3 ms·

Am I the only one who thinks mirror test is stupid? You won't realize your recorded voice is yours. You won't realize it's you on a CCTV footage. It's not gran

Am I the only one who thinks mirror test is stupid? You won't realize your recorded voice is yours. You won't realize it's you on a CCTV footage. It's not granted. On the other hand, it's trivial to make robot pass the mirror test, indeed necessary in the environment where mirrors are common.

lutusp 13y ago
> Am I the only one who thinks mirror test is stupid? It's pretty reliable, and the scientists who measure it have a number of ways to be sure they're witnessing what they think. One way is to put a colored spot on an animal's forehead in advance of the test. If the animal looks in the mirror, sees the spot, and then reaches to his own forehead to remove it, we can be pretty sure he grasps that he's looking at a reflection of himself. But as to the outcome supporting or not supporting some theory of consciousness in animals, that's a different story. > On the other hand, it's trivial to make robot pass the mirror test, indeed necessary in the environment where mirrors are common. That's certainly true, and it's a good argument for the limited significance of the mirror test in ideas about consciousness.
- mdda 13y ago
  While the mirror test isn't a positive test for consciousness, for a species not to be able to pass the mirror test indicates that their 'mental model' of the world is lacking in something that would seem to be a necessary component of self-awareness and (therefore?) consciousness.
  - lutusp 13y ago
    I think the mirror test, and consciousness, are separate issues. 1. As someone else here pointed out, a computer can be programmed to pass the mirror test, indeed a sophisticated robot must be able to do this, to be able to successfully navigate a complex environment. But that robot need not be conscious to do so. 2. A conscious creature (a creature that shows classic signs of being conscious) isn't necessarily able to pass the mirror test. In this article -- http://en.wikipedia.org/wiki/Mirror_test http://en.wikipedia.org/wiki/Mirror_test -- we read that "Primates, other than the great apes, have so far universally failed the mirror test." I think many primates would meet various criteria for consciousness, but can't pass the mirror test. Interestingly, in the linked article, birds of the family Corvidae (crows, ravens, European magpies, gray jays (aka "camp robbers"), and others) pass the mirror test, possibly the only bird type that can. Corvids are very smart birds. http://en.wikipedia.org/wiki/Corvidae http://en.wikipedia.org/wiki/Corvidae

lutusp 13y ago

> Am I the only one who thinks mirror test is stupid? It's pretty reliable, and the scientists who measure it have a number of ways to be sure they're witnessing what they think. One way is to put a colored spot on an animal's forehead in advance of the test. If the animal looks in the mirror, sees the spot, and then reaches to his own forehead to remove it, we can be pretty sure he grasps that he's looking at a reflection of himself. But as to the outcome supporting or not supporting some theory of consciousness in animals, that's a different story. > On the other hand, it's trivial to make robot pass the mirror test, indeed necessary in the environment where mirrors are common. That's certainly true, and it's a good argument for the limited significance of the mirror test in ideas about consciousness.

mdda 13y ago
While the mirror test isn't a positive test for consciousness, for a species not to be able to pass the mirror test indicates that their 'mental model' of the world is lacking in something that would seem to be a necessary component of self-awareness and (therefore?) consciousness.
- lutusp 13y ago
  I think the mirror test, and consciousness, are separate issues. 1. As someone else here pointed out, a computer can be programmed to pass the mirror test, indeed a sophisticated robot must be able to do this, to be able to successfully navigate a complex environment. But that robot need not be conscious to do so. 2. A conscious creature (a creature that shows classic signs of being conscious) isn't necessarily able to pass the mirror test. In this article -- http://en.wikipedia.org/wiki/Mirror_test http://en.wikipedia.org/wiki/Mirror_test -- we read that "Primates, other than the great apes, have so far universally failed the mirror test." I think many primates would meet various criteria for consciousness, but can't pass the mirror test. Interestingly, in the linked article, birds of the family Corvidae (crows, ravens, European magpies, gray jays (aka "camp robbers"), and others) pass the mirror test, possibly the only bird type that can. Corvids are very smart birds. http://en.wikipedia.org/wiki/Corvidae http://en.wikipedia.org/wiki/Corvidae