Monday, April 4, 2011

The Smartest Dogs

Will is probably the smartest pooch that ever crossed our doorstep. I admit that I am bias, but there are so many actions I witnessed her do, too many to line up in this post, that confirm to me that she is special. Once, at the dog park, her buddy Kiwi had a stick she badly wanted. Kiwi was still a pup then, weaker and submissive, but nevertheless Will heeded the golden rule that possession equals ownership and did not force it from her. Instead, she walked a few feet away and sniffed the ground so intensely that Kiwi joined in, dropping the stick in the process, which Will instantly ran to and snatched. Another time she took ingenuity a step further: swimming side by side with another dog, she raised her paw and dunked his head under water to make him release the stick he was retrieving back to shore. Again, she grabbed it the moment her friend let go of it, swam to shore and placed it right in front of my feet – grinning I swear.
In my line of work I frequently hear tales of canine tricksters that hoodwink to get what they want. A common one is the pooch barking out the window at nothing to entice his canine cohabitant off the sofa, so he can hog the cushiest spot himself. Anecdotal evidence regarded with skepticism by science purists, but I believe because I meet many dogs with exceptionally connected brain cells – and lately more often than I used to.
It appears that dogs are smarter these days, and not just the intelligent poodles, collies and shepherds. Pure-blooded ones of all breeds, and mutts, are in Dr. Phil’s words plugged in. They are attentive, coiled for action, and learn in a flash. Perhaps, I pondered, this is an evolutionary result of humans beginning to perceive dogs differently? Not like brawny, status-challenging predators, but more… here it goes, what heresy, anthropomorphically?
What shouldn’t surprise anybody is that the smarted dogs have smart owners who understand and take an interest in their dog, and invest the time and effort it takes to achieve whatever they’re aiming for. Think about David Hartwig and his amazing Australian Cattle Dog Skidboot. Labeled by his previous owner as a problem, David saw the potential in the handsome heeler and took trick training to the 11th degree. As amazing as their acts were, Skidboot was also David’s steady companion and working partner.
And there is Bama, a search and rescue trained Labrador retriever who demonstrated incredible self-thinking. I first read about Bama in 2008, in the Fall issue of Animal Behavior. The article by Gary Wilkes discussed false positives in tracking dogs, which is when the dog indicates without actually having found anybody - fibbing so to say. That is more common than one might think, because dogs are reinforced when they locate a lost person, and if that takes too long during a difficult search they can become frustrated and signal anyway to elicit the beloved ball, or whatever else their reward is for a job done as expected. Of course, false positives are undesired. In life or death search operations the dog has to be reliable. Bama’s owner, Theresa McPherson, had a brilliant idea how to solve that problem, and it involved teaching her a new "word".
Bama already knew two signals to discriminate between having found the person alive or dead. Alive was a bark, and someone dead she marked with a quiet lay down by her handler’s feet. The new signal she was about to learn was to touch a stick on her owner’s belt to indicate that she hadn’t found anybody yet, and for which she was also rewarded. That allowed Bama to return frequently to McPherson and inform her on the status of the search, and because she was reinforced for reporting correctly, and not just for finding someone, the frustration stopped and with it the false positives.
Bama’s own brilliance became evident when, during a training exercise, she was confronted with a dilemma. Her task was to locate the scent of a dead person that was planted in a building. Bama obligingly searched, found and returned, but then apparently made a profound mistake. She lay by McPherson’s feet and barked, conflictingly telling her that she found someone who's dead and alive. Naturally, that is impossible and had McPherson puzzled, but she followed her dog to the building anyway and, through an open window, heard people talking outside. The clever Lab, when she found the dead person’s scent, obviously heard or smelled the alive humans too and accurately informed her owner that she found the dead one, but also alive ones. Reporting correctly required that she combined two separate signals into a compound word – and that’s remarkable.
The newest member of the “gifted canine club” has got to be Chaser, and if you google “Border Collie Chaser” you find a whole bunch of articles and video clips. The gist of it is that Chaser comprehends 1,022 proper nouns and 3 verbs. No kidding, and verified that it is not a Clever Hans type hoax. Chaser’s human, 82-year-old retired psychologist John W. Pilley, accomplished that by showing her an object and repeating the name up to 40 times. During daily sessions the collie learned 1-2 new words, and recapped the already known ones. Aside from the unbelievable vocabulary Chaser accumulated, how Pilley taught caught my attention.
The continuous repetition of commands, both in the learning phase and after a dog connects a certain word with a behavior or object, is a big no-no with the dog-training crowd. Immediate, one-command obedience is the mantra, and command repetition during the exercise almost blasphemous. But maybe we need to rethink that, now that Pilley demonstrated that this is exactly the way to teach a dog to comprehend many, many words.
Talking with your dog is okay? More than that, essential if you want a really smart pooch? I am strongly leaning toward it. Not ceaseless chatter; not flooding with non-stop and meaningless acoustic sound, but teaching precise verbal information the dog can use as tools to succeed; words humans use and you need to repeat patiently until the dog gets it. How many repetitions it takes depends on the dog. 40 with Chaser the Border collie; likely a few more if you own a beagle or Saint Bernard.
I never counted how many words my dogs comprehend, and never scientifically tested if they, indeed, understand the spoken word or are just excellent readers of my micro-expressions. But I am certain their intelligence goes beyond prompted actions. I believe that to be true for all dogs. If they were but cute stimulus-response machines, many of us wouldn’t love them the way we do. And if we recognize and foster cognizance some, like Bama who created a new word, and Chaser who is able to pick out an unknown object by distinguishing it from all the familiar ones, leap to a level science is just beginning to explore.

1 comment:

  1. Very neat post. I have seen luna hood wink quite a few dogs. When she really wants to keep playing keep away with a dog she works them over. Whether this means she acts disinterested in the toy so they venture over to it, only to have her dash off with it... or barks to cause them to drop it out of startle.. or the dunking... or sucking up to wyatt all flirty so he thinks she wants to play and he gets off the couch only to have her jump up and take his spot.

    She is not a good word dog, or has not been so far. But maybe she needs more repetition is all. I will work on it.
    Thanks for posting