Carrots and sticks. Rewards and punishments. As humans, we resist the notion that these are the primary elements that will dictate our behaviour. We see it all the time; the story of the individual who overcomes adversity and hardship to achieve great things. Our own experiences of feeling motivated to do better after a personal setback. Clearly behaviour must be dictated by more complex forces than the simple rules of B.F. Skinner and operant conditioning. As dog trainers, this can make things very confusing.
Recent research into human motivation in the business world has provided some startling insights into human motivation and performance in the workplace. Remarkably, it appears that the common business practice of offering financial rewards (i.e., bonuses) for reaching performance goals can be ineffective and, in some cases, can even have a negative impact on performance. The research seems to contradict what we are learning in modern dog training. That positive reinforcement training, using rewards for behaviour, produces effective and even superior performance compared to other training methodologies. What’s going on here?
Paradox
In his book Drive – The Surprising Truth About What Motivates Us, author Daniel Pink describes a study done in 1962 by Samuel Glucksberg, a professor of psychology, to investigate the effect of rewards and incentives on worker performance. One notable experiment involved asking participants to solve a proposed problem. One group was offered financial incentives for those accomplishing the task in the shortest time. The other group were simply told that they were being timed to establish an average for how long it took a person to complete the task (no incentives or punishments were introduced). Remarkably, Glucksberg’s results showed that the group that was offered the financial incentives performed the task MORE SLOWLY than those in the group that was given no incentive!
This seems to fly in the face of what we know about rewards and behaviour. But Glucksberg repeated the experiment with the same setup for the two groups. However, this time Glucksberg simplified the components of the task so that the means to accomplish it were obvious. This time the group that was offered the financial incentive easily outperformed the group with no incentives. What was the difference? What’s going on here? Daniel Pink’s analysis of these results shows something remarkable.
Different jobs, different motivations
It seems that rewards and punishments can be remarkably effective when it comes to simple or mechanical tasks where we ask for specific behaviours. But when it comes to more complex, problem solving behaviours, these basic rewards and punishments will be less effective. What Pink is suggesting is that there must be a more complex model of motivation beyond just what is provided as the result of behaviour. Although the work of B.F. Skinner showed that the consequences of a behaviour will determine whether that specific behaviour will be more or less likely, clearly the externally provided consequences are just one factor in a bigger puzzle of motivation.
More creative and complex tasks and behaviours seem to derive from more internal motivations. The satisfaction of a job well done, contributing to a community project, or even just making a loved one happy can all be motivations for human behaviour. But just as these internal motivations have their place in driving complex behaviour, the operant conditioning model of rewards and punishments has proven to be incredibly effective at teaching rudimentary skills and behaviours not just in animals but in humans as well. It seems that different motivators are better suited to different jobs.
For his book, Pink looked at research done at MIT, Carnegie Mellon, University of Chicago, the London School of Economics, and other institutions. What he found was that, across the board, external rewards and punishments had a huge influence on performance when it came to simple, mechanical tasks that had clearly defined criteria and structured rules for how the behaviour was to be done. But when problem solving and creative thinking were required, three internal factors had a tremendous impact on the subject’s performance – the level of autonomy the subject had in pursuing their results, their mastery or ability to gain mastery of the task, and the ultimate purpose and meaning of the work they were doing.
Now, while all of this is interesting when looking at human behaviour, how does this relate to dogs and how we work with them? I think the important aspect of this work on human behaviour for us dog people is recognizing the problems that can come from ignoring the internal motivators that Daniel Pink refers to. It is unlikely that our dogs will ever be called upon to do any serious problem solving or creative work. But these internal motivators can impact even simple tasks as show by the Glucksberg experiments.
So much of our human experience involves complex, problem-solving behaviours in daily life. It shouldn’t come as a surprise that we are very familiar with the internal motivators that Pink describes. In fact, much of traditional notions of dog training that resist the use of rewards also rely on ascribing these internal motivators to our dogs. Our dogs should perform for us when asked “because they love us” or out of a sense of loyalty or duty. These are motivators that we humans understand well and respond to almost daily. But the truth is, our dogs live very different lives than we do. Their life with us does not consist of the creative, problem-solving complex behaviours that we experience. Theirs is a much simpler, task based existence where we require only “sit”, “wait”, “come”, or a few other basic behaviours.
Autonomy, mastery, and purpose
Although our dogs life with us consists mostly these simple tasks, that does not mean that opportunities for internal motivation are not present. In fact, depending on your lifestyle, there are any number of ways that a series of these basic behaviours can come to be seen as a more complex behaviour chain and where internal motivators can indeed become an important factor.
Let me give you an example from my own life. I work with my dog in the sport of dog agility. It is an interesting sport because it can be viewed and played in different ways. One way to look at agility would be akin to pinball; an obstacle course is laid out and the job of the human is to prevent the dog from going to the wrong place until all of the obstacles have been completed in the correct order. I chose to look at the sport another way. It was my job to teach my dog all of the skills necessary to complete an obstacle course and then to ask her for those behaviours at the appropriate times so that she completed the course without an error. My approach is far more risky and much more demanding of myself and my dog to work as partners.
This is where Pink’s description of internal or “intrinsic” motivators starts to make sense to me when I look at my dogs. My approach to agility definitely grants a certain level of autonomy to my dog. As I stand 15 feet away and direct her over a jump, she is certainly free to go to the tunnel instead. I can’t prevent her from doing that and she knows this. So she has a healthy degree of autonomy since I ask for the behaviour rather than demand it.
Using rewards in teaching my dog individual skills and behaviours means that my dog is successful 70% of the time or more. I control that through adjusting my expectations as I teach individual behaviours. I think my dog’s high success rate may convey a sense of “mastery” of these skills and behaviours. Certainly my dog appears happy and confident as she runs through an agility course whether I direct her to the correct obstacles or not. She has successfully practiced those behaviours hundreds of times.
As to Pink’s assertion that a sense of purpose or meaning for behaviour providing an internal motivation, this one is the hardest for me assess with regard to my dogs. There are difference in the cognitive abilities of humans and dogs. I cannot be sure if my dog understands that she is working toward a successful agility run with me and that it gives her a sense of purpose. And so Daniel Pink’s conclusions can only be stretched so far in trying to apply them to dogs. But it is certainly something to think about.
The human-ness of dogs?
This question of external (rewards/punishments) versus internal motivations seems to be something that we dog owners ignore at our peril. Clearly our dogs are not simple input/output machines whose behaviour can be easily and predictably manipulated by the consequences we provide. Our human understanding of internal motivation has impacted our approach to dog training for centuries. But our dogs are not human and we should avoid the mistake of assuming too much. For as much as science has been able to learn about dogs, we still don’t know how much of our human cognitive model can be applied to our dogs with any accuracy. There are still too many questions.
Can we say for certain that a Search and Rescue dog feels a sense of purpose in the work they do? All we can do is continue to work with our dogs and continue our observations. What I am convinced of is that our expectations can be unrealistic. We can assume either too much or too little regarding the internal motivators of our dogs. As the science moves forward we are learning more and more but, in the mean time, shouldn’t we be putting the dog’s physical and mental welfare above all else? We have to keep our expectations realistic. They are dogs, not people.
I am convinced that dogs do have internal motivators. Some of them are biological or instinctive. Some are learned. And some, like those internal motivators described by Daniel Pink, are probably worth much more scientific study before we can understand them better. But in the mean time, we can all be scientists. We can all work with our dogs and observe the results. We might be amazed at what we find if we provide a little more autonomy, teach our dogs a little more mastery, and try to convey to our dogs that what they do for us and with us, matters to us.
Until next time, have fun with your dogs!
Be sure to check out our Canine Nation ebooks in the Canine Nation store and Dogwise. Join our conversation on Facebook in the Canine Nation Forum!
The NEW Canine Nation ebook is now available –
“Relationships: Life with Dogs”
Photo credits –
Pug and Carrot- Robert Occhialini copyright 2010 from Flickr
Dogs – Chris Scheufele copyright 2007 from Flickr
video – Petra Wingate copyright 2014
Motivation – Samuel Cockman copyright 2011 from Flickr
Kevin Behan says
Great article and believe you’ve put the research into the proper perspective. I would like to suggest that the source of this internal motivation and self-directed behavior can be found in the title of Pink’s book, “Drive” as opposed to looking for higher cognitive processes. I know the existence of Drive has been minimized if not dismissed in modern behaviorism but this is why the current models are incomplete. In my view Drive evolved to overcome resistance. And because dogs have more Drive than more neurologically advanced creatures, such as apes, chimps, monkeys, etc., is why they are able to perform work related service for humans (S/R, Police, Protection, Bomb, etc.), most especially under extreme conditions, which these brainier animals are not capable of. Drive runs deeper than instinct and engaging the world on this level is where the feeling of autonomy and purpose arise.
Eric Brad says
Thanks for your comments, Kevin! 🙂
Tanya says
Gear article, Eric. I need to absorb this one and come back to read again tomorrow. So much to think about.