Re: Do dogs have a perception of time?

Your ten-year-old is asking a sophisticated question that is not so easy to 
answer! I could only find a very few, very technical scientific papers that 
have some relevance to time perception in dogs (I'm not posting them here 
because I don't think they'll be very helpful to you or your son). There 
are a number of things that make this kind of question difficult to study. 
First, since a concept of time is not a thing you can observe, you can't 
measure it directly. Second, you can't get at it indirectly by asking your 
dogs questions the way you could with a person. So you have to define some 
behavior or behaviors you *can* observe and measure directly that you 
think are a result of, or related to, having a concept of time. To identify 
this behavior, you have to know some things about your dogs. 

Maybe you've noticed that your dogs are excited to see you (that 
is, they jump all over you, wag there tails, lick your face) whenever you 
come home, but they seem more excited after you've been gone longer. 
So you might want to measure how long they wag their tails, and/or 
bark, or whatever they do when you come home. These measures are 
called dependent variables in scientific research. The independent 
variable, in this case, would be the amount of time you are away. In 
behavioral research, we often set up experiments to see whether the 
independent variable affects the dependent variable. 

Your idea of having him leave and return at different intervals is one way 
you could your dogs' ability to discriminate between shorter and longer 
durations, is you do it carefully. By doing it carefully, I mean you'd have 
to try to keep everything constant except the time intervals from one trial 
to the next, so you can minimize other things that might be influencing how 
the dogs react when you return. For example, how do you know they're not 
just hungrier because you've been gone longer? Or maybe they really need to 
go outside and relieve themselves! So you would want to start at the same 
time each day (preferably when the dogs aren't hungry or napping)and then 
increase the duration away from one day to the next. Also, when you return 
you should keep your behavior consistent, because how long the dogs jump 
all over you could have something to do with how you react to that 
behavior.

   In a controlled laboratory, it would be possible to set up an experiment 
that would give you more clear behavior to measure, and would therefore 
give you clearer results. But I'm sure a laboratory approach is more 
involved and complicated to do at home than you really want to attempt. 
Here's a quick example of the sort of thing I might try in a laboratory. 
You have a lever the dog can push to get food. The dog can only get food if 
he pushes the lever after he hears the sound of a bell. BUT if he pushes 
the lever right after the bell, he doesn't get any food. He has to learn to 
wait 10 seconds after the bell goes off to get the food. It may take  him a 
few tries to learn, but if he can learn the association between the sound, 
the 10-second delay, and getting food, then he has some sense of time. I 
couldn't find any articles with that simple a study, but I believe a dog is 
capable of learning that kind of time discrimination.