There are lots of “positive-only” trainers that label trainers that don’t train their way as “unscientific”. It’s really a form of criticism of “non-food treat oriented” training and trainers. Sometimes, you’ll hear the slightly more finessed and stealthy criticism framed this way, “Well, you see food treat (positive-only) trainers are ‘modern’ dog trainers and everyone that doesn’t train our way is “old fashioned.” It’s a clever framing of the issue, to be sure. It’s designed to depict people who train with food treats as; good, smart, humane, scientific, new etc. and people who don’t use food as a primary part of their training as; bad, dumb, inhumane, unscientific and old fashioned (and I almost forgot – abusive).The good stuff about operant conditioning
Now, let me say that operant conditioning (that’s the fancy name for what has become ‘positive only’ food treat training) certainly has a scientific basis to it. It was developed by a researcher named B.F. Skinner, who popularized this form of conditioning decades ago. Also, I should say, that I’m not “anti”-operant conditioning. Operant conditioning is a very useful tool in many venues. Use of this sort of conditioning has transformed the zoo keeping industry by allowing keepers to condition zoo animals to feeding times, enclosure cleaning procedures, as well as many veterinary procedures. In addition, there is a lot you can do when training some dogs (and cats and rats or gerbils for that matter) with operant conditioning. Further, I use operant conditioning with puppies under six months old as a primary form of training. It’s only once a puppy gets to that 6-7 month stage of development that we start to form a more mature social relationship with the pup. Further, there are some specific tasks and behaviors with dogs that I find conditioning like this to be really useful.The weakness of ‘positive only’, operant conditioning
In any “scientific” approach it’s important, if you’re careful, to not over-state the conclusion. It’s important to know what the weaknesses, and limitations are in any conclusion. With regard to positive only operant conditioning, the weaknesses that I see are two-fold:
- The effectiveness of the conditioning is proportional to the motivation that the “reinforcer” (or food treat) provides. For example, I have many clients that simply say, “We were told our dog isn’t trainable because they’re just not interested in food treats or toys.” This is solved in dog training classes by telling clients to “get better treats” or “don’t feed Sparky on the day of class” This sort of advice comes straight from B.F. Skinner himself. To solve this “motivation issue”, Skinner kept the rats and pigeons (Skinner was never a canine researcher) that he did his studies with at ¾ of their normal body weight, literally starving them, to ensure when he brought a subject out for an experiment that they would be “consistently motivated” by the reinforcer. Since most of us feed our dog regularly and aren’t willing to keep them at starvation levels to improve training, this solution becomes problematic for most people.
- How motivating the “reinforcer” (food-treat) is, or how reliable the conditioned response is, when exposed to other environmental stimuli (distractions around your dog). Another set of clients that come to me have said things like, “Our dog could do the obedience command in our house or in the training room by himself. But if there were other people around, noises or any sort of distraction (like on a walk), none of the food-treat conditioning or food-luring worked!” This problem was dealt with by Sinner as well. In fact, he became famous (or infamous, depending on who you talk to) for his Skinner box. He conducted his experiments on rats and pigeons in highly controlled enclosures or boxes. In this way, he could control outside (distracting) factors and ensure an animal’s focus on the structured task in front of them. Skinner became intrigued with this idea of a highly controlled environment and even extended his ideas to child raising. Skinner raised his own daughter in a Skinner box ( which was a sort of rolling crib) that was completely enclosed with a glass window as the child’s only interaction with the outside world, with the exception of the several times a day where a child would be changed and fed. He marketed these “boxes” in the 1940s as “baby tenders” to “alleviate the inconveniences of mothering young children”. They didn’t sell very well because many mothers weren’t willing to keep their babies in a box 90% of the day.
There’s no doubt that some dogs can be trained to some level of “functionality” using a food treat based “positive-only” methodology. It’s probably important to define the word “functional” too. I’ve had many clients that competed in agility, competition obedience, rally, and other “sports”. They came to me because, while they were able to be “conditioned” to the tasks of their “sport”, they couldn’t perform very well when going to the beach, picnic, and park or be sane when someone tried to use a vacuum cleaner. So, I’m not saying don’t do “positive-only” training. In fact, my first exposure to training was in these methods. If that training works for you, then I’m happy for you. Again, I’m not on a mission to make everyone train dogs the way I do. I just don’t want people, who are continually frustrated living with their dogs, who have been told that their dog “doesn’t learn very well”, to believe that they have no other option. In my experience, using my method that takes a balanced approach to training and takes into account a dog’s natural communication style, works well for the vast majority of dogs.