force, energy, momentum, etc., all have loose colloquial usage, but they also have fairly strict scientific usage, although I will admit that energy can sometimes have multiple meanings even in science, depending on whether one is dealing with mechanics or thermodynamic).

This is why one has to start any discussion on intelligence with a clear definition as to how one is using the word (which is what I tried to do above), and base all discussion on that definition (not on presumed colloquial usage).
« Last Edit: 29/02/2008 13:14:52 by another_someone »
 

lyner

  • Guest
Intelligence as an Illusion?
« Reply #19 on: 29/02/2008 23:57:35 »
If we start with an axiomatic definition of intelligence then there is no room for opinion. If we don't then opinions can be anything.
I know when I see a sign of intelligence and when I see stupidity -but that's just my personal point of view. I base my assessment of intelligence on the ability to solve complex problems and to communicate effectively. But this involves a window with an upper as well as a lower limit - due to my personal limits.
I probably could not recognise a 'God - Like' intelligence at all. Some excellent plans and decisions would appear totally stupid to me as I could not appreciate them.
This is why I base my assessment from a 'like me - but a bit smarter '  standpoint.
 

another_someone

  • Guest
Intelligence as an Illusion?
« Reply #20 on: 01/03/2008 02:36:26 »
If we start with an axiomatic definition of intelligence then there is no room for opinion. If we don't then opinions can be anything.
I know when I see a sign of intelligence and when I see stupidity -but that's just my personal point of view. I base my assessment of intelligence on the ability to solve complex problems and to communicate effectively. But this involves a window with an upper as well as a lower limit - due to my personal limits.
I probably could not recognise a 'God - Like' intelligence at all. Some excellent plans and decisions would appear totally stupid to me as I could not appreciate them.
This is why I base my assessment from a 'like me - but a bit smarter '  standpoint.

But you cannot do science on anything but axiomatic definitions.  If you have a definition that changes from place to place, person to person, then you cannot do objective science on it - so you have to create an axiomatic base in order to build an objective science upon it.  This must be true even if the axioms are arbitrary.
 

lyner

  • Guest
Intelligence as an Illusion?
« Reply #21 on: 01/03/2008 09:56:59 »
Quote
But you cannot do science on anything but axiomatic definitions.
What sort of Scientist would launch into an investigation of something which has an unknown number of parameters, about which you can do no measurements and in which any observations are almost totally subjective? He'd just have to be a Cosmologist.
 

another_someone

  • Guest
Intelligence as an Illusion?
« Reply #22 on: 01/03/2008 11:32:33 »
Quote
But you cannot do science on anything but axiomatic definitions.
What sort of Scientist would launch into an investigation of something which has an unknown number of parameters, about which you can do no measurements and in which any observations are almost totally subjective? He'd just have to be a Cosmologist.

LOL - cosmologists aside, I think a more pertinent analogy would be evolutionary science, because IMV evolution, like intelligence, is ultimately about problem solving and information processing (hence my analogy above between mycobacteria responsible for TB, and the scientists trying to develop antibiotics effective against bacteria).  The difference is ofcourse, that evolution works on an intergenerational development of solutions, while what we regard as intelligence is an intragenerational processing of information.  This allows intelligence to work faster than genetic inheritance  (especially for long lived organisms, for which generational times are too long to allow pure reliance on intergenerational solutions to problems - less of an issue for short lived bacteria).

As for doing science on the unmeasurable - try measuring entropy.  Again, entropy, like intelligence, is about information.  We just don't yet have the tools to adequately be able to develop abstract models of information in all but the simplest systems (i.e. in simply communication systems, we can develop good models of how information works, although even there one has ambiguities - but such systems are inherently one dimensional - in a very physical sense - they often travel down a piece of wire, or can be altered to a model that equates to travelling down a single wire).
« Last Edit: 01/03/2008 11:48:27 by another_someone »
 

lyner

  • Guest
Intelligence as an Illusion?
« Reply #23 on: 01/03/2008 14:41:01 »
Quote
As for doing science on the unmeasurable - try measuring entropy
Not really fair; entropy change can be calculated using readily measurable quantities.
I suppose you could have some sort of definition of intelligence involving entropy, efficiency and a load of information theory. But the 'usefulness' intelligence should / would surely be taken into account and that wrecks the Science involved.
 

another_someone

  • Guest
Intelligence as an Illusion?
« Reply #24 on: 01/03/2008 15:08:27 »
Quote
As for doing science on the unmeasurable - try measuring entropy
Not really fair; entropy change can be calculated using readily measurable quantities.

Can you?  What units is entropy measured in?

You can do qualitative studies in relative entropy (i.e. you can say that A had more entropy than B - sometimes), but how do you do quantitative studies in entropy?

In some cases, I am sure we could make qualitative statements about changes in intelligence, but we cannot make quantitative measures of it (excepting to the degree that one has faith in IQ tests - but at best they can only be applied in limited circumstances).  In very crude terms, one can compare two chess playing programs, and because the purpose of the programs is very one dimensional, one can say that one chess playing program is more intelligent than the other - to the same degree that one can say on a one dimensional measure that one system has less entropy than another - but neither is quantitative, and neither works in a multi-dimensional environment).

I suppose you could have some sort of definition of intelligence involving entropy, efficiency and a load of information theory. But the 'usefulness' intelligence should / would surely be taken into account and that wrecks the Science involved.

Why would it wreck the science involved?

Surely, usefulness is contextual, but science should be context free.

I can measure the energy of fusion of hydrogen, but it does not mean that this energy is always useful - that rather depends on the context.
« Last Edit: 01/03/2008 15:14:58 by another_someone »
 

The Naked Scientists Forum

Intelligence as an Illusion?
« Reply #24 on: 01/03/2008 15:08:27 »

 

SMF 2.0.10 | SMF © 2015, Simple Machines
SMFAds for Free Forums