Eureaka! Problem Solving Makes Dogs Happy

What's in this article? 

• Solving a problem means you learned something new, meaning you have more control over your environment, which makes you feel happier. This behaviour is also seen in dogs.
• Researchers tested the happiness level of dogs who problem solve by teaching beagles to solve three puzzles and testing their response to receiving a reward in comparison to dogs who did not solve a problem but received a reward anyway. 
• During the test, dogs who completed a problem ran faster and wagged their tails harder on their way to receive their reward than dogs who received a reward without solving a problem.
• Upon entering the test, the dogs who learned to solve the three puzzles became excited upon seeing a problem they had solved before.
• The study showed that dogs are happier receiving a reward after solving a problem than they are when getting a reward for no reason.

______________________________________________________________________________

Do dogs experience pride of accomplishment when they solve a problem? A new study set out to answer this question. For human beings, the simple act of solving a problem can be quite rewarding in and of itself. That is why people spend time trying to solve crossword and jigsaw puzzles, Sudoku, and so on. When a person fills in the correct word in a crossword, they are not given a piece of food, money, or social praise—their only payout is knowing that they managed to solve the problem. I am familiar with this process since my wife is a great fan of crossword puzzles and will work at them for hours. If I happen to wander by when she is filling out one and I notice a set of blank spaces for which I happen to know the answer, I am always tempted to suggest the solution to her. However, from previous experience I know that the response that I am going to get is something like, “Don't do that. It's my puzzle and it's no fun if somebody else gives me the answers!” In other words, by solving a bit of the problem for her I am taking away some of her reward.

Given the fact that there are a lot of similarities in the emotional responses of dogs and humans, one might guess that simply solving a problem is rewarding for canines as well. From an evolutionary perspective, solving problems should be rewarding since each time you solve a problem you demonstrate that you have learned something new about your world and are a little bit more in control of your environment. The more you know about and can control in your world, the more likely it is that you will survive. This should be true for all animals, not just humans. 

The sudden surge of positive feeling that we get when we solve difficult problems is often referred to as the “Eureka Effect.” The term comes from an incident involving the Greek scientist and inventor Archimedes, who was asked to determine whether goldsmiths had adulterated what was supposed to be pure gold in the crown of Hiero II, the king of Syracuse, with some other metals. While looking at the water level rise as he immersed himself in a bath the answer to his problem came to him. Archimedes leapt up and took to the streets—stark naked no less, so happy and excited by his discovery that he had forgotten to take time to dress—all the while shouting “Eureka!” (from the Greek heureka, which means “I have found it!”). Archimedes was demonstrating an example of an intense reward feeling associated with solving a major problem, however lower levels of the Eureka feeling are what reward us for solving everyday problems. It is this same gush of positive feeling which provides the reward that keeps us working at puzzles and computer games involving problem solving.

A team of researchers headed by Ragen McGowan of the Swedish University of Agricultural Sciences in Uppsala, Sweden decided to see if this same Eureka effect appeared in dogs. The essence of the experiment was to show that the act of solving a problem to get a reward has a far more positive effect on a dog than simply getting the same reward without working through the problem.

"Solving the problem to get the reward made the dogs feel better than simply getting the reward without any intellectual accomplishment."

The animals tested were a group of female Beagles. There were six different tasks which the dogs could be trained to do: pressing a lever, pushing a box off of a stack, tipping over a plastic construction, pushing a ball off of a table, pressing a paddle to ring a bell, or pressing a key on a toy piano. When the dog was successful there was a sound cue, such as a click or a bell, followed by a reward. Each dog was trained to perform three out of the six possible tasks.

After a week-long break, the actual testing began. A brand-new test environment was used that had a starting compartment with a gate that opened to a large arena. When the gate opened, the dogs gained access to a reward at the far end of the room. Possible rewards the dogs could receive included food, social contact with a human who would pet them, or a chance to have social interaction and play with two other dogs. At the beginning of each test session there were two trials where the gate was opened and the dog got to see what kind of reward she would encounter that day.

For the actual testing, the dogs were measured in matched pairs. Each test begins with a piece of test apparatus in the start area. In the problem-solving condition, one of the dogs was tested with a piece of apparatus that she had been trained to operate. What the dogs had to learn was that, although each problem worked in the same way that it did before, the results were different. Now, working the apparatus resulted in a sound signal and the gate swinging open so that the dog could go and get her reward. This problem-solving dog's behaviour would be compared to the other member of the matched pair who was placed in the test area with a piece of apparatus that she had not been trained on; no matter what she did, there was no effect and thus she was unable to solve the problem. Regardless, the dog that did not get to solve a problem was still given a reward (sound and gate opening) at the same time interval that the problem-solving dog in the previous session had obtained her reward. Thus, one dog gets the reward for actually solving the problem while the other dog gets the same reward without having to solve the problem first.

To determine whether getting to solve the problem had a positive effect on the dogs, a number of different measures were used. One involved assessing how quickly the dogs shot out of the gate to retrieve their reward. Investigators also measured the dogs’ activity level (believe it or not, these researchers counted every individual paw movement) and tail wag (recorded on video and then counted). The activity level indicated the dog's excitement, with tail wagging and the speed at which the dog went for the reward indicating how positive the dog was feeling.

The results clearly indicated that solving a problem was rewarding to the dogs. When the dogs correctly manipulated the apparatus, thus solving the problem before they got the reward, their activity level was higher, and they showed many more positive indicators (such as tail wagging) than when they simply received a reward without earning it. In other words, solving the problem to get the reward made the dogs feel better than simply getting the reward without any intellectual accomplishment.

It is interesting to note that the dogs seemed to be quite happy and interested when they were brought into the testing sessions and could see that they would be interacting with a piece of apparatus that they were familiar with, one that posed a problem that they knew they could solve. When presented with a piece of apparatus that they had not been trained on, the dogs seemed to show frustration and reluctance, even when they received random rewards that they otherwise would have had to work for. It was their lack of control over the situation that seemed to bother them.

The experimenters are quite certain that the dogs were experiencing the "Eureka effect." At the study’s conclusion they reported, “It was success in the problem-solving that elicited a positive affective state in the experimental animals.” In other words, just as for Archimedes, the act of simply solving a problem gives dogs a burst of positive feeling—but unlike the Greek scholar they do not feel any subsequent shame when they find that they have rushed out into the world naked because they were overcome with happiness at their accomplishment!