|
| Author |
Messages |
|
cwalt
MH
Posts:180
 |
| 08/16/2008 3:48 PM |
|
| Natural populations follow Bergmann's rule, animals that live closer to
the poles are larger. The largest wolves are arctic specimens that can
easily reach 100 kg in weight and are fearsome predators of large
ungulates such as moose. Wolf species closer to the equator are smaller
and more typically like the coyote and are fearsome predators of
rodents. Larger animals have bigger brains and so it is difficult to
compare brain weights between a Canadian wolf and a Gordon setter for
any meaningful answers to which animal is brighter. Both dogs and
wolves depend a great deal on innate behaviors and fixed action patterns
for a lot of their observable behaviors. The entire panoply of body
language signals partially shared between wolves and dogs are individual
innate behaviors that can be expressed singly or in simultaneous
groups. The elevation of the tail, as an index of emotional state isn't
so much under conscious control of the dog as it is simply an innate
behavior reflecting the dog's internal state. These innate patterns are
so consistent that with study a human can almost instantly recognize the
emotional state of a dog and predict with great confidence the animal's
next behavior. We aren't quite as certain of the emotional state of
wolves, partly because of unfamiliarity and partly because the wolf body
language may well exhibit fewer innate behavior patterns.
Almost every behavior of a bird, such as the graylag goose, is a series
of innate behaviors that comprise the bird's ethogram, the sum total of
behaviors the animal can exhibit. If the behavior isn't programmed the
goose doesn't do it. Innate behaviors are shaped by evolution and save
the birds from having to make conscious decisions about everything,
things which could make birds slower and more susceptible to fatal
accidents. Dogs are not so constrained and they originate a lot of
behaviors that are not preprogrammed. Compared to man, however, dog
behavior is dominated by innate motor patterns. A lot of these innate
motor patterns are more complex than simple reflexes and we describe
these as fixed action patterns. A dog teaches a young boy to be
faithful, to love and to turn around three times before lying down. The
turning around before lying down is a fixed action pattern triggered by
an internal state, the intention to lie down. Pointing is another fixed
action pattern but this is triggered by an external source, the sight
or scent of a bird. It is important for the bird hunter to realize
that pointing is between the bird and the dog and humans have nothing
to do with it. Pointing cannot be taught, it is built into the dog's
psyche. We also have to realize that some complex behavior patterns are
learned, response to the whoa or halt command is such a learned
behavior. Realize that pointing and the halt are totally different
behaviors triggered by different releasers that are unrelated. The whoa
is not pointing and cannot ever be pointing because it is an interaction
between a human and the dog and has nothing to do with the bird. It is
quite common in North America for dog trainers to assume that pointing
can be trained and they induce a great many problem behaviors because of
this mistaken belief.
Some complex canine behavior patterns result from totally different and
unexpected releasers; whenever there is an emotional conflict
displacement behaviors may arise. The most startling feature of
displacement behaviors is that they are completely unrelated to the
factors that trigger them. When commanded to retrieve a dummy from
water a dog instead drinks water every time the command is given. This
is called displacement drinking the dog is fearful about the water and
cannot enter it so it does something else. In similar circumstances a
dog might run up and down the shore and seize a stick or clump of grass
and bring it to the handler. In birds, such as seagulls, an innate
fight or flee conflict can lead to grass pulling, an unrelated
displacement behavior that relieves the emotional tension. Dogs are
capable of learning far more things than geese or seagulls and increased
experience can lead to a "smart" dog. This doesn't mean that the animal
is intelligent, it does mean that the animal is capable of learning a
lot. Both wolves and dogs are smart animals but they are smart about
different things.
Cj |
|
|
|
|
jmurr
MH
Posts:158
|
| 08/16/2008 5:07 PM |
|
| Nice work, Clem but some comments are hurriedly presented.
On brain size animal size dog vs wolf.
From Coppinger previously cited, page 55
Wolfhound Wolf Anatolian
Body weight (lbs) 150 100 180
Skull size (cm²) 348 352 343
Brain volume (cm³) 303 329 312
Dogs 50% to 80% larger than the wolf have heads of approximately the same size but
brains 5% to 8% smaller. On page 56 they present a graph and state: "...Big dogs
have the same growth trajectories as wolves, but their brain growth slows and their
brains don't [ever] get much bigger than those of four-month-old wolf puppies."
Your ending: "Dogs are capable of learning far more things than geese or seagulls
and increased experience can lead to a "smart" dog. This doesn't mean that the
animal is intelligent, it does mean that the animal is capable of learning a lot."
Doesn't this statement contradict the commonly used concept of intelligence? Our
efforts to define a measure of human intelligence typically measure scope and depth
of learning (acquisition of information etc) over a wide range of information at
given ages and compare these scores within the population of that age, assigning
high IQ to the upper reaches of the distribution and low to the lower reaches - or
something like that.
"Both wolves and dogs are smart animals but they are smart about different things."
Sure, but that's why I deliberatly mentioned observations of wolves and dogs
confined to cages and their relative success at devining escape methods when I
tried to give an example of differing intelligence. Here is the wolf readily
solving a problem it has never seen before while the dog fails in spite of frequent
challnge.
I think you over state the case wrt wolf body language. Their body languge may be
quite clear but the triggers for certain behaviors are certainly more touchy.
There are some things one is best advised NOT to do near or to a wolf which s/he
could do with impunity around most dogs.
Now you've worked on the innate behaviors; IAPs and FAPs, what about those trained
behaviors with no counterpart in the repertoire of innate behaviors? What is the
neurophysiological difference, if any, between them and the innate behaviors? I
don't have a clue myself but wonder whether their frequent need for reinforcement
through touch up training to prevent fadeing with time could be a distinguihing
characteristic and a clue to what may be different in the brain circuits which
control them. Muddy enough?
Jere
> Natural populations follow Bergmann's rule, animals that live closer to
> the poles are larger. The largest wolves are arctic specimens that can
> easily reach 100 kg in weight and are fearsome predators of large
> ungulates such as moose. Wolf species closer to the equator are smaller
> and more typically like the coyote and are fearsome predators of
> rodents. Larger animals have bigger brains and so it is difficult to
> compare brain weights between a Canadian wolf and a Gordon setter for
> any meaningful answers to which animal is brighter. Both dogs and
> wolves depend a great deal on innate behaviors and fixed action patterns
> for a lot of their observable behaviors. The entire panoply of body
> language signals partially shared between wolves and dogs are individual
> innate behaviors that can be expressed singly or in simultaneous
> groups. The elevation of the tail, as an index of emotional state isn't
> so much under conscious control of the dog as it is simply an innate
> behavior reflecting the dog's internal state. These innate patterns are
> so consistent that with study a human can almost instantly recognize the
> emotional state of a dog and predict with great confidence the animal's
> next behavior. We aren't quite as certain of the emotional state of
> wolves, partly because of unfamiliarity and partly because the wolf body
> language may well exhibit fewer innate behavior patterns.
>
> Almost every behavior of a bird, such as the graylag goose, is a series
> of innate behaviors that comprise the bird's ethogram, the sum total of
> behaviors the animal can exhibit. If the behavior isn't programmed the
> goose doesn't do it. Innate behaviors are shaped by evolution and save
> the birds from having to make conscious decisions about everything,
> things which could make birds slower and more susceptible to fatal
> accidents. Dogs are not so constrained and they originate a lot of
> behaviors that are not preprogrammed. Compared to man, however, dog
> behavior is dominated by innate motor patterns. A lot of these innate
> motor patterns are more complex than simple reflexes and we describe
> these as fixed action patterns. A dog teaches a young boy to be
> faithful, to love and to turn around three times before lying down. The
> turning around before lying down is a fixed action pattern triggered by
> an internal state, the intention to lie down. Pointing is another fixed
> action pattern but this is triggered by an external source, the sight
> or scent of a bird. It is important for the bird hunter to realize
> that pointing is between the bird and the dog and humans have nothing
> to do with it. Pointing cannot be taught, it is built into the dog's
> psyche. We also have to realize that some complex behavior patterns are
> learned, response to the whoa or halt command is such a learned
> behavior. Realize that pointing and the halt are totally different
> behaviors triggered by different releasers that are unrelated. The whoa
> is not pointing and cannot ever be pointing because it is an interaction
> between a human and the dog and has nothing to do with the bird. It is
> quite common in North America for dog trainers to assume that pointing
> can be trained and they induce a great many problem behaviors because of
> this mistaken belief.
>
> Some complex canine behavior patterns result from totally different and
> unexpected releasers; whenever there is an emotional conflict
> displacement behaviors may arise. The most startling feature of
> displacement behaviors is that they are completely unrelated to the
> factors that trigger them. When commanded to retrieve a dummy from
> water a dog instead drinks water every time the command is given. This
> is called displacement drinking the dog is fearful about the water and
> cannot enter it so it does something else. In similar circumstances a
> dog might run up and down the shore and seize a stick or clump of grass
> and bring it to the handler. In birds, such as seagulls, an innate
> fight or flee conflict can lead to grass pulling, an unrelated
> displacement behavior that relieves the emotional tension. Dogs are
> capable of learning far more things than geese or seagulls and increased
> experience can lead to a "smart" dog. This doesn't mean that the animal
> is intelligent, it does mean that the animal is capable of learning a
> lot. Both wolves and dogs are smart animals but they are smart about
> different things.
> Cj
>
|
|
|
|
|
cwalt
MH
Posts:180
|
| 08/16/2008 6:40 PM |
|
| Jere Murray wrote:
> Dogs 50% to 80% larger than the wolf have heads of approximately the same size but
> brains 5% to 8% smaller. On page 56 they present a graph and state: "...Big dogs
> have the same growth trajectories as wolves, but their brain growth slows and their
> brains don't [ever] get much bigger than those of four-month-old wolf puppies."
>
>
~~~~~~~~~~~~~~~~~~~~~
Interesting data and of course we know that there are significant
differences between skulls of dogs and wolves. Even wolf-like dogs can
be separated from wolves by skull morphology because there is so much
difference between the species. Yet there are no reliable data to
suggest that larger brains are more intelligent than smaller ones.
Porpoises have bigger brains than humans but are they more intelligent?
Perhaps they are.
Cj
~~~~~~~~~~~~~~~~~
> Your ending: "Dogs are capable of learning far more things than geese or seagulls
> and increased experience can lead to a "smart" dog. This doesn't mean that the
> animal is intelligent, it does mean that the animal is capable of learning a lot."
>
> Doesn't this statement contradict the commonly used concept of intelligence?
~~~~~~~~~~~~~
No, I don't think so. Data storage doesn't necessarily mean that the
animal can put the parts together to arrive at novel conclusions. We
have savants who can memorize telephone books but they cannot do
anything else in the intellectual sphere. You may have remembered from
your last IO test that there were a lot of questions that asked you to
put together things you haven't learned and animal intelligence is all
about putting things together to derive new meanings or solutions. Some
tests have shown that wolves are considerably smarter than dogs at
solving "how do I get the food?" problems and resolving spatial problems
(heading off a moose). It is easily possible to teach a dog a hundred
or more commands, I never could get my children to obey more than two
or three commands. That begs the question of which is smarter, the
father or the offspring.
Cj
~~~~~~~~~~~~~~
... I deliberately mentioned observations of wolves and dogs
confined to cages and their relative success at divining escape methods
when I
~~~~~~~~~~~~~~~~~~
I have had several dogs that learned, without human intervention, to
open doors by turning the door knob and I have had a wirehair bitch that
was better than a raccoon at undoing the latch on her dog box. There
are pluses on both sides of the issue. I term this observational
learning and there have been some recent experiments that show
conclusively that dogs can can think and interpret some things. So far
most of these thinking results have been associated with dogs learning
from the body language of other dogs and putting things together
without being shown what to do. Dogs can learn to think about the
body language of other dogs but in experiments do not appear to think
about human body language, a conclusion that raises doubt in my mind.
Cj
~~~~~~~~~~~~~~~~~~~~~~~~~
> Now you've worked on the innate behaviors; IAPs and FAPs, what about those trained
> behaviors with no counterpart in the repertoire of innate behaviors? What is the
> neurophysiological difference, if any, between them and the innate behaviors? I
> don't have a clue myself but wonder whether their frequent need for reinforcement
> through touch up training to prevent fadeing with time could be a distinguihing
> characteristic and a clue to what may be different in the brain circuits which
> control them. Muddy enough?
>
> Jere
>
>
~~~~~~~~~~~~~~~~~~~~~~~~~
Well I haven't worked through very much on IAPs and FAPs, we could go on
for hours about those and related ethological issues. There is quite a
difference between learned and innate action patterns in terms of what
part of the brain is used and how fixed the behavior might be. Simple
action patterns such as whoa can, and should, be trained until they are
reflexive. If you give a command and the dog hesitates it means that
the dog is 'thinking' about it. Simple direct one action commands whoa,
sit, and so on can be trained to be so reflexive that they simulate
fixed action patterns and are not readily distinguishable from innate
behaviors. Other commands should never be trained as reflexive, these
include the recall in which you want the dog to thin k before
responding. Calling a dog off a point isn't as serious as calling a dog
and having it return across a busy highway without hesitation. Chained
trained sequences require more thought since they are always taught, and
expressed, as sequences. Because of this they are rarely, if ever, as
smoothly executed as fixed action patterns. Trained retrieving
sequences are often chained and always exhibit a fraction of a second
lag in sequential execution since movement a triggers movement b and so on.
The first action in a chained sequence can be drilled so it is very fast
in response to a command since your body language telegraphs your
intent. With intermittent reinforcement chained sequences can be
recalled and repeated so long as the dog can recognize your signals. I
find that the closest learned responses to training can closely mimic
innate responses when they're cued by your body language, dogs can read
human intention movements faster than they can respond to an auditory
or visual signal. That response to a body language signal is faster
than a response to a hand or other visual signal is explained by the
fact that responses to human body language and dog body language are
largely innate and can always proceed faster than signals requiring some
consideration.
Some chained sequences are drilled until they become almost reflexive,
this is commonly observed in FT retrieving where the handlers don't want
the dog to think about anything but just do it the way they were
trained. This can be relatively fast but it might not be efficient or
wise. Every year veterinarians in my area treat retrievers that have
been impaled on sticks, branches or stumps while leaping into the water
to start a trained retrieve.
Cj
|
|
|
|
|
rospigan
MH
Posts:372
|
| 08/18/2008 3:25 AM |
|
|
>>>>Yet there are no reliable data to
suggest that
larger brains are more intelligent than smaller ones.
Porpoises have
bigger brains than humans but are they more intelligent?
Perhaps they
are.
Cj>>>
"I think, therefore I exist". I do not remember who
said this in ancient times. I do not know much about this phenomena
"intelligence" from a scientific point of view but I have been "thinking" about
it now and then.
One of the virtual experiments that I make in my
mind is to put a new born human baby among apes to be raised by them. I could of
course "raise" the kid in a totally sterile environment but I do not like the
idea since it somehow feels very cruel. So my baby's brain will get a lot
of stimuli but of a much different kind from what human baby's usually get.
The inherited pre-requisite for becoming or being
intelligent is there - the human brain. We know that our experimental brain
could, if developed and trained in a human environment, become at least as
intelligent as the average human brain is.
I believe that we all know that the environment
will to some degree guide the development of the infant brain. I also believe
that there is some kind of a program in the infant brain that governs in what
order and what pace things will develop in the brain.
We have in earlier discussions come to the
agreement that a dogs working ability mainly consist of two factors: heredity
and environment. This is true with humans also.
We said that the infant brain has the heredity to
become as bright as the average human brain - provided the environment is
favourable for it to develop into a human brain that has to solve human
problems.
What happens if the infant human brain develops in
an environment controlled by apes? The environment will affect the physical
development of the brain so that the brain will function smoothly in the ape
environment. As Cj mentions dogs respond faster to body language that to spoken
commands. Perhaps apes do this too. In that case will the infant brain for
example develop into a direction to respond faster to body language than to
sounds? Will this development be irreversible?
You know what I mean?
Now, if things will develop in this way - I mean
the brain develops to suit the apes environment smoothly - how can we possibly
measure the intelligence of it at a certain age. To start with we can not
communicate with it since we do not have a common language. If we still could
communicate we would not have the same values and view of life. Perhaps we would
try to measure the brains problem solving ability by subjecting it to the same
problems as we do with apes, namely "how do I get to this banana?". Maybe we
forgot that it is a human brain with a heredity for intelligence?
Now back to dogs. Take a puppy and put it in a
kennel and leave it there for 1 - 2 years without any stimuli. We
have now created a "stupid" dog, not capable of learning anything.
Take another puppy and let it live and develop in a
(dog-) active family and let it follow them everywhere for a couple of years. We
have now created a very smart - intelligent - dog.
There is without doubt a hereditary limit for how
much a certain kind of brain, and an individual brain of any kind, can "learn to
learn" from the favourable environment. There is a definite
hereditary limit for dogs, porpoises as well as humans. However it seems
like the brains ability to reach as close to this definite limit as possible
will to a dominating degree depend of the environment it was develop in from the
start.
Well, what was I trying to prove? If nothing else I
have proved how very mixed up and confused I am when facing the
subject!
Torsti
Borta Med Vindens Kennel "Ask not what your dog can do for you. Ask
what you can do for your dog." www.rospigan.net |
|
|
|
|
mcotton
MH
Posts:87
|
| 08/19/2008 2:27 AM |
|
|
The problem to me seems to be we measure
intelligence of animals by comparing them to human intelligence.
Dogs are intelligent in their own way.
I believe that the biggest difference between us
and animals is our ability to
put the past future and present into one
package. Does that make sense?
Nevermind, I'm on my 3rd wine
|
|
|
|
|
jmurr
MH
Posts:158
|
| 08/19/2008 11:31 AM |
|
| Cj said:
> There is quite a
> difference between learned and innate action patterns in terms of what
> part of the brain is used
Is there? Can you be more specific? The input (sensory, detection and processing)
aparatus would appear to be identical, as would the output (motor control)
apparatus. Where does the difference lie? Is it in the specific areas of the
brain used, the wiring details or extent of wiring there or are some areas left out
of the mix in learned patterns vs innate?
> and how fixed the behavior might be.
Is there a fundamental limit to the level of fixation inherent in the processes
whereby learned behaviors are conditioned or fixed? The conventional "wisdom"
states that conditioned behaviors fade or extinguish over time if they are not
reinforced (via additional conditioning exercises). Do some innate behaviors do so
also (outside the fact that some are programmed to end at certain stages of life)?
> Simple
> action patterns such as whoa can, and should, be trained until they are
> reflexive.
And when you have achieved this or seen it at the highest level of your experience
was it then resilient over time - not subject to the above mentioned fading w/o
reinforcement?
Along this line may I suggest it is possible, with a suitably bred and motivated
dog, to arrange training for some behaviors, including the "whoa," in such a manner
that they become self-reinforcing in the everyday experience of the dog and
therefore should not fade. But this would be another topic.
Jere
|
|
|
|
|
cwalt
MH
Posts:180
|
| 08/20/2008 4:11 PM |
|
| Jere Murray wrote:
> Cj said:
>
>> There is quite a
>> difference between learned and innate action patterns in terms of what
>> part of the brain is used
>>
>
> Is there? Can you be more specific? The input (sensory, detection and processing)
> aparatus would appear to be identical, as would the output (motor control)
> apparatus. Where does the difference lie? Is it in the specific areas of the
> brain used, the wiring details or extent of wiring there or are some areas left out
> of the mix in learned patterns vs innate?
> ~~~~~~~~~~~~~~~~~~~~~~~
>
Don't know about wolves but I would hazard an educated guess for dogs.
It would appear that innate behaviors originate deeper in the older
parts of the brain whereas learned behaviors reside largely in the
cerebral cortex, the outer surface of the brain. An exemplary case is
the snap reflex that is triggered by hissing noises originating near a
dog's head. Even under anesthesia a dog will move the head and snap at
a hissing noise, a more basic and primitive reflexive action.
> ... The conventional "wisdom"
> states that conditioned behaviors fade or extinguish over time if they are not
> reinforced (via additional conditioning exercises). Do some innate behaviors do so
> also (outside the fact that some are programmed to end at certain stages of life)?
>
> ~~~~~~~~~~~~~~~~~~~~~~~~
I have trained a number of dogs to the reflexive response level for whoa
and the response to the command or its alternative signals remains
active into senility if the commands are exercised regularly. The first
reflexive response to fade is usually the visual one to the hand signal
simply because vision declines in older dogs.. If a dog becomes deaf of
course the verbal or other sound signal releasers decline earlier in
life. I believe that, if trained properly and used consistently most
simple direct commands become self-rewarding and are preserved if
exercised. Alternate or secondary signals will tend to yield confused
responses as the dog ages and they do not seem to last as long as the
primary releasers, normally as dog and handler age the secondary learned
signals tend to be used less by the handler. In my limited experience
learned commands must be exercised if they are to be maintained. This
is another variable that depends a lot on the trainer/handler's body
language since responsive behaviors by a well trained dog are reinforced
by the handler's body language. Note that commands trained by a third
party tend to fade faster with disuse than those trained by the dog's
pack leader partly because of the differences in the handler's body
language. |
|
|
|
|
|
| You are not authorized to post a reply. |
|
|
|
ActiveForums 3.7
|
You must be logged in to use this module.
|
 |
Membership: |
 |
Latest:
ADAM/REMMY |
 |
New Today:
1 |
 |
New Yesterday:
1 |
 |
Overall:
3210 |
 |
People Online: |
 |
Visitors:
86 |
 |
Members:
0 |
 |
Total:
86 |
Online Now:
|
|
|
|
|