4 ms·

I always liked the competing theory the wolves/dogs were responsible for domesticating humans. Mostly because I have seen first hand dogs train humans. For ex

I always liked the competing theory the wolves/dogs were responsible for domesticating humans. Mostly because I have seen first hand dogs train humans. For example, my parent’s Goldendoodle knocks (punches) a sliding glass door when she wants to be let in/out. No one taught her this, she did it and got the response from the human she was looking for and continues the behavior getting the same expected result from the human everytime.

elboru 8y ago
I never thought about it in that way, but my dog trained me too. When she's done with her business outside, she will scratch the carpet, I learned to open the door when I hear that noice, I never taught her that, so she trained me.
lainga 8y ago
Well, the two questions are: (1) Do dogs have more human-like behavioural traits compared to wolves? (2) Do humans have more dog-like behavioural traits compared to primates? I have always thought it was a a sort of symbiotic domestication, both happening at the same time.
mikestew 8y ago
For example, my parent’s Goldendoodle knocks (punches) a sliding glass door when she wants to be let in/out. No one taught her this Dogs do what works. You have no idea the plethora of other things the dog tried before tapping the glass. Training, OTOH, involves "I would like the humans to open the door when I tap the glass. Using positive training, I shall..." The dog would have been just as content to bark if it got the door to open. Edison didn't train a light filament to glow without burning up, his assistants kept trying until they found a method that gave the results they desired.
- asdkhadsj 8y ago
  Is there a meaningful difference between > "I would like the humans to open the door when I tap the glass." and > "I would like the humans to open the door so I'll keep trying stuff until it works" ? I say that because I feel like your description means training is a means of applying a specific input and obtaining the output. Where as, even humans often don't even train dogs - they try random inputs until they obtain the output they desire. So what would you call it if a human tried random inputs to a dog, eventually getting the desired behavior? If that's not training, what is it? Is the input just as crucial as the output to "training"?
  - mikestew 8y ago
    So what would you call it if a human tried random inputs to a dog A waste of time? When I train a dog, I want the dog to respond to a specific input, for instance saying the word "sit". I'm not going to stand there trying different words until the dog's ass hits the ground. The difference lies in that the dog cannot be said that it wanted the door opened when it tapped the glass. The dog simply wanted the door to open. Of the many things the dog (might have) tried, tapping the glass was the only thing that worked. So the dog goes with that from now on. Is the input just as crucial? Yes, it is. If for no other reason than a standardization of language. Because teaching your two year old to just use whatever random word she makes up for what the rest of us call "strawberry" isn't going to get her very far in life. In a similar manner, standardization of language with dogs is just as essential. I've got two dogs, you think I'm learning a different language for each? Duh, no. Consistency of input/desired-output mappings is a key element to training in general.
    - asdkhadsj 8y ago
      > A waste of time? When I train a dog, I want the dog to respond to a specific input, for instance saying the word "sit". I'm not going to stand there trying different words until the dog's ass hits the ground. Agreed, it's far from efficient - but it's what I've seen a ton of people do. Throw random inputs at the dog to try to get them to go pee, or go outside when peeing, or sit on command, or etc. > The difference lies in that the dog cannot be said that it wanted the door opened when it tapped the glass. The dog simply wanted the door to open. Of the many things the dog (might have) tried, tapping the glass was the only thing that worked. So the dog goes with that from now on. My question was mainly centered around the idea that, the way you describe it, I think is the same that a lot of people use in "training" dogs. They don't care what the input is, they care what the output is. Even during simple things like Sit. They don't pay mind to all the random variables they're throwing in like body language, tone, gestures, food, etc. > Is the input just as crucial? Yes, it is. If for no other reason than a standardization of language. Because teaching your two year old to just use whatever random word she makes up for what the rest of us call "strawberry" isn't going to get her very far in life. In a similar manner, standardization of language with dogs is just as essential. I've got two dogs, you think I'm learning a different language for each? Duh, no. Consistency of input/desired-output mappings is a key element to training in general. Well I think you misunderstand me. I'm not asking if input matters, I'm asking if input matters to your definition of "training". Ie, if a dog tries random inputs and eventually figures out a way to get me to let it outside - it sounds like you did not consider that training. Yet, I feel like frequently people do the same to "train" dogs. Hence my question towards you.

elboru 8y ago

I never thought about it in that way, but my dog trained me too. When she's done with her business outside, she will scratch the carpet, I learned to open the door when I hear that noice, I never taught her that, so she trained me.

lainga 8y ago

Well, the two questions are: (1) Do dogs have more human-like behavioural traits compared to wolves? (2) Do humans have more dog-like behavioural traits compared to primates? I have always thought it was a a sort of symbiotic domestication, both happening at the same time.

mikestew 8y ago

For example, my parent’s Goldendoodle knocks (punches) a sliding glass door when she wants to be let in/out. No one taught her this Dogs do what works. You have no idea the plethora of other things the dog tried before tapping the glass. Training, OTOH, involves "I would like the humans to open the door when I tap the glass. Using positive training, I shall..." The dog would have been just as content to bark if it got the door to open. Edison didn't train a light filament to glow without burning up, his assistants kept trying until they found a method that gave the results they desired.

asdkhadsj 8y ago
Is there a meaningful difference between > "I would like the humans to open the door when I tap the glass." and > "I would like the humans to open the door so I'll keep trying stuff until it works" ? I say that because I feel like your description means training is a means of applying a specific input and obtaining the output. Where as, even humans often don't even train dogs - they try random inputs until they obtain the output they desire. So what would you call it if a human tried random inputs to a dog, eventually getting the desired behavior? If that's not training, what is it? Is the input just as crucial as the output to "training"?
- mikestew 8y ago
  So what would you call it if a human tried random inputs to a dog A waste of time? When I train a dog, I want the dog to respond to a specific input, for instance saying the word "sit". I'm not going to stand there trying different words until the dog's ass hits the ground. The difference lies in that the dog cannot be said that it wanted the door opened when it tapped the glass. The dog simply wanted the door to open. Of the many things the dog (might have) tried, tapping the glass was the only thing that worked. So the dog goes with that from now on. Is the input just as crucial? Yes, it is. If for no other reason than a standardization of language. Because teaching your two year old to just use whatever random word she makes up for what the rest of us call "strawberry" isn't going to get her very far in life. In a similar manner, standardization of language with dogs is just as essential. I've got two dogs, you think I'm learning a different language for each? Duh, no. Consistency of input/desired-output mappings is a key element to training in general.
  - asdkhadsj 8y ago
    > A waste of time? When I train a dog, I want the dog to respond to a specific input, for instance saying the word "sit". I'm not going to stand there trying different words until the dog's ass hits the ground. Agreed, it's far from efficient - but it's what I've seen a ton of people do. Throw random inputs at the dog to try to get them to go pee, or go outside when peeing, or sit on command, or etc. > The difference lies in that the dog cannot be said that it wanted the door opened when it tapped the glass. The dog simply wanted the door to open. Of the many things the dog (might have) tried, tapping the glass was the only thing that worked. So the dog goes with that from now on. My question was mainly centered around the idea that, the way you describe it, I think is the same that a lot of people use in "training" dogs. They don't care what the input is, they care what the output is. Even during simple things like Sit. They don't pay mind to all the random variables they're throwing in like body language, tone, gestures, food, etc. > Is the input just as crucial? Yes, it is. If for no other reason than a standardization of language. Because teaching your two year old to just use whatever random word she makes up for what the rest of us call "strawberry" isn't going to get her very far in life. In a similar manner, standardization of language with dogs is just as essential. I've got two dogs, you think I'm learning a different language for each? Duh, no. Consistency of input/desired-output mappings is a key element to training in general. Well I think you misunderstand me. I'm not asking if input matters, I'm asking if input matters to your definition of "training". Ie, if a dog tries random inputs and eventually figures out a way to get me to let it outside - it sounds like you did not consider that training. Yet, I feel like frequently people do the same to "train" dogs. Hence my question towards you.