5 ms·

Computers can solve puzzles too. Would you say, then, that computers can reason? The problem with your argument is computers and animals are not cogitative, in

Computers can solve puzzles too. Would you say, then, that computers can reason? The problem with your argument is computers and animals are not cogitative, in any way similar to humans. There might be instances when animals come close, but that is no reason to imply all animals can. That means some come close.

posterboy 9y ago
animals solve without being told how to by a human, in contrast to a computer. of course newer AI is different, and it's called AI for a reason.
- mabub24 9y ago
  Newer AI is entirely limited by the algorithmic and logical frameworks assigned to it by others than itself. Machine learning is not "true" AI. It's called AI because it's slick marketing. Animals can solve puzzles. Solving puzzles, however, doesn't mean you can reason in the same way humans can. For instance, humans can ask why they are solving the puzzle. That's an expression of cogitative ability that animals don't have.
  - Dylan16807 9y ago
    If you ask a human why they are doing something they will often make up a reason and not even realize they're doing so. Especially when emotions are involved.
    - mabub24 9y ago
      That's true, but rather than "make-up" they are assigning a reason to their actions because they did them voluntarily. That is a rational thing to do, that gives a why for their action through their introspection, or sheer need for justification. If it is not voluntary, they'll often say that or it is immediately apparent, or they are ill. And we are fully capable of investigating that reason, that why, to find out whether it is reasonable or whether a better why can be provided. Animals cannot assign a reason, nor can we "investigate an animal's reason", because that's nonsense. Animals don't give why's.
  - posterboy 9y ago
    This whole discussion is just about people unsure whether something else is smarter than thdm or not.
dkersten 9y ago
If they can learn a technique on one problem and then use that technique on a completely different problem based on some observation, then sure. Why not? Monkeys, crows, ravens, octopuses and other animals have been shown to be able to do stuff like this. > but that is no reason to imply all animals can I don't think anyone was making the claim that all animals can. (In case you're not a native english speaker, GP said "all sorts of animals", which means "many different kinds of animals" and not "all animals")
jdmichal 9y ago
Computers are tools. They don't reason; they do exactly what they're instructed to. A stick can solve a problem too, but it's still a tool, because it only does exactly what you tell it to. This will be true until we develop an AI with intrinsic agency. Furthermore, I'd say that your over-focus on reasoning is misguided. There are many aspects of cognition and consciousness, and reasoning is just one of them. (And, as others have mentioned, there's plenty of evidence for various forms of abstract reasoning in studied animals.) Wikipedia's article on Animal Cognition has a convenient listing of different aspects that have been studied: https://en.wikipedia.org/wiki/Animal_cognition https://en.wikipedia.org/wiki/Animal_cognition
- mabub24 9y ago
  This whole comment chain started because the quote missed the core part of first sentence. "The core difference between humans and animals is that humans can reason; we possess cogitative powers." (I have emphasized the point that was missed.) A core difference between humans and animals is the cogitative powers of humans. Animals simply do not show the same cogitative powers. Animals do have cognition. I don't question that at all. I'm saying, in regards to cogitation (introspection), animals pale to humans.
  - jdmichal 9y ago
    Disagree. This has been a "fact" which has been steadily eroded as time goes on. It started with humans using tools... Until we found animals using tools. And it's been a series of dominos falling ever since, with the current studies on animal linguistics just being another one in the line. For instance, your invocation of "introspection" is covered in the linked article by "theory of mind". And maybe in the second paragraph covering metacognition in "consciousness". I think this quote covers the ideas succinctly: "Convergent evidence indicates that non-human animals have the neuroanatomical, neurochemical, and neurophysiological substrates of conscious states along with the capacity to exhibit intentional behaviors. Consequently, the weight of evidence indicates that humans are not unique in possessing the neurological substrates that generate consciousness. Non-human animals, including all mammals and birds, and many other creatures, including octopuses, also possess these neurological substrates." [0] We've found evidence in some animal or another for pretty much everything we've ever thought was unique about human cognitive abilities. So far, the only unique thing about humans is how broad and powerful our cognition is. Which is maybe what you mean, and are having a hard time expressing. [0] http://fcmconference.org/img/CambridgeDeclarationOnConsciousness.pdf http://fcmconference.org/img/CambridgeDeclarationOnConscious...
coldtea 9y ago
>Computers can solve puzzles too. Would you say, then, that computers can reason? Obviously yes. There's nothing special to reason that requires a human.
celticninja 9y ago
computers solve puzzles because we program them to do so. animals solve puzzles because it provides a reward to them, the same way it does for us.

posterboy 9y ago

animals solve without being told how to by a human, in contrast to a computer. of course newer AI is different, and it's called AI for a reason.

mabub24 9y ago
Newer AI is entirely limited by the algorithmic and logical frameworks assigned to it by others than itself. Machine learning is not "true" AI. It's called AI because it's slick marketing. Animals can solve puzzles. Solving puzzles, however, doesn't mean you can reason in the same way humans can. For instance, humans can ask why they are solving the puzzle. That's an expression of cogitative ability that animals don't have.
- Dylan16807 9y ago
  If you ask a human why they are doing something they will often make up a reason and not even realize they're doing so. Especially when emotions are involved.
  - mabub24 9y ago
    That's true, but rather than "make-up" they are assigning a reason to their actions because they did them voluntarily. That is a rational thing to do, that gives a why for their action through their introspection, or sheer need for justification. If it is not voluntary, they'll often say that or it is immediately apparent, or they are ill. And we are fully capable of investigating that reason, that why, to find out whether it is reasonable or whether a better why can be provided. Animals cannot assign a reason, nor can we "investigate an animal's reason", because that's nonsense. Animals don't give why's.
- posterboy 9y ago
  This whole discussion is just about people unsure whether something else is smarter than thdm or not.

dkersten 9y ago

If they can learn a technique on one problem and then use that technique on a completely different problem based on some observation, then sure. Why not? Monkeys, crows, ravens, octopuses and other animals have been shown to be able to do stuff like this. > but that is no reason to imply all animals can I don't think anyone was making the claim that all animals can. (In case you're not a native english speaker, GP said "all sorts of animals", which means "many different kinds of animals" and not "all animals")

jdmichal 9y ago

Computers are tools. They don't reason; they do exactly what they're instructed to. A stick can solve a problem too, but it's still a tool, because it only does exactly what you tell it to. This will be true until we develop an AI with intrinsic agency. Furthermore, I'd say that your over-focus on reasoning is misguided. There are many aspects of cognition and consciousness, and reasoning is just one of them. (And, as others have mentioned, there's plenty of evidence for various forms of abstract reasoning in studied animals.) Wikipedia's article on Animal Cognition has a convenient listing of different aspects that have been studied: https://en.wikipedia.org/wiki/Animal_cognition https://en.wikipedia.org/wiki/Animal_cognition

mabub24 9y ago
This whole comment chain started because the quote missed the core part of first sentence. "The core difference between humans and animals is that humans can reason; we possess cogitative powers." (I have emphasized the point that was missed.) A core difference between humans and animals is the cogitative powers of humans. Animals simply do not show the same cogitative powers. Animals do have cognition. I don't question that at all. I'm saying, in regards to cogitation (introspection), animals pale to humans.
- jdmichal 9y ago
  Disagree. This has been a "fact" which has been steadily eroded as time goes on. It started with humans using tools... Until we found animals using tools. And it's been a series of dominos falling ever since, with the current studies on animal linguistics just being another one in the line. For instance, your invocation of "introspection" is covered in the linked article by "theory of mind". And maybe in the second paragraph covering metacognition in "consciousness". I think this quote covers the ideas succinctly: "Convergent evidence indicates that non-human animals have the neuroanatomical, neurochemical, and neurophysiological substrates of conscious states along with the capacity to exhibit intentional behaviors. Consequently, the weight of evidence indicates that humans are not unique in possessing the neurological substrates that generate consciousness. Non-human animals, including all mammals and birds, and many other creatures, including octopuses, also possess these neurological substrates." [0] We've found evidence in some animal or another for pretty much everything we've ever thought was unique about human cognitive abilities. So far, the only unique thing about humans is how broad and powerful our cognition is. Which is maybe what you mean, and are having a hard time expressing. [0] http://fcmconference.org/img/CambridgeDeclarationOnConsciousness.pdf http://fcmconference.org/img/CambridgeDeclarationOnConscious...

coldtea 9y ago

>Computers can solve puzzles too. Would you say, then, that computers can reason? Obviously yes. There's nothing special to reason that requires a human.

celticninja 9y ago

computers solve puzzles because we program them to do so. animals solve puzzles because it provides a reward to them, the same way it does for us.