'Do as I do' poised to be a credible method in training, highlights dog's capability as social learners

One unseasonably-warm spring day in 2000, I was knelt down in a soon-to-be flourishing flower bed. Having ditched the hand shovel that I had been using, my gloved hands seemed a more efficient tool to dig small pockets in the soil in order to get the items that I'd picked up planted. 

Unbeknownst to me, Gretchen – who was about 8 months old at the time – had been spying me instead of chewing on the bone that I'd given to occupy her mind. Her curiosity got the better of her it seemed, since before too long she had claimed a spot next to me, peering over at my methodic digging and mounding. 

Suddenly, she set to jabbing one front paw at the freshly cultivated earth, then the other. Looking over at my work, and back to her own poking and digging, tail wagging all the while and her face wearing an expression as if to say, 'you and I – we're doing the same thing!'. 

It's not just a fond memory for me, but an illustration that dogs are fully capable of social learning. 

This type of learning has been studied in different non-human species (including lizards) 

Social learning, as the name implies, bears the hallmark of one learning a new behavior by watching someone else perform it, then copying that behavior. It was first thought that only humans were the only species capable of the processes that are considered to be high-order. 

Seeing how in-tune Gretchen was even at that age, I was excited to use her willingness to follow my lead to help me facilitate better communication and our training process. It was a fun way interact with her and proved to be an effective tool in unfolding her. Little did I know that years later, it would become a recognized asset in the world of canine behavior.

Any positive direction that dog training takes is a good one, and an interesting method called "Do as I do" (DAID) is an example of this and is catching on. By using a dog's ability to learn socially, DAID enables the handler to demonstrate a set of actions to the dog that can then be imitated (thusly, learning by doing). 

In a recent study done by researchers Adam Miklósi and Claudia Fugazzi from Eötvös Loránd University, Budapest, Hungary, set out to see how a more traditional positive reinforcement method using clicker/shaping training (which uses individual learning) stacked up against DAID. 

The pair expanded on their findings in the study published in August, Social learning in dog training: The effectiveness of the Do as I do method compared to shaping/clicker training. 

Fugazzi, who created the DAID method, is also author of the book Do As I Do: Using Social Learning to Train Dogs. 

Most interestingly, the results from the study indicate that DAID is more effective than shaping/clicker training when it comes to learning things that involve interacting with an object. For the study, opening a sliding door on a cabinet was chosen because for each dog involved, as it was a novel behavior.

Dogs who were trained with DAID learned faster than the dogs who were only exposed to clicker training for the skill. They also retained the ability to perform the behavior when later cued verbally better than dogs in the clicker group – even after 24 hours had passed from first exposure to learning the skill. Furthermore, the DAID dogs could apply the same skill more reliably in a new context. (This could prove to be helpful in "proofing". For more on that, click here.) 

Though DAID incorporates social learning, it's a more simplified form than is seen in other instances. 

Social learning often plays out by way of an individual learning a new behavior simply by observing others in a group setting, and without any cue to reinforce that they're getting it right.

In DAID, the dog learns by observation, then performs a behavior and has positive reinforcement – that all important form of communication – when it is performed correctly. (The positive reinforcement is a hallmark of operant conditioning.) In essence, a handler - treats in hand, ready to reward - gets the dog's attention and performs an action with the animal paying close attention. Next, the handler exclaims, "Do it!", at which point the dog mimics the action and if performed accurately, is rewarded with a treat and a verbal affirmation.

Using this multifaceted approach is fun, engages dogs in a positive way and enhances their natural learning style, since they are social animals.

More research is needed on how reliably dogs can learn body-movement behaviors (another area of importance in canine training) as was referenced in the study. I think it'll be interesting to see how reliable this kind of training is for those desirable behaviors.

The best part of this research is that it highlights the importance of establishing a solid human-animal bond between a dog and their people, and how we can make it work to enhance our communication with our canine counterparts.

Watch the video below to see a DAID training in action.






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