Many people understand that an e-collar can be used as a punishment to stop unwanted behavior in a dog. It doesn’t take a rocket scientist to come to that conclusion. Of course that limited way of thinking is what continues to promote use of the term Shock Collar.
This is unfortunate. 🙁 Because the tool can be used in a way that is quite gentle.
There is so much finesse with today’s electronics and the sensation can be associated with any number of meanings. One awesome way to associate the idea of paying attention is to pair the sensation with the use of treats.
This combination can be used to create a reliable recall and a well behaved dog who happily pays attention to you.
Skinner’s model of learning theory says that the term *positive* is to be used in a mathmatic sense (to add) so if we are adding stimulation in order to reinforce the likelihood of a behavior repeating……food for thought.
Check out the video clip for a quick example of this.
Happy Training,
Robin
Although the dog’s behavior does look relatively “fine”, there are tell-tell signs of discomfort such as the decrease in activity, the head drops, the ears back. They are indeed subtle and it is obvious that the amout of shock being given is not earth shattering, however it is punishing. The fact that it is being somewhat paired with treats is possibly enough to make it more than tolerable, but it doesn’t change the fact that for something to be Postive Reinforcement, it must actually increase a behavior. I see the shock being used at the point that the dog is furthest away (positive punishment) and then when the dog turns back to the handler it is released (negative reinforcement). The treats are actually given when the dog returns to the handler (positive reinforcement). If the shock were truly being used the same way as a secondary reinforcer (clicker), then the dog would be actually work to increase its distance AWAY from the handler (since that is what is being “marked”). I didn’t get the idea that this was the goal of the session. The goal (IMO) was to teach the dog to feel the shock, look back at the handler and come to it to mitigate the continuation of the shock and get a bonus treat for coming close. That was my perception….
Not at all Janet. I don’t know how you can come away with the idea that I agree with Sarah at all unless you have an agenda to put forth. And, it isn’t borrowing terms from the “positive reinforcement crowd” at all. Again, this shows your bias. There are four “science based” quadrants of learning and not just one or two as the “positive only” (which does not exist) crowd tries to promote.
Actually, Sarah has correctly identified what is happening. The fact that there are four quadrants, and that they all work to alter behavior is NOT the main issue. The issue is whether we can use the quadrants that are the least aversive to affect the same change in behavior, and the answer is yes. Expediency is never an excuse to abuse a dog. Anyone who has had the “odd sensation” of being shocked while scuffing across a carpet and then touching something metal usually does not want to repeatedly experience that sensation, yet you seem to feel it’s OK to do it to dogs. And, I submit that trainers who are not able to affect change quickly may not have the patience you might have, and WILL turn that dial up until the dog submits to +P. It’s unethical, in my opinion, to train dogs into learned helplessness and claim it’s obedience. A dog does not have to look unhappy to have given up trying to get away from the shock. Martin Seligman’s conclusions from the 1967 study are still valid today – I have yet to see any peer reviewed study that shows otherwise. Science should be combined with ethics, and you don’t have to cause discomfort to dogs in order to train them.
Ah, Michael – you have just proved Sarah’s point, so eloquently made and explained. If shock training is really so great, you wouldn’t need to constantly borrow and use terms and phrases used correctly by the positive reinforcement crowd. But, thanks for the compliments.
Sarah,
I believe you misunderstood Robin’s analogies. When Robin mentioned that adding something such as the stimulation from a remote collar would be a positive reinforcer…she was speaking to the general public and demonstrating how remote collars can be used humanely (via negative reinforcement) along with positive reinforcement (praise and food treats) in a comprehensive program.
The general public usually is not aware of learning theory and the four quadrants (positive reinforcement, negative punishment, negative reinforcement and positive punishment) nor do they want to learn about theory. They are interested in learning how to train their dog effectively and humanely. As such, Robin does an excellent job in breaking down theory and talking in terms that are easily understood by the public. If her targeted audience had been professional trainers at a a learning theory seminar, she probably would’ve scripted it using quadrant language and examples. But this is not who she was addressing.
DogTra collars have the capability of providing a stimulation at “just the right level” that gets the dog’s attention without harming the dog. Yes, it is something the dog wants to avoid experiencing (an aversive) but so is a raised voice, a stern look, withheld treats, etc. that many “positive only trainers” will inadvertently use at one time or another.
In your comments, you stated the remote collar is a positive punisher when the shock starts and a negative reinforcer when it stops. However, at what point does it become a positive punisher and then becomes a negative reinforcer?. It’s not simply the start and end of the shock if one is using the “continuous feature” of the collar (the shock continues until the undesired behavior, e.g. inattentiveness, discontinues). If one were to use the “nick feature” of the transmitter, punishing the undesired behavior, then it would be a positive punisher unless the dog was to learn that the nick will continue at a steady predictable pace until the undesired behavior discontinues. Then we are back to calling it negative reinforcement. However, for the purpose of Robin’s article, definitions of the four learning quadrants isn’t really necessary since the target audience was the general public.
You also stated,
“On a side note, it is possible to pair just about anything aversive with food and change the dog’s feelings about it (but I’m not so sure you have come close to building that association here).”
That is exactly what Robin was doing and I suspect you gathered that based upon your explanation as to how aversives can be paired with food. Robin was also using food as a positive reinforcer for the dog being attentive following the negative reinforcer which stopped the inattentiveness.
You also stated,
“That (needle injections) doesn’t prove that they never hurt in the first place (or that that amount of pain would be humane to use in training) simply because I am able to change her mind about it.”
The use of a remote training collar at the just right level does not cause pain. It’s an odd sensation that the dog seeks to avoid but it doesn’t hurt. I’m sure you would agree if you tried it on yourself at a low setting that simply obtains your attention or if you attended one of Robin’s dog training seminars.
As a recap, these were the main talking points I picked up from Robin’s article and video:
The collar can be used as negative reinforcement in a gentle way so people don’t think it’s only use is as a positive punisher set at too high of settings.
Using a remote training collar does not take away other training tools such as positive reinforcement (treats).
She was speaking to the general public who do not care about scientific terms.
In the past, remote collars didn’t offer the versatility that they do now and all too many trainers used too high of settings, thus inhumane positive punishers. Robin has been training trainers to use low settings that are sufficient to get the dog’s attention, like a tap on the shoulder. In the past, there were two camps….those who used food in training and despised remote collars and those who didn’t use food when using remote training collars. Robin is showing us that both can be used in a comprehensive training program.
Robin MacFarlane is a renowned professional dog trainer who is dedicated to educating the public and trainers how remote training collars can be used humanely and effectively. In fact, Robin once stated, “….the collar can be used in a way that is kind and gentle and yet effective. I am trying to create a paradigm shift in peoples’ thinking so that if they do chose to use a collar they will use it as something other than a “bigger hammer”. And, I am hoping to open peoples’ eyes that even if they don’t want to use a collar with their dog that they can understand it is not a tool to be banned or bashed because there can be responsible use and education is the key.” This is the reason she created the website, http://www.thetruthaboutshockcollars.com .
In addition to providing a forum for the proper use of remote collars, Robin is open to discussion on the matter from others who may disagree. I have no problem with you disagreeing with her viewpoint, but it is not professional nor appropriate for you to be so disrespectful in your tone and words. It’s sad when positive reinforcement trainers are so quick to “positively punish” those who have another training style and viewpoint. I sincerely hope you will come to one of Robin’s dog training seminars to observe her training methods first hand.
The definition of a positive reinforcer, as you mention, is something added(positive) contingently that increases(reinforces) the probability of the behavior being repeated.
If the shock were truly a positive reinforcer in this situation (see video) as you claim, then its application would serve to increase the future probability of the behavior it follows.
This (as I assume you know) is how a clicker works. It is paired with food and can then be used to “mark” desirable behaviors. As the clicker is a (secondary) positive reinforcer, the behaviors it marks will increase in the future.
You are asking us to believe that you are using the shock in the same way, pairing it with food and then using it as a positive reinforcer. This is not what is occurring in this video. No (competent) clicker-trainer would use a clicker to GET a dog’s attention. Clicking a dog as he is moving away or being inattentive might cause the dog to turn and look to you to get his treat, but it does not cause an increase in attentive behavior in the future. Remember, a positive reinforcer INCREASES the future probability of the behavior it follows. What behavior (once paired) did the shock follow? Moving away from the handler.
This is the same old stuff. The dog has learned that turning back to the handler will turn off the shock (negative reinforcement) AND will earn a treat (positive reinforcement). If inattentive behavior and moving away from the handler decrease over time, it will be because the beginning of the shock served as a positive punisher. Let me make it more simple for you, sometimes these fundamentals can be confusing.
Behavior: Dog is wandering off.
Consequence: Shock starts.
Prediction: Dog will wander off less.
Quadrant: Positive punishment
Behavior: Dog turns back to owner.
Consequence: Shock ends.
Prediction: Dog will turn back to owner more (particularly under condition of shock).
Quadrant: Negative reinforcement
Behavior: Dog turns back to owner.
Consequence: Cookie given.
Prediction: Dog will turn back to owner more.
Quadrant: Positive reinforcement
(Of course, there may be other contingencies occurring that are worth consideration in a proper analysis, but I’ll stop at the primary function of the shock and the food for this discussion.)
On a side note, it is possible to pair just about anything aversive with food and change the dog’s feelings about it (but I’m not so sure you have come close to building that association here). For example, my dog comes running for her weekly injections and stands and waits for them, because they have been paired with food. That doesn’t prove that they never hurt in the first place (or that that amount of pain would be humane to use in training) simply because I am able to change her mind about it. Besides, I would never STOP pairing the needle with food, because it would quickly regain its aversive property she would stop volunteering for her medical care. Same goes with your collar. It IS an aversive, and even if you had paired it with food properly (like I said, I don’t think this is what you have done here, so this is theoretical) you would need to maintain that pairing or it would quickly become a positive punisher/negative reinforcer again.
Back to the video. If the shock were truly functioning as a positive reinforcer, no proper trainer would use it when the dog was moving away as was being done toward the end of the video.
Simple fact is, the collar you use is aversive to dogs. That is how it is intended to work, and that is how it does work. You are using negative reinforcement and positive punishment when you are using that collar. The application of shock can positively punish the behaviors it follows. Following behaviors with the cessation of shock can negatively reinforce them. I’m assuming you don’t have a problem with that, so just stick with what you know.
I am choosing to assume that you are sincerely confused about these principles, rather than deliberately trying to deceive people. Please go back and review basic learning theory before you muddle these concepts for your readers again.
I believe the very first thing that needs to change is the name of this tool. I refer to it as a Control Collar, because that is what it provides me. The term “Shock” immediately gives the wrong impression.
I disagree. Euphemizing does not change how the collar works. It delivers a shock to the dog’s neck via metal conductive contact points that are part of the collar, and the shock is delivered when a human being presses a button on the remote activator. The fact is that if it were not aversive it would not work, and for some dogs, that means turning the level of shock up beyond what some trainers like to call a “reminder” or a “tap” or a “stim.” If you cannot be honest about that, AND refer to what you are doing in scientific terminology that makes sense to those in the behavior disciplines (psychology, ethology, zoology, etc.) then you are fooling yourselves and the public, which is unconscionable if you call yourself a professional. If you cannot correctly define the following terms, you have no business taking money from anyone to modify their dog’s behavior: acquisition, generalization, fluency, reinforcement (positive and negative), punishment (positive and negative), primary reinforcer, secondary reinforcer, conditioned response, unconditioned response, Premack principle, conditioned stimulus, unconditioned stimulus, event marker, extinction, extinction burst, learned helplessness, continuous reinforcement, variable reinforcement, classical conditioning, counter-conditioning, desensitization, behavior chain, latency, successive approximations, social or observational learning, modeling, luring, capturing, shaping, targeting, the five D’s (duration, distance, distraction, diversity, and difficulty). If you cannot do that, you don’t understand what you are doing, cannot adequately compare one method to another, and cannot decipher whether what you are doing could be equally well accomplished by a method that is more palatable to the dog. It is the dog, after all, who decides whether something is reinforcing or punishing. If I gave you pecan pie as a thank you for doing something nice for me, thinking you would like it, but you hated pecan pie, you might be polite and take it, but that wouldn’t change the fact that I could have made you much happier with chocolate cream pie.
Hi Anne,
You mistakenly assume my mission here is to make sense to those in the behavior disciplines. While I have had numerous people (including some who promote themselves as “all positive”) study and compliment me on my techniques, they can call it what they wish for it is little consequence to my individual journey.
I have not embarked on this path to impress the scientific community. It has always been my passion to simply help the frustrated pet owner. My background history gave me much first hand experience with those people who sought more final alternatives than a remote collar.
I know some trainers who can very eloquently explain each of the terms you list, however, they can’t seem to train dogs, particularly to the owners satisfaction. Ultimately it is the owner who is positively reinforced by the results they get. So for the time being, I will continue to do what I do, teach people how to use this tool in the least aversive way possible. And I will continue writing in way that creates discussion, for it is in midst of disagreement that we often achieve an understanding of middle ground.
all the best,
Robin
Hi Robin and All,
It seems to me that much of the argument against your use of the e-collar in this video is not actually a critique of the e-collar as a tool or of your use of the tool but against the use of all pressure in training.
If one side enters the discussion about this video with a base premise that any use of pressure is wrong then there is no way, or for that matter any reason, to discuss this particular use of pressure.
I wear shoes because the rocks in the world are uncomfortable to step on. I wear a coat in the winter because I am more comfortable when I wear a coat.
The universe is not evil or wrong because there are negative consequences built into my life experience.
I do not need to step on a rock every day before I put shoes on. I have learned to wear shoes to avoid the discomfort of the stones.
However, I do sometimes need to put my coat on in the middle of the day because conditions change outside. I have learned to put a coat on when I get cold.
I have even learned to read the signs of temperature change and I sometimes put my coat on before I get cold so that I stay warm.
I am not afraid of rocks, I do not worry if the weather changes nor do resent the changing temperature.
Thanks,
Pat
Good for you Robin, fill this site with videos of professionals working with dogs using “shock” collars correctly. The public has to know that the fear mongering all positive, all the time, ignore bad behavior, reward good behavior trainer wannabees spouting off about this tool are simply talking without any real knowledge.
In fairness though as is the case with any training tool, including treats and clickers there is a dark side to shock collar training that can be harmful to dogs. I’d also like to see fair light shed onto those practices. Sort of a “What Static Shock Collar Training is Not” I’d like those practitioners exposed for what they are so they aren’t tarring the reputation of true professionals.