I usually try to stay pretty neutral and open in my training posts. I have some pretty strong beliefs regarding dog training, but I try not to jump into the fray on the debate that often because one of those beliefs is: do what is working for you and your dog. If that means the word "No" never leaves your lips, that's fine as long as both you and your dog dog are happy. If that means you're working your dog on a prong collar, also fine as long as both human and dog are working well together.
All that being said, there are certain buzz words I cannot abide by. I see them and clench my fists to keep from typing a long comment that no one needed or wants.
The first one of those is "science based".
First, I want to discuss the fact that dog training and psychology has been studied, but those studies are largely small and few. There are so many times when someone will ask me: has there been a study done on this? Unfortunately, the answer is usually no. Though I absolutely love when I can pull out a study for a client to look over when it's available.
Next, I want to have a discussion about the quality of the science that goes into many studies about dogs, but most especially studies about training efficacy and ethics. My first issue is that sample size is often extremely low. I've seen sample sizes of under ten dogs. That's miniscule. I know people with litters of puppies more than ten. Yet often I see the value of a formal study on a tiny sample size extolled and breeders who have a sample size of hundreds of their dogs reviled when they make statements about their extensive anecdotal experiences.
Another issue is that often, a control group is lacking within the studies. I'll be frank, I did well in science in school, but it was never my first love. That being said, I'd love for everyone to think back to high school science and the importance of a control. Without that control group you have no baseline to compare your study group against. I will say that even when I see control groups in studies they are not usually a truly neutral control, and that's one of the hardest things about studying living beings with emotions, perfect control is difficult, especially when discussing behavior.
I'll add that often experimental design is either flawed, biased, or overreaching, and in some cases all three. First of all, check the funding for a study. Is the financial backing out to prove a certain result? Are backers part of the study? (I've seen this on at least one occasion). While it can be argued (well) that true neutrality is impossible, I'd like to see studies that are not single-mindedly seeking to prove (or often disprove) a hypothesis that it effects that experimental design.
Finally, regarding the quality of the science that goes into these studies, I beg of everyone, read the studies yourself. Don't just listen to a media blurb that usually sensationally overgeneralizes the results. How often do you hear: Science proves X is inhumane. or Science proves X is the only way to ethically train. Then go to the actual study and the study finds mild correlation of Y. Or their seems to be a correlation of Y. Science does not make sweeping general conclusions. It just doesn't. There's too much precision needed for generalities.
Another issue I have with the phrase "science based" or "science backed" or "science driven" or my least favorite "scientifically proven" in regard to dog training (though I see the phrase misused and abused in other contexts regularly) is that one of the longest established, most scientifically tested and verifiable, is the concept of operant conditioning*. I have discussed the basics of operant conditioning before. In short, operant conditioning is the process by which organisms seek to receive reinforcement and avoid punishment (with reinforcement being a scientific term that means increasing likelihood of a behavior and punishment being a scientific term meaning decreasing the likelihood of a behavior).
Operant conditioning involves four quadrants that interact with an organism differently and thus influence behavior in different manners. Two of those quadrants are reinforcement (reward) based. Two of those quadrants are punishment (aversive) based. All of those quadrants impact behavior.
Thus, I get a bit testy when people call themselves science based trainers and they cannot provide an adequate summary of the actual study they are extoling. Or they consider themselves science backed trainers but do not allow that the punishment quadrants of operant conditioning have been thoroughly studied and explored and found to affect behavior in a predictable manner (the organism avoids punishment).
Now, certainly there are extremely legitimate arguments to be made regarding the ethics of using punishment (aversives). I am not arguing with that statement. What I am responding to is the attempt to take an ethical high ground using partially true or unfounded statements like science based.
Another series of phrases that I find...lacking comprehensiveness are common phrases that trainers use to describe themselves. "R+" and "Force free" and "reward based" and "Balanced". Even LIMA (least invasive, minimally aversive has a lot of wiggle room, though CPDT has recently released very stringent new guidelines on that).
R+ is used to refer to trainers who primarily work within the positive reinforcement quadrant of operant conditioning. However, discuss this with most trainers and they'll confess they also use negative punishment. Have an in depth conversation about practicality and realities and they will admit that it's impossible to only use R+ and most I've talked to will admit that the environment naturally will apply positive punishment and positive reinforcement, even if they decline to do so on an ethics basis. While it is a descriptive phrase, it's also not a complete one, which you'll soon find is a problem I have with most descriptors.
Force free (and fear free for that matter) is another phrase that I find not telling the whole story. The idea is that you are as force free as possible (same with fear). But talk to a force free trainer, pose them ethical questions. A dog is bolting for the street. Do you jerk their leash back to keep them from being run over? A dog has you pinned in a corner, jumping up. They outweigh you. It's friendly but there's no one around to help and they're physically injuring you. You weren't expecting to encounter a dog and you don't have treats. What do you do? The reality is that life (for dogs and for humans) is not force or fear free. Yes, the label tries to market itself by capitalizing on our desires for ourselves and our dogs. I don't think that's inherently harmful, but it is clever marketing and the fact remains, life isn't ever going to be completely force or fear free despite our best efforts.
Reward based is a descriptor that doesn't grate on me as much as many. However, again, my issue with it is clever marketing and a bit of misdirection. It's really easy to point at these theoretical trainers that never use positive reinforcement and just use punishment. In my experience, and in the experience of my colleagues I have talked to, the majority of trainers use a lot of a large amount of positive reinforcement and a sprinkle (this sprinkle varies widely) of punishment, and it's just the edges of the bell curve that are 99% positive reinforcement or 99% positive punishment/aversive. And by claiming that you are a reward based trainer, you are claiming that you are not one of those trainers who doesn't use rewards, when the reality is most trainers are primarily reward based.
Balanced is a term I take different issues with. It is generally defined as a trainer who uses all quadrants and all tools to suit the individual dog in front of them. The very fact that I had to define the term illustrates the issue. The general public doesn't know what balanced means. It's catchy in terms of philosophically desiring balance, but less descriptive than many. I'd also argue that it's entirely too broad and can mean essentially anything an individual trainer would like it to mean.
Ultimately, I guess this post boils down to the fact that I think our labels regarding training philosophy are at best inadequate and at worst horribly misleading. I don't think this is necessarily a problem that will be solved with better or catchier marketing (and yes, it is marketing. Trainers believe in their training style and want to sell it).
The solution? I can't say I have a perfect one. I will always recommend you have a discussion with a trainer you're interested in using. (That does mean sign up for a consultation and pay their fee. Don't expect them to work for free. I don't know a single trainer who's not constantly overworked). A discussion will allow you to get a feel for each other and see if it's worth moving forward. Not every trainer is a fit for every person or dog. Not every person is a fit for every trainer. I generally try to make myself approachable, available, and amenable to each individual owner. But there are certainly some client traits that I will pass. For me that usually means: rigidity in an unresearched training philosophy and an inability to listen. I'm generally pretty talented at giving homework that clients will actually accomplish, one of my secret super powers that I know can often be an issue for trainers. That being said, other trainers are going to have different lines, desires, and thresholds.
When it comes down to it, talk to the trainer. Please don't just choose someone based on buzz words.
*Some links below with some actual studies regarding operant conditioning (which includes punishment), including BF Skinner's findings, who was the first to formally posit operant conditioning. These studies do not necessarily involve dogs, but they do explore operant conditioning in a variety of animals:
Comments