6 ms·

"This is because cats evolved as solitary hunters with little need for reading social cues, especially those for behavior modification." [...] “How the hell i

by jchendy 12y ago

"This is because cats evolved as solitary hunters with little need for reading social cues, especially those for behavior modification." [...] “How the hell is your cat supposed to know that you’re yelling at him because you want him to stop scratching the couch?” [...] "When your cat does the thing you want her do to, reward her with a treat, or affection." If we accept that cats can't learn from negative reinforcement, why would we think that positive reinforcement would work?

cstross 12y ago
Because positive reinforcement does work on cats -- the problem is, finding the positive behavioural pattern to reinforce! Positive and negative reinforcement patterns are not opposites, they're entirely separate processes. (Cat owner here.)
readerrrr 12y ago
Perhaps because yelling isn't necessarily interpreted as something negative, while food is certainly something positive.
XorNot 12y ago
This is all wrong though. Domestic cats have been domesticated for a long time. Cats which live with humans (or even other cats - they do naturally form colonies) adopt a number of pro-social behaviours to interact with us. They are in fact very adept at reading social cues - but conversely they are not pack-animals. They don't have the same instinctive need to identify a hierarchy of command, so much as "mess with" and "don't mess with".
- lingben 12y ago
  Among the most common being meowing. Cats do not mew at other cats, only at humans.
- puzzlingcaptcha 12y ago
  >Domestic cats have been domesticated for a long time. That sounds like lamarckism. Or do you suggest that cats were selectively bred for their social behaviour? I find it dubious.
  - pessimizer 12y ago
    >I find it dubious. You should check your detector.
  - Sharlin 12y ago
    How do you think domestic cats came into existence? Prosocial behavior and tolerance of humans are traits that are equal parts nature and nurture in cats. Obviously current housecats are descended from those wild cats that were more accepting of human presence and had traits that made them accepted by humans as well. Still, individual cats must learn to interact with humans, and if they aren't habituated to human presence during the first couple of months of their lifes, they end up very different psychologically compared to if they are.
    - puzzlingcaptcha 12y ago
      I questioned artificial selection (breeding), you replied with natural selection (increased reproductive success by adopting to living near humans). Briefly skimming some sources, we are both correct http://www.ncbi.nlm.nih.gov/pmc/articles/PMC2702791/#__sec6title http://www.ncbi.nlm.nih.gov/pmc/articles/PMC2702791/#__sec6t...
    - warfangle 12y ago
      The domestication of dogs happened a lot longer ago, and probably a lot more organically (more like what you're thinking). The wolves would follow us around while we hunted (or, alternatively, we'd happen on one of their meals and chase them off), and the less human averse among them got more food. Cats happened about the same time agriculture did. They were attracted to the rodents, and less rodents meant we were happier - and they were happier with a reliable food source. At the beginning it might have helped to not be afraid of humans, but we probably didn't mess with them too much because they were helping us out with all the mice in our grains. So the difference in behavior is clear, not just from a social species standpoint, but from the domestication path: The European Wolf was domesticated to dogs as a hunting partner; the African Wildcat was domesticated to kill the rodents that devoured our foodstuffs. For the few few thousand years that was probably enough - they didn't have to make friends with us, just keep our food safe. Earliest cat domestication evidence is 9,500 years old; earliest dog domestication evidence is about 36,000 years old. They've still got quite a ways to go ;) Now if we could just start breeding them for social/intelligence instead of how weird and fluffy we can make them. Hard to morally advocate for breeding when the shelter/feral population is so damn high, though. (I feel the same way about dogs, though, tbh).
Proleps 12y ago
Punishment also works on cats. You just have to be very consistent when applying it (this is exactly the same for dogs). If you punish your cat(or dog) you have to do it immediately. If the cat is done scratching its nails, and you squirt water, you are too late. [Edit] Changed negative reinforcement to punishment, rbehrends was correct when he pointed out that negative reinforcement is something different.
rbehrends 12y ago
That's not negative reinforcement. Negative reinforcement is when behavior is reinforced by removing an aversive stimulus. Yelling at a cat because it's scratching the couch is positive punishment [1]. And punishment works quite differently from reinforcement to begin with; it is typically less effective for far better socialized species such as humans and dogs, too. [1] http://en.wikipedia.org/wiki/Operant_conditioning#Consequences_that_shape_behavior:_Reinforcement.2C_punishment.2C_and_extinction http://en.wikipedia.org/wiki/Operant_conditioning#Consequenc...
- angersock 12y ago
  Thank you! People mix this up all the time, and it gives me an eye twitch. Psych 101 people, cmon.
mcv 12y ago
Well, positive reinforcement is better than negative reinforcement because instead of presenting yourself as an unpredictable angry ape, you're presenting yourself as a source of treats, which is better. Who cares why the cat is getting treats?
wuliwong 12y ago
They think that because they don't logic well. :) I thought the same thing. I hate that any comment on a HN post adds to it's score. I would like this comment counted as a negative. This article is terrible.
azernik 12y ago
The issue (which the article explained badly) isn't that punishment doesn't work; it's that cats don't understand human social cues. For a dog, yelling is a punishment on its own, because they understand what the tone of voice means. A cat will just think "hmm, the human got louder - I wonder why?" If you're trying to punish a cat for things, you either need a physical stimulus they don't like (spray bottle, for example) or a social cue they understand (hissing is moderately effective, though that's more of a warning than a punishment).

by jchendy 12y ago

cstross 12y ago

Because positive reinforcement does work on cats -- the problem is, finding the positive behavioural pattern to reinforce! Positive and negative reinforcement patterns are not opposites, they're entirely separate processes. (Cat owner here.)

readerrrr 12y ago

Perhaps because yelling isn't necessarily interpreted as something negative, while food is certainly something positive.

XorNot 12y ago

This is all wrong though. Domestic cats have been domesticated for a long time. Cats which live with humans (or even other cats - they do naturally form colonies) adopt a number of pro-social behaviours to interact with us. They are in fact very adept at reading social cues - but conversely they are not pack-animals. They don't have the same instinctive need to identify a hierarchy of command, so much as "mess with" and "don't mess with".

lingben 12y ago
Among the most common being meowing. Cats do not mew at other cats, only at humans.
puzzlingcaptcha 12y ago
>Domestic cats have been domesticated for a long time. That sounds like lamarckism. Or do you suggest that cats were selectively bred for their social behaviour? I find it dubious.
- pessimizer 12y ago
  >I find it dubious. You should check your detector.
- Sharlin 12y ago
  How do you think domestic cats came into existence? Prosocial behavior and tolerance of humans are traits that are equal parts nature and nurture in cats. Obviously current housecats are descended from those wild cats that were more accepting of human presence and had traits that made them accepted by humans as well. Still, individual cats must learn to interact with humans, and if they aren't habituated to human presence during the first couple of months of their lifes, they end up very different psychologically compared to if they are.
  - puzzlingcaptcha 12y ago
    I questioned artificial selection (breeding), you replied with natural selection (increased reproductive success by adopting to living near humans). Briefly skimming some sources, we are both correct http://www.ncbi.nlm.nih.gov/pmc/articles/PMC2702791/#__sec6title http://www.ncbi.nlm.nih.gov/pmc/articles/PMC2702791/#__sec6t...
  - warfangle 12y ago
    The domestication of dogs happened a lot longer ago, and probably a lot more organically (more like what you're thinking). The wolves would follow us around while we hunted (or, alternatively, we'd happen on one of their meals and chase them off), and the less human averse among them got more food. Cats happened about the same time agriculture did. They were attracted to the rodents, and less rodents meant we were happier - and they were happier with a reliable food source. At the beginning it might have helped to not be afraid of humans, but we probably didn't mess with them too much because they were helping us out with all the mice in our grains. So the difference in behavior is clear, not just from a social species standpoint, but from the domestication path: The European Wolf was domesticated to dogs as a hunting partner; the African Wildcat was domesticated to kill the rodents that devoured our foodstuffs. For the few few thousand years that was probably enough - they didn't have to make friends with us, just keep our food safe. Earliest cat domestication evidence is 9,500 years old; earliest dog domestication evidence is about 36,000 years old. They've still got quite a ways to go ;) Now if we could just start breeding them for social/intelligence instead of how weird and fluffy we can make them. Hard to morally advocate for breeding when the shelter/feral population is so damn high, though. (I feel the same way about dogs, though, tbh).

Proleps 12y ago

Punishment also works on cats. You just have to be very consistent when applying it (this is exactly the same for dogs). If you punish your cat(or dog) you have to do it immediately. If the cat is done scratching its nails, and you squirt water, you are too late. [Edit] Changed negative reinforcement to punishment, rbehrends was correct when he pointed out that negative reinforcement is something different.

rbehrends 12y ago

That's not negative reinforcement. Negative reinforcement is when behavior is reinforced by removing an aversive stimulus. Yelling at a cat because it's scratching the couch is positive punishment [1]. And punishment works quite differently from reinforcement to begin with; it is typically less effective for far better socialized species such as humans and dogs, too. [1] http://en.wikipedia.org/wiki/Operant_conditioning#Consequences_that_shape_behavior:_Reinforcement.2C_punishment.2C_and_extinction http://en.wikipedia.org/wiki/Operant_conditioning#Consequenc...

angersock 12y ago
Thank you! People mix this up all the time, and it gives me an eye twitch. Psych 101 people, cmon.

mcv 12y ago

Well, positive reinforcement is better than negative reinforcement because instead of presenting yourself as an unpredictable angry ape, you're presenting yourself as a source of treats, which is better. Who cares why the cat is getting treats?

wuliwong 12y ago

They think that because they don't logic well. :) I thought the same thing. I hate that any comment on a HN post adds to it's score. I would like this comment counted as a negative. This article is terrible.

azernik 12y ago

The issue (which the article explained badly) isn't that punishment doesn't work; it's that cats don't understand human social cues. For a dog, yelling is a punishment on its own, because they understand what the tone of voice means. A cat will just think "hmm, the human got louder - I wonder why?" If you're trying to punish a cat for things, you either need a physical stimulus they don't like (spray bottle, for example) or a social cue they understand (hissing is moderately effective, though that's more of a warning than a punishment).