Dog (Canis
lupus familiaris) domestication and their mutually beneficial relationship
with humans is one of the oldest; working, hunting and living together for over
10 000 years. During that time dogs have been selected for a wide range of physical
and behavioural traits such as protection,
hunting, guidance and more recently appearance and responsiveness to
verbal/physical cues. Controllability and docility in temperament
is assumed to be an early selection factor in the original human adoption of
the ancestral dog and remains a pivotal character throughout the domestication process. Frank (1980) looked into the underlying
differences between learning techniques favoured under conditions of natural
and artificial selection whilst questioning the notion of animal intelligence.
Dogs originated
from the grey wolf (Canis lupus) which relies on primitive instincts in response
to basic survival needs as well as a cognitive system that evolved in correlation
with group hunting. Wolf behaviour is consistent in response
to very specific stimulants. When comparing learning ability between wild
wolves and domesticated dogs, wolves excel at observational learning, which in
a natural setting add to their survival by being able to problem solve. However
domesticated dogs excel at trainability or instrumental conditioning which is learning
by extrinsic reinforcement in a human environment. The animal learns a response to a command
that has no discernable functional connection with the feedback received. For
example: a dog is trained to sit; when it completes the required response (sitting)
from the cue (vocal command) it receives food (extrinsic
reinforcement/feedback). In nature, a wolf does not receive food for ‘sitting’
therefore it is not bettering its survival to obey commands with no obvious function or beneficial outcome.
Image courtesy of: http://www.shardabakersdogworld.com/potty-training/dog-potty-training/
In the process of domestication it appears
that the wolf's capacity for observational learning and automated
instinctual responses has been superseded by the capacity for trainability.
Artificial selection favoured docility and human intervention between the dog
and the consequences of his behaviour have relaxed those pressures favouring capacity
for insight. Now modern breeds have acquired a tolerance for boredom, human socialization, a reduced wariness of environmental and social
stimuli and are highly dependent upon their human caregivers.
It
could be argued that in training our dogs to be what we consider ‘smarter’ (by completing
cue-response tasks) we have dulled their capacity to problem-solve independently
and they now lack the strong ability to learn on their own through observation
and trial and error as do their wild counterparts.
 | |
| Comic courtesy of: http://jsburner.edublogs.org/2010/08/04/neolitic-revolution/ |
Awesome post Elsea, I was wondering do you see an increase problem solving in wild dogs? Does the absence of human attention increase their ability to fend for themselves?
ReplyDeleteI would believe that wild dogs (feral/strays) have a much better chance at survival once they'd learnt to fend for themselves and they 're-developed' their own hunting/scavenging techniques so they no longer are dependent on humans. However I don't think that there would be a great change in their problem solving/way of learning as domestication has structured them specifically for human-dog interactions and to respond to a partner. So going 'solo' and surviving as a stray/wild dog would require some sort of partnership (with other dogs etc) to be successful.
DeleteA fascinating, well written post. Do you known if anyone looked at the brain structure of domestic dogs and wolves and compared neural firing patterns during different cognitive tasks? This would be an interesting way of assessing whether cognitive abilities are different (more or less).
ReplyDeleteHey Tasmin, most papers are done on behavioural testing and I couldn't find anything on differences of brain structure. I would expect there to be a difference however it would be hard to test as many variables may affect the response (if one is even received) with wolves, such as introducing human interaction in the first place. If research were to be done in this area in more depth I would be very interested in finding out how they designed the cognitive tasks to be suitable to both dogs and wolves to produce measurable results.
Delete