Dogs learn by error and success. They are also opportunists and constantly calculate what is the best possible option for them to behave in the situation.
We humans like to rationalise everything. Our way of thinking is much more complicated than dogs’. When we see something happening, we try to explain it and understand why it happened. When we interact with dogs, we like to interpret their behaviour the way we think another human would act in that same situation.
For instance – A dog has broken or chewed on something during a work day. The owner arrives home, sees the mess and gets upset. Often the dogs reaction to this is to keep some distance to the owner, lower it’s head, move it’s ears back and avoid owners eye contact. Why does the dog react like that? Is it feeling ashamed? Is it being afraid? Did it do the damage as a revenge, because the owner left for work? Or is it reacting to it’s owners radical mood change?
In a book “Millaista on olla eläin?” (2015, Eng. How is it like to be an animal?). Helena Telkänranta, biologist and a science writer, introduces some latest studies conducted on dogs’ feelings and their ability to think and reason. These studies show that dogs are extremely talented in reading body language. Even so talented, that it’s tempting for us humans to think, that they can even read our thoughts.
Like humans, dogs are also capable of feeling fear and worry. They are. however, not capable of thinking from other individuals point of view, which is essential for one to feel ashamed. Dogs’ genes are built to make it react with specific gestures when they get afraid or worried. This is why they look worried, even when they have no clue what to be worried about.
Dogs are often misinterpreted in the training field. People say that “It (the dog) knows what I’m asking, but it just won’t do it!” and then they punish the dog for “not obeying”. So the person interprets, that the dog knows what to do, but it just doesn’t feel like doing it. What is the truth? How can we find out if she’s interpreting right what we wish for? Does the dog really know what it’s being asked for?
Sometimes the dog really just doesn’t feel like doing what it’s told, but usually there is something that the handler misses.
Has the dog done it before in the same environment with all the same distractions? Has something changed after that? Is the dog physically fine, or could it be sick or not feeling well? Is the reinforcement good enough compared to the reinforcements environment offers?
In scent work the interpreting seems to be even more in common. Probably this is mainly because we can’t experience the scent world ourselves and we need to use our imagination to fill those parts in exercises that we can not see or otherwise experience. As handlers we of course want our dogs to succeed and find the target odour that it’s suppose to find. If the dog finds the target odour, we get to reward the dog and feel good about our successful work with the dog. It is, after all, usually a result of long time of training and hard work. The dog has learned to search and find!
But how can we know for sure, that the dog knows what it’s doing? It looks like it’s really efficiently searching and finding, but how can we be sure, that it’s not a case of misinterpretion?
A classic example for the impact of body language on the animal behaviour is a story about the “Clever Hans”. Clever Hans was a horse, that appeared to know how to answer highly intellectual mathematical questions by tapping his hoof. No matter how hard the question was, Hans nearly always got it right. The skill that Hans had was so remarkable that it soon became famous nation wide. It wasn’t until psychologist Oscar Pfungst did a formal investigation about his skills that they realised, that the skill was not about the ability of Hans to do maths and all but to read the body language of the audience.
After knowing about Clever Hans in the beginning of 20th century, what do you think about this extraordinary jack russell terrier on BBC Earth in the 21st century?
When training scent work alone, the trainer always knows where the target odour is hidden. As long as the dog is able to see or sense the handler who knows where the odour is, there’s a possibility that the dog reads the handlers body language and does the search based on that. Dogs pick up small gestures like the direction of handlers leers, movement of eye brows, breathing tempo, the direction handler moves to or stands towards to, etc.
What if it would be done as a blind search, so that the handler doesn’t know the hide?
If someone else makes the hides, so that the handler doesn’t know where they are, but is still present while the dog is searching, the effect is the same. As in Clever Hans’ case, dogs can also read other peoples body language, not just their owners or handlers. As we humans tend to think we know better than the dog, we try to rationalise the hides also based on where we assume the other person must have put them. This leads to handler directing the dog to those assumed places or directing it off those places we think the hide just could not be. And again the handler interferes the dogs work.
So what we have left, is a training situation called a double blind search. That means, that someone makes the hides, but does not tell the position of the hides to the handler nor is present when the dog searches. If no-one in the search area knows the hide, the dog needs to use it’s nose to find it and can’t get any help from people.
Here comes the most interesting part: does the dog find it then or not? There has been many dogs that have fooled their handlers, or in fact, have been extremely talented reading their or others body language and played a very nice role as a working detection dog. Washington Post wrote a worrying story about drug detection dogs that alerted based on their handlers suspicions in last August (2015). You can read the article here. The fact that even highly trainers police dogs can be affected by they handlers this severely gives us a good idea what kinds of mistakes other dog handlers can make.
At this point it’s fair to say, that double blinds are a tool to train a good and reliable scent detection dog. It’s not something you should do always and every time. There’s a lot more to it. Double blind searches are a way to see what point the training is in. And a way to gain trust to the dog, when it succeeds. When the dog is ready for work and the basic training is done, then the double blinds are a way to measure it’s level of proficiency.
Scent work is a difficult sport. There are many things that can go wrong without us noticing it before we already have a problem. One of them, is the problem with human reading dogs. As said before, we as humans, can not fully understand scent work. And that is the main reason, why we need to test our dogs with double blind searches regularly. That is the only way to find out what they really know: to search and find or to read human body language.
This post is also available in: Suomi (Finnish)