Author, Cindy Ludwig
There are as many ways to train a dog as there are people who train them. There are television shows about dog training and blogs, books and a variety of training equipment for sale in pet stores. How does the average pet owner who just wants a well behaved dog go about sorting through all this information to determine which equipment and methods of training are the ones he should choose for his dog? Do different dogs require different methods of training?
Training methods exist on a continuum
The way I see it, dog training philosophy and methodology exist on a continuum between hypothetical endpoints of reward-based training and correction-based training. In reality, there is no such thing as reward-based training without punishment (correction) because reward-based (aka “positive”) trainers withhold rewards which is a form of punishment, and there is no such thing as correction-based training without rewards since even correction-based trainers give some positive reinforcement (praise, food treats, etc.). Most trainers then lie somewhere on the continuum between the two endpoints, tending to be either reward-based or correction-based, depending on where they sit on the continuum.
Balanced training
The trainers that occupy the middle ground sometimes refer to themselves as “balanced” trainers, using a “balance” of correction and rewards. If one imagines the reward-based end of the continuum as positive and the correction-based end of the continuum as negative, balanced training that hovers around the midpoint of zero could be conceived as having little or no effect, with the negative cancelling out the positive. In reality the result is more of a diminutive and unreliable response than an actual zero effect.
While it is totally possible to get a well behaved dog using this approach, the trade-off is likely to be a dog that is inhibited and does not offer behaviors for fear of being punished. Think about it, if you don’t know whether your next move will be rewarded or punished, how likely are you to experiment and offer new behaviors? What will your demeanor look like? Will you be very creative? Have much initiative?
Is correction necessary to train dogs?
Many people who are not familiar with the science behind reward-based (aka “positive reinforcement”) training wonder how an animal can learn if not corrected. I think the question is not, how can you teach a dog without correction, but how can you teach a dog with it, the operative word being teach? We know, for example that stress inhibits learning and we also know that we learn more from our successes than our failures (MIT, 2009). The use of punishment – or correction in training in my opinion is really something that has its origin in outdated cultural norms.
So then, how does an animal learn what he did wrong if not corrected?
Quite simply. The basic premise of positive reinforcement training is that animals repeat what is rewarding and do not repeat what is not rewarding. So rather than correct or punish an animal for an undesired behavior, positive reinforcement trainers simply ignore unwanted behaviors and continually revise their training plan to set an animal up for success while observing for desirable responses on which to build their goal behaviors. Desired behaviors grow stronger and undesired behaviors extinguish or go away. A motivated animal learns very quickly what works and what doesn’t to get him what he wants! For dog and handler, it’s a win-win situation!
Food
Some wonder if an animal trained with food rewards will always need food to continue to perform a desired behavior and the answer is yes and no – depends how you use it. Once a behavior is learned, the rate of reinforcement (rewards) is decreased and put on a variable schedule, meaning the animal doesn’t get a reward for every correct behavior but instead receives rewards on a less frequent and irregular basis, which actually serves to keep the learned behaviors strong.
Skilled trainers know how to use food appropriately in training so that an animal is neither distracted by the presence of food or dependent on the food to perform.
Some people would like to train their dogs without food for a variety of reasons, but food is a necessary part of animal training, at least in the early stages. It is what behaviorists call a primary reinforcer, meaning it is something that all animals need and do not have to learn to like that serves to increase the likelihood a behavior will be repeated.
Food can be delivered at the reinforcement rate necessary to facilitate learning and keep an animal engaged in the training process. Later on in the training process, food will be systematically replaced with secondary reinforcers such as play, petting and praise.
Food is also integral to treating animals with fear and anxiety because food activates the parasympathetic nervous system, causing an animal to relax and develop what we call a positive emotional response. In training food is also useful as a gauge of stress. An animal that cannot eat is too stressed to learn, and something needs to change before asking the animal to continue with a training session.
Where people get into trouble is using food as a bribe rather than a reward. If we do too much luring with food and food is always within easy view of the animal, the animal may become dependent on food and refuse to perform without it. If instead, we keep the food out of sight and make rewards contingent on performance, food is used as a reward.
Clicker training
Clicker trainers are positive reinforcement trainers who minimize luring with food and use a small mechanical device called a clicker to signal to the animal being trained when it has performed a behavior worthy of a reward. The click-click sound of the clicker marks a behavior at a precise moment in time and signals to the animal that a reward is forthcoming for that behavior.
In clicker training there is no “wrong” or “no,” only right or try again! In clicker training we speak of cues versus commands. In compulsion training, commands are given. In clicker training we teach a behavior before teaching the animal the name (“cue”) of the behavior, whereas in compulsion training and some other forms of positive reinforcement training the command is taught at the same time a behavior is taught. A cue represents an opportunity for reinforcement, whereas a command is an order. If you give a command and your dog disobeys your only option is to increase the forcefulness of your command, but if you cue your dog to perform a certain behavior and he misses it the first time, you’d better believe he will listen better the next time!
The benefits of clicker training include 1) no harmful effects, even if used incorrectly, 2) rapid learning, 3) enthusiasm for learning, 4) deepened bond between owner/trainer and his or her animal, and 4) long term memory for what has been learned.
“Clicker training” is a highly precise and versatile method of training that can be used to teach almost any animal anything the animal is physically able to do. It is used in a variety of settings these days from zoos to wildlife preserves to veterinary offices to guide dog training centers such as Guide Dogs for the Blind on the west coast. Clicker training can be used to teach scent detection dogs, to teach canine musical freestyle, and to teach agility to cats, horses, rabbits and sheep! It has even been used to teach rats to sniff out land mines! Now it is available for dog owners to use with their pets!
Clicker training encourages creativity and teaches an animal how to learn. Clicker trained animals are easy to spot –they’re enthusiastic learners with happy faces and bright eyes! When not working, they’re calm and content, but ready to start their next training session when the opportunity arises.
Lure-reward training
Lure-reward trainers are generally positive reinforcement trainers who use food as a lure moreso than clicker trainers and may or may not use a clicker or other marker signal in training. Lure-reward trainers vary greatly in their methodology and may employ some degree of correction.
Correction-based training
In stark contrast to positive reinforcement training is correction-based training, aka compulsion, or force-based training. Correction-based trainers believe they must correct or punish undesired behaviors in order for an animal to learn.
Correction-based trainers use such training aids as prong (“pinch”) collars, choke chain collars and shock collars, aka “remote trainers” or “e-collars.” All of these devices work by applying what is known in behavior science as positive punishment and negative reinforcement. Quite simply, an aversive stimulus is applied for an undesired behavior and it is discontinued when the animal complies with a desired behavior.
Positive reinforcement, negative reinforcement – what do all these terms mean?
Don’t get confused about the term, “positive punishment.” “Positive” in behavioral terms simply means to add something. Negative” means to take something away. “Positive reinforcement” means to add something that is reinforcing, and “reinforcement” is something which increases the likelihood a behavior will be repeated. “Positive punishment” means to add punishment, aka correction. “Punishment,” in behavioral terms means something which decreases the likelihood of a behavior being repeated.
“Negative reinforcement” is a little trickier – it means to remove something which will cause an animal to repeat a desired behavior. i.e., comply. Stopping electrical shock when an animal lies down or comes when called would be an example of negative reinforcement. Using a “force retrieve” by pinching a dog’s ear or his toe until he takes a dumbbell in his mouth is another example of negative reinforcement. The pain stops when the animal complies. Negative reinforcement is a very powerful method of training, but it raises ethical questions in the minds of positive reinforcement trainers who contend that we have just as effective if not more effective means of training without the unwanted effects of punishment, aka correction.
“Negative punishment” is yet another option for getting a dog to respond the way that we would like by manipulating consequences. In negative punishment, we are “punishing” the dog by removing something the dog desires, for example, our attention, in order to decrease the likelihood a behavior will be repeated. A good example of this is turning your back to your dog when he jumps on you for attention. If you turn your back consistently your dog will learn that jumping up doesn’t get him what he wants and this behavior will extinguish, or disappear. Positive trainers also teach the dog what to do instead of jumping, so they will reward the dog with attention when he sits or at least isn’t jumping. This is positive reinforcement
So in summary, “positive reinforcement” trainers rely on positive reinforcement and negative punishment to train dogs and modify behavior. Correction-based (aka compulsion) and “balanced” trainers employ positive punishment and negative reinforcement whereas positive reinforcement trainers, particularly clicker trainers do not.
The vastly different paradigms of reward-based trainers and correction-based trainers
The overarching difference between positive reinforcement (aka reward-based) trainers and correction or force-based trainers is where they place the locus of control. Positive reinforcement trainers, especially clicker trainers work hard to positively motivate and empower their animals to give them ample opportunity for success while correction-based trainers work to control their animals through external devices and the threat of punishment.
The problem with punishment
Positive reinforcement trainers recognize that the fallout from punishment can be harmful to their animal’s mental health, their relationship and the training
process. They know that for punishment to be effective it must meet a number of criteria:
· It must be applied at the right intensity the first time and every time; otherwise the animal may get used to (“habituate” to) the pain and increasing levels are required to be effective
· It must be applied consistently; otherwise it can actually have the unintended effect of strengthening the undesired behavior
· It must be applied at the precise moment an undesired behavior occurs for the animal to make the intended association between the punishment and the undesired behavior
Punishment can result in:
· Suppressed fearful or aggressive behavior and worsening of the animal’s underlying emotional state
· An unintended association between the punishment and the punisher, another animal, object or situation
· Physical harm due to electrical burns, trauma to the trachea, damage to the nerves of the eyes or life-threatening pulmonary edema (AVSAB, 2007)
In a recent study, investigators found that dogs whose owners managed aggression with confrontational techniques such as yelling, growling at the dog, hitting, kicking, grabbing the dog and shaking, physically forcing the dog down onto its side, physically forcing a dog to release something from its mouth, alpha rolls (physically rolling the dog onto its back and holding it there), or staring developed increased aggression. The researchers concluded that these maneuvers were “fear-eliciting and may lead to owner aggression.” (Univ. of PA, 2009)
Do we really need to show our dogs who’s boss?
It is a misconception derived from outdated information about wolf packs that we need to dominate our dogs and be their “pack leader.” We now know, based on the work of Dr. Ray Coppinger that dogs, although close relatives of the wolf, aren’t pack animals as we previously thought, but rather scavengers that form loose associations with one another around food resources.
Of course all that has changed somewhat with selective breeding and integration into the human family unit. But one thing remains certain – dogs are dogs and humans are humans and we humans don’t do a very good job of imitating canines. Don’t think for a minute that your dog thinks you are his canine pack leader – he recognizes you as a species apart from his own and tries very hard to understand you as a human. We form a social unit with our pet dogs, but it is probably not best described as a “pack.”
Conclusion
If we give up the idea that we need to be the pack leader and dominate our canine companions, we should also be able to give up outdated training methods aimed at subduing them in favor of tools and techniques such as clicker training which bridge the communication gap and deepen our relationship.
While impeccable technique is no guarantee that correction-based training will have no untoward effects on our canine partners, the worst that can happen with amateur clicker trainers is that the dog gets a few ill-timed clicks and treats and the bond between owner and dog is strenghtened. Dog and human handler alike will still learn – and have fun doing it!
It is debatable whether correction-based or reward-based training works faster. If punishment is applied absolutely correctly, it will stop undesired behavior without having to be repeated. But since punishment also inhibits learning, new behaviors may be acquired much more slowly, requiring far more repetition than with positive reinforcement training.
Positive reinforcement (reward-based) training is not only humane, but highly effective without side effects. It is founded on well established principles of behavior science. We know that this type of training is valid and reliable because it works on virtually any animal and the same consistent results can be replicated time and time again.
The choice of which training method to use really comes down to one simple question. Do you want to make your dog obey using external control or do you want to motivate your dog through positive reinforcement of desired behaviors? If you want a partnership with your dog, a deeper relationship, if you want to develop an animal that thinks and is emotionally healthy, that is joyful and loves learning the choice is clear.
Training that rewards and builds on what a dog does right rather than what he does wrong is not only a training philosophy, it is a way of life. Clicker trainers for example, use the same positive reinforcement methods with their human students as they do with their canine students. If you can relinquish control to gain it you will discover the wonderful world of positive reinforcement dog training and when you do, you will find that what you can accomplish with your dog is only limited by your imagination and what your dog is physically able to do!
Cindy Ludwig is a
Karen Pryor Academy Certified Training Partner and owner of Canine Connection
LLC in Dubuque, Iowa www.dubuquedogtraining.com. She is a member of the Association of Pet Dog Trainers and the International Association of Animal Behavior Consultants.
References:
Finding Fuzzbutt Four: http://findingfuzzybuttfour.blogspot.ca/2011/10/different-theories-of-training.html
American Veterinary Society of Animal Behavior – Behavior Professionals Statement
http://www.avsabonline.org/avsabonline/images/stories/Position_Statements/behavior%20professionals.pdf
American Veterinary Society of Animal Behavior. (2008).
Position Statement on the Use of
Dominance Theory in Behavior Modification of Animals. http://www.avsabonline.org/avsabonline/images/stories/Position_Statements/dominance%20statement.pdf
American Veterinary Society of Animal Behavior. (2007) Position Statement: The Use of Punishment for Behavior Modification of
Animals. http://www.avsabonline.org/avsabonline/images/stories/Position_Statements/Combined_Punishment_Statements.pdf
Coppinger, R., & Coppinger,
L. (2002). Dogs: A New Understanding of Canine Origin, Behavior,and Evolution.
New York:
Scribner.
Halber,
D. (2009). Why we learn more from our successes than our failures, MIT News
Retrieved
from http://web.mit.edu/newsoffice/2009/successes-0729.html
Herron, M.E. Shofer, F.S., & Reisner I.R. (2009).
Survey of the use and outcome of confrontational and non-confrontational
training methods in client-owned dogs showing undesired behaviors. Applied Animal
Behaviour Science 117, 47–54 http://vet.osu.edu/assets/pdf/hospital/behavior/trainingArticle.pdf
Hiby, E.F., Rooney,
N.J., & Bradshaw, J.W.S. Dog
training methods: their use, effectiveness and interaction with behaviour and
welfare. Animal Welfare, 13
(1), 63-69(7). http://www.ingentaconnect.com/content/ufaw/aw/2004/00000013/00000001/art00010
Mech, L.D. (2008). What
ever happened to the term alpha wolf? International Wolf, 18 (4), 4-9.
http://www.wolf.org/wolves/news/pdf/winter2008.pdf
Open Letter
From Dr. Karen Overall regarding the use of shock collars. (2005). http://www.joelwalton.com/shockcollars.html
Overall,
K.L.(2007). Why electric shock is not behavior modification. J. Vet. Behav.: Clin. Appl. Res. 2 (1),
1-4. http://www.k9behavioralgenetics.net/resources/Articles/Why%20electric%20shock%20is%20not%20behavior%20modification.pdf
Polsky, R. (2000), Can aggression in dogs be
elicited through use of electronic pet containment systems? Journal of Applied Animal Welfare Science,
3(4), 345-357
Pryor, K. (2002).The poisoned cue. http://www.clickertraining.com/node/164
Schilder,
M.B.H., van der Borg, J.A.M. (2004). Training dogs with the help of the shock
collar: Short and long term behavioural effects. Appl. Anim. Behav. Sci. 85, 319-334
http://eldri.ust.is/media/ljosmyndir/dyralif/Trainingdogswithshockcollar.pdf
Resources:
American College of Veterinary Behaviorists. How to select a trainer – for owners. http://www.dacvb.org/resources/for-the-public/
APOPO: Detection Rats Technology http://www.apopo.org/home.php
Association of Pet Dog Trainers: Trainer Certifications: What Do All Those Letters After Everyone’s Names Mean? http://www.apdt.com/petowners/choose/certifications.aspx
Association of Pet Dog Trainers – Trainer Search. http://www.apdt.com/petowners/ts/default.aspx
Association of Pet Dog Trainers: Understanding Training Equipment Options – Collars, Leashes and Crates. http://www.apdt.com/petowners/choose/equipment.aspx
Karen Pryor Academy – find a great trainer. https://www.karenpryoracademy.com/find-a-trainer
Luck, Laurie. (2010). A Surprising Look at Balanced Training. http://smartdog.typepad.com/smart_dog/2010/04/a-surprising-look-at-balanced-training.html
Ludwig, Cindy. (2011). Clicker Training. What Is It?
http://animalinfo.com.au/fact_sheets/view/2/25/175/Clicker_Training_What_is_it
Ludwig, Cindy. (2011). Positive Reinforcement Training. What Is It? http://animalinfo.com.au/fact_sheets/view/2/25/177/What_is_Positive_Reinforcement_Training
San Francisco Academy for Dog Trainers: Academy-Certified Trainers. https://www.academyfordogtrainers.com/sc/Academy_Graduate_Referral_List.pdf
The Humane Society of the United States: Dog Collars: Which Type is Best For Your Dog? (2010). http://www.humanesociety.org/animals/dogs/tips/collars.html
Truly Dog-friendly Trainers. http://www.trulydogfriendly.com/blog/?page_id=4
VanArendonk Baugh, L. (2010).Should You Use No Reward Markers? Examining the Debate.
Add Your Comment