Friday, July 26, 2013

In positive reinforcement training with dogs, which reward is most effective? A study sheds some light

Thumbnail image for gretchenpuppy.jpg
Gretchen, around 8-10 weeks of age. Training had by that time already started.
Ideas in dog training and understanding canine behavior have certainly come a long way in the past 15 years or so.

My Gretchen is almost as old, and I have to say that I wasn't comfortable with every aspect of some of the popular methods that were around back then when she was a puppy. No, let me rephrase that: I wasn't comfortable with most of them.

Those were the days before puppy classes and hiring the expertise of a professional dog trainer were the norm, I should note. But being totally engaged with the prospect of unfolding what would become an adult dog who was able to think independently and control herself with a fair amount of finesse in our very-human world was paramount.

"Doing the hard work now will pay off for a lifetime," was my mantra.

But the lack of what I felt were good resources in dog training back then did get me thinking about what I knew in my gut: forging good communication, trust and a solid relationship with her was the only way.

The former was really the only way for me to attain and keep that solid relationship and trust with Gretchen.

The problem? We didn't share a language yet. But it was my mission to make that happen. And punishment or aversive-based methods weren't going to enable that to come to fruition — they were only going to make her fear me, cause her to retreat and tune out.

Bringing the chubby mass of fur home at five weeks was already a challenge — that was much too young to have been weaned and not have much contact with her mother — so I knew that I had a lot of diligence to put forth. So I followed my gut: I set about establishing a strong bond through daily interaction/positive touch and play, kept a routine and "listened" to what she most favorably responded to.

What I quickly learned is that she not only loved yummy treats, but praise and interaction (including but not limited to games, a walk to her favorite haunts). I knew that the latter two were exceptionally important to cultivate with care, as they were going to be the foundation for great communication in our daily life for many years to come.

Depending on the skill I was teaching, I would use praise, interaction/positive touch, games/play and sometimes food rewards.

Admittedly, I did not use food rewards for everything, and certainly not every time that I wanted to convey that she had gotten something right. I did find that food rewards were especially useful for difficult training like "stay," since the skill is graduated into different distances apart as a dog gains ability to control herself over time.

Gretchen was a eager learner, and did so quickly.

Fast forward so many years later, and Gretchen is a settled, happy dog whom has been able to enjoy her life — and has been a pleasure to be around.

There seems to be considerable dissent between those in the canine training and behavior community with regard to which methods and approaches yield the best results.

Typically, you'll hear discussion in these circles with regard to aversive or punishment-based methods vs. those that employ positive reinforcement only. But as Stanley Coren, Ph.D. discusses in a recent article on PsychologyToday.com, positive reinforcement trainers may not always be in agreement, either.

The question of whether food rewards, verbal praise or positive physical touch is most effective in training can be a sticking point amongst professionals, but hasn't really been formally tested until recently.

Coren talks about results of a study recently published in the Journal of Veterinary Behavior which reported research by co-authors Megumi Fukuzawa and Naomi Hayashi in the Department of Animal Resource and Sciences of the College of Bioresource Sciences at Nihon University in Japan.

The study, Comparison of three different reinforcements of learning in dogs, divided 15 dogs into three sets, and in each set, a different reward was used: praise, stroking/petting and food.

As part of the study, observing the dogs while they were learning the "sit-stay" and the "recall" (or "come") command seemed to make the most sense, since those are core commands. All three groups were trained identically.

Overall, the trainings that employed food rewards were more effective than the other two methods used. Interestingly, early in training for the recall command, there seemed to be a slight advantage in food reward.

Used in tandem with the right timing to reward the dog, as well as keeping training sessions short and fun, food rewards can make all of the difference, especially when introducing a new skill.

There are some out there who dismiss the notion of using food rewards at all in training, but hopefully this study will broaden the conversation with regard to how useful the practice can be, and prompt more studies like it in the future.

Click here to read the article published on PsychologyToday.com.

Lorrie Shaw is a blogger and owner of Professional Pet Sitting. Shoot her an email, contact her at 734-904-7279 or follow her adventures on Twitter.



No comments:

Post a Comment

Thanks for your comment!