Positive Reinforcement Training Needs Oxygen
For many years, on-leash obedience classes for adult dogs were the only training option for dog owners. Certainly, teaching basic manners, such as to Come, Sit, Down, Stay and Heel on cue, improve many aspects of living with a dog. Training is all about improving communication. But training on-leash takes a very long time and provides little in terms of preventing or resolving behavior and temperament problems. Also, the transition from on-leash to off-leash can be extremely challenging.
Back then, dog owners wanted and needed so much more. And so, in 1982, along came off-leash, puppy socialization and training classes. The syllabus was enormous, comprising every aspect of behavior and temperament modification as well as teaching off-leash manners by utilizing half a dozen different positive reinforcement training techniques, selecting those that were easiest, quickest, and most effective for each exercise: Lure-Reward training to teach the meaning of verbal cues for instruction and as guidance when dogs err; Wait & Reward Training to teach inattentive hyperactive puppies and adolescent dogs to calm down and focus; Shaping to increase the duration of Stays, Shush, and Off; Autoshaping to decrease activity, barking, anxiety and stress; and Physical Prompting to solidify Stand-Stays and Sit-Stays.
Nowadays though most trainers use just a single positive reinforcement technique —Shaping — which is the most time-consuming and complicated of techniques that requires the highest skill set on behalf of the trainer, and so is the least suited positive reinforcement technique for companion dog training. Consequently, many positive reinforcement training programs are flagging and off-leash cued-reliability has decreased substantially.
Dog training needs to be revived to the way it was at the turn of the Century. Off-leash puppy classes that allow pups to socialize freely with unfamiliar people and puppies to: 1. To prevent predictable temperament problems, especially anxiety, stress, wariness, fearfulness, and maybe aggression towards unfamiliar people or dogs; 2. To enable the puppies to develop all-important bite inhibition and social savvy so that they learn how to act around other dogs that may not be so similarly socialized; 3. To enable owners to master off-leash control from the outset; and of course, 4. For trainers to be on the lookout for incipient signs of developing temperament problems, such as over-excitability, fearfulness, or pushy ply styles, and nip them in the bud.
Off-leash Puppy Classes
PREVENTION is the name of the game. Puppies are trained off-leash to make certain that they all develop stellar bite inhibition, and to prevent the development of otherwise predictable, quality-of-life-crippling temperament problems, such as shyness, anxiety, stress, and fear of people or other dogs. Aggression towards people and dog-dog reactivity quickly become non-issues. Yes, in the first session some pups show incipient signs of anxiety or fear, such as hiding and backing away but shy and fearful puppies quickly gain confidence to engage in play. Any dog-dog reactivity is simply nipped in the bud. Check out "Look Inside Our Classes" at siriuspup.com to see this in action.
Our extensive online resources (at DunbarAcademy.com) describe and illustrate simple solutions to prevent the BIG three behavior problems (housesoiling, destructive chewing, and excessive barking) plus separation anxiety.
Basic manners are taught off-leash within the distraction of play because lure-reward training makes teaching cued-behaviors both quick and easy. Many puppies learn to come when called from play and to Sit, Down, Stand and Stay in their first session. A solid vocabulary of reliable cue-instructions is essential for effectively dealing with misbehavior and no-compliance.
We let puppies be puppies, but gradually and progressively mold their temperaments until they are mannerly and confident and so, better prepared to navigate adolescence without anxiety, fear, or stress. Teaching puppies to overcome their fears and prioritizing the prevention of aggression towards people and dog-dog reactivity makes it possible for older puppies to continue in off-leash classes throughout adolescence and so, teach impulse-control, increase attention, and focus on off-leash reliability.
The ultimate goal is to produce an adolescent dog that promptly and reliably responds to single verbal request, when off-leash, at a distance, amidst distractions, and without the continued need of any training tool whatsoever, especially including, leashes, hand-contact, food lures, food rewards, and specialized collars, halters, or harnesses.
Puppy classes introduced: early socialization and handling, off-leash training, luring, positive reinforcement, and the dogs' and owners' points of view to the world of dog training, i.e., owner-friendly and dog-friendly dog training.
Positive Reinforcement
The success of puppy classes was the ease, speed and effectiveness of temperament modification and teaching basic manners using a wealth of positive reinforcement techniques: classical conditioning, lure-reward training, wait and reward training, shaping, autoshaping, and physical prompting. We embraced them all: Shaping a quick Sit into a lengthy Sit-Stay, building a momentary glance into avid attention, and increasing a second of silence into a cued multi-minute Shush. Autoshaping puppies in class and at home to settle down promptly, calmly, and quietly. (Feeding only from hollow chewtoys reduces activity and barks by 90% in just a couple of days.) Using the oppositional reflex of Physical Prompting to solidify Sit-, Down- and especially, Stand-Stays (for examination). Wait & Reward Training became the magical method of choice for puppies that didn't engage and for inattentive, hyperactive adolescent dogs. The game-changer though was Lure-Reward Training — by far the easiest, quickest and most effective way to teach cued-responses. Owners absolutely loved 'quick', 'easy', and 'effective', and dogs loved learning our language, so they understood what we would like them to do and so, avoid our frustration when they didn't understand.
Lure-Reward Training is as Simple as 1-2-3-4
1. Request, 2. Lure/Handsignal, 3. Response, and 4. PRAISE and maybe Reward. Such a simple little training sequence but packed with a whole bunch of science. Not only do Praise and Reward reinforce the desired Response, which increases in frequency and is more likely to occur in the future but also, the Reward reinforces the associative (predictive) relationship of the Request-Handsignal-Response sequence. After just half a dozen repetitions, the dog learns that when you verbally request "Sit" (difficult for dogs to understand), you will then lure/signal 'Sit' (easy for dogs to understand because they are fluent in body language), and when they Sit, you'll praise and maybe offer a reward — a food reward, or a life reward, such as, "Go Play".
Lure-reward training is the easiest, quickest and most effective way to teach ESL — English as a Second Language — to open communication channels. Once a dog understands the meaning of our words, training transcends to an entirely more communicative and effective stratosphere.
Phasing out Training Tools
Food Lures are replaced entirely by handsignals in the first session, whereafter handsignals are used as lures to teach the meaning of verbal requests. The only potential Achilles heel of lure-reward training is not phasing out food lures, causing food in the hand or bait bag to become a bribe.
Food Rewards are drastically reduced (to increase their reinforcing power) and largely replaced by far more powerful life rewards, such as walks, play, sniffing, and interactive games. Reducing food rewards is falling-off-a-log easy by asking more for less — more and more responses for fewer and fewer food rewards, and longer for less — longer and longer durations for fewer and fewer food rewards, i.e., by Shaping, and objectively monitoring progress by calculating Response:Reward Ratios. For perspective, first week in puppy class, the average Golden pup will gladly perform 20 puppy push-ups (Down-Sits) for the prospect of a single food reward, i.e., a Response:Reward Ratio of 40:1.
Are we offering too many food rewards and so, reducing their reinforcing power? Yes.
Response Reliability
Verbally cuing responses enables us to objectively quantify Response-Reliability as a percentage score, i.e., a single index that measures the dog's increasing 'degree of verbal comprehension' and 'motivation to respond'.
Training is not just 'based on science', TRAINING IS SCIENCE, and each training session generates oodles of irrefutable data as proof of the speed and effectiveness of training. Science dictates that we always ask the questions: Did training Work? How well? and How quickly? Without quantifying and monitoring progress, training is simply not science.
Quantification of comprehension is essential because the most common reason for non-compliance is that the dog did not fully understand our instruction in that situation. Response-Reliability is progressively proofed via several extremely simple troubleshooting exercises, for example, by asking a dog to Sit-Stay-Watch every 25 yards on a walk. Each time the dog Sits, they may be rewarded by saying, "Let's go!" to signal resumption of the walk, i.e., 211 training interludes during a three-mile walk, all reinforced by life rewards, and every training interlude occurring in a different scenario. After just a few walks, most dogs begin to generalize their response-reliability to any and every situation — anytime, anywhere, and no matter what is going on.
Off-leash, lure-reward trained Response-Reliability Percentages are off the charts, compared to other training methods, including the five other positive reinforcement techniques, and any aversive technique.
The Power of Words
Once dogs understand what we say, we may direct their behavior using verbal instructions. Additionally, and this is the Jewel in the Crown of my latest book, Barking Up the Right Tree, we can quickly and precisely resolve misbehavior and non-compliance in an obvious and courteous manner — by offering verbal guidance (correction) when dogs err.
A single comprehended word instantly curtails most misbehaviors, for example, Sit, Settle, Shush, Hustle, Steady, Chewtoy, Bed, Outside, etc., and without even raising one's voice. Similarly, non-compliance becomes a non-issue when using a simple technique that has been championed for ages by kindergarten teachers, enlightened grandparents, and I am glad to say, my father and grandfather — calm, gentle, persistent insistence. (FMI: Read the book!)
The Glory Days
During the heyday of off-leash puppy classes and lure-reward training throughout the 80's, 90's, and beginning of the Century, the highlights were the mannerly, well-behaved, friendly and confident dogs totally devoid of any quality-of-life-destroying temperament problems, joyfully playing and training together with proud and happy owners, devoid of the frustration of misbehavior and non-compliance. The obvious ease and success of dog owners and their dogs walking sidewalks and visiting dog parks soon convinced dog trainers that positive reinforcement is easier, quicker, and more effective than any aversive means.
Easier because reinforcement increases the frequency of desirable behaviors making them easier and easier to reinforce and further increase in frequency until most misbehaviors are squeezed out, whereas punishment decreases the frequency of undesirable behaviors making it harder and harder to 'catch the dog in the act'. Consequently, many dogs reserve their 'misbehavior' for times when they know they won't or can't be punished. Punishment unintentionally creates owner-absent behavior problems including, 'owner physically-present but functionally-absent' and 'owner physically-present but mentally-absent problems'.
Quicker because there's ONLY ONE RIGHT WAY — only one expression of the behavior that we need to reinforce, which takes just a finite amount of time, for example, eliminating in their toilet area, (which leaves less urine and feces for elsewhere). On the other hand, there are an infinite number of ways to express undesirable behavior, for example, in how many different places might a Yorkie pee in your bedroom? Infinite. And to punish the dog in each instance would require an infinite amount of time. Since finite is shorter than infinite, punishment-training takes longer than reinforcement.
More Effective because inhibiting undesirable behavior, (often normal and natural doggy behaviors), is only part of the puzzle. An equally, or more important objective is to get the dog back on track, doing precisely what we want, which of course, may be accomplished so easily with a single verbal instruction. To this day, every time I see an OTT, OOC, hyperactive, rambunctious and rumbustious, adolescent dog Sit on request, Shush on cue, run to their Toilet, or Find a Chewtoy to take to their Bed and Settle Down to chew, I feel a shiver of delight. Magic? Nope! Simply lure-reward training doing what it does best — communicating.
But... THE WORLD OF POSITIVE REINFORCEMENT HAS CHANGED
Over the past decade or so: off-leash puppy socialization and training classes have become the exception rather than the rule; classes have gone back on-leash; individual consultations (many by Zoom) are replacing classes; the syllabus has shrunk considerably; verbal communication is rare — few instructions, no verbal guidance, and limited praise when dogs excel. There's little attempt to phase out temporary training tools, such as, food rewards, leashes, halters, harnesses etc., (which have become permanent management tools). And most disturbing, ongoing quantification and quality control (that were built into lure-reward training) have disappeared and so, we lack objective data as to how well and how quickly training is working, or even, whether training is working.
In the behavioral sciences, rewards, punishment, training, and trainers themselves are all defined by their observable and quantifiable effect on behavior. However, these days, so many trainers focus on the nature of training, often decrying and denigrating other trainers for how they are training, and usually without offering a quicker and more effective alternative. My view is that of course dog training should be dog-friendly and owner-friendly; why on earth should owners be forced to treat their best friend like their worst enemy in the training arena? Dog training (teaching) should be entirely non-aversive. But also, we must offer proof that our preferred methods are at least equally, if not more effective at producing observable behavior change in the intended direction.
The Companion Dog Syllabus Has Shrunk
1. No longer is there an extreme and urgent emphasis on teaching bite inhibition and preventing behavior and temperament problems via environmental enrichment, handling and safely socializing neonates and very young puppies with unfamiliar people, socializing older puppies with other puppies, or molding temperament by building confidence in shy and fearful pups and toning down pushy play-styles. Consequently, fear, stress, anxiety, reactivity, and aggression are rife once more.
2. In terms of teaching manners, positive reinforcement has lost half its syllabus by neglecting to classically reinforce the antecedent cascade (Request-Handsignal-Lure) that precedes the desired Response. In essence, so many dogs are denied the opportunity to learn our language. Luring to teach ESL, phasing out training tools, and quantifying response-reliability are no longer priorities, aside from those trainers that still teach off-leash classes and especially those who compete with their dogs.
These days, positive reinforcement almost entirely focusses on reinforcing behaviors that are offered spontaneously and progressively refining criteria until the behaviors approximate some 'final performance'.
Teaching ESL is No Longer Priority
By no longer positively reinforcing the predictive association between antecedents (verbal requests and handsignals) and the desired behavior, teaching cued-responses went out the window, and trying to teach our dog specifically how to act without using words has become time-consuming and cumbersome. Also, without cuing responses, quantifying response-reliability has become next to impossible. Without objectively monitoring progress (effectiveness), positive reinforcement training has become more of an art than a science.
When we instruct a dog to "Sit', just half a second later we can ask, "Is the dog sitting?" Yes or No? And so, with repeated trials, calculate response-reliability percentages, as a precise, combined index of 'degree of comprehension' and 'level of motivation to respond'. But if we don't instruct a dog to "Sit", we cannot objectively quantify reliability and so, we have no realistic view of our control over our dog, when walking on-leash, when off-leash in a park, or when the dog is running out an open front door.
Without teaching the simplest of cued responses from the outset, for example to Sit-Stay-Shush-Watch, we have no hope of offering verbal guidance to help dogs quickly overcome their fear, stress, anxiety, hyperactivity, reactivity, and aggression. The absence of verbal guidance when dogs go off track has always been the massive Achille's heel of those reward-training techniques that do not teach cue-responses from the outset but instead, wait for spontaneous expression of specific desired behaviors.
Not being able to help dogs quickly to overcome their fears is just not fair for dogs (or their owners). Dogs are hurting and so our verbal guidance should be immediate.
It is essential to change how a dog feels about other dogs and people to change the way they act. But that takes time. Much quicker is to change how dogs act (to be quiet, still, and focused on us) to change how they feel (calmer). A calm and controlled dog presents a much better optic to other dogs and people, one that changes their feelings and their perceptions, causing them to feel less threatened by your dog and so, threaten your dog less.
PRAISE???
I think deprioritizing teaching cued-behaviors (ESL) for instructions and corrective verbal guidance is just plain silly, but even sillier was replacing PRAISE with a 'click'! And for no reason. Saying "G'dog" is both a precise 'marker' and a natural primary reinforcer in its own right, and one that acquires enormous secondary reinforcing power during everyday life. And a secondary reinforcer that keeps both hands free to pet and hug your dog.
I know a lot of you might accuse me of clicker-bashing. Well yes, I am. I really don't like gizmos of any type in training, and I just don't see the point of using a clicker. I prefer to use my voice, as did my great-grandfather when he and his horse won a straight-line ploughing competition with no reins.
However, I have always been a massive fan of the concept of Shaping and it has always been a major ingredient in our puppy training curricula (as I mentioned above). The Breland's Animal Behavior was my choice as a class textbook when teaching behavior at UC Berkeley. And so, even though I didn't like the practice of 'clicking' and 'bonking' dogs, I helped promote Karen Pryor's and Gary Wilkes' very first clicker-training workshop because I knew that Shaping offered a wonderful non-aversive alternative to the Force Retrieve (the default method at that time for obedience, hunting, and protection training). Shaping was such a brilliant way to reinforce extremely inhibited behaviors, for example to teach a retrieving breed to retrieve a dumbbell or pheasant's wing, if the retrieval object had been previously 'poisoned' by aversive feedback.
As we all know, shaping is the method of choice for creating behaviors that are not in a dog's normal behavior repertoire, (of which there are very few). However, shaping (with or without a clicker) is an extremely time-consuming way to 'teach' naturally occurring behaviors, such as Come, Heel, Sit, Down, Stand, Stay, Rollover, Bang! Beg, Back-up, High Five, etc., etc., that may all be lure-reward trained on cue, in a flash.
And of course, a massive Acchille's heel, shaping is next to hopeless for resolving misbehavior and non-compliance. (Turn your back on the dog? Distract the dog? Teach a DRO or DRI?) It is so much easier to teach a dog to Sit or Down on cue and use these simple instructions as emergency commands when we are at a loss for what to do.
A Shadow
The facts are evident: today's positive reinforcement training is but a shadow of its former self. The big three losses would be communication, quality control, and feeling.
Dogs have lost the reassuring help of our voice as clear instruction and guidance when they are not sure how to act and they have lost our praise and congratulations.
Without quantifying the speed and effectiveness of training and generating oodles of data points, training is no longer science. Lack of effectiveness (response-reliability) has prompted a resurgence of temperament problems and allowed aversive techniques to make a comeback.
Today's positive reinforcement has become time-consuming, unnecessarily complicated, clinical, impersonal, and virtually wordless. Our dog is not a rat in a cage! Our dog is our companion in life.
I think we must urgently return to the days when we embraced all positive reinforcement techniques and especially focus on teaching ESL. In Barking Up the Right Tree, I objectively compare the relative pros and cons of each reward-based technique in terms of ease, speed, effectiveness, and enjoyment. All techniques have pros and cons, and depending on each individual exercise, some are easier, quicker, more effective, and more enjoyable than others
SO, WHAT DO PEOPLE WANT?
Given that all dog training (teaching) should be dog-friendly and thoroughly enjoyable for dogs and owners alike, I think most dog owners simply want to learn the easiest, quickest, and most effective way to teach their dog the meaning of the words they use as instructions to direct their dog's behavior at home, on walks, and in parks. People want to know how to raise their puppy to learn household manners, and as an adult, to welcome household visitors and thoroughly enjoy meeting strangers and unfamiliar dogs on walks and in parks. They don't need many words. A reliable Sit is a simple solution for so many annoying misbehaviors.
What do new puppy owners need to know?
Above all, that training doesn't happen by magic, and that without preventative intervention, as puppies grow older, they will naturally develop a few utterly predictable behavior problems, such as housesoiling, destructive chewing, and excessive barking, and temperament problems, especially 'fear of the unfamiliar', i.e., fear of unfamiliar people, fear of unfamiliar dogs, fear of unfamiliar places/scenarios, and fear of being left alone.
The good news? Preventing these potential problems is falling-of-a-log easy:
1. During the first couple of days, set up an errorless housetraining and chewtoy-training schedule, that also decreases barking and prevents separation anxiety.
2. Over the first month, socialize the puppy with numerous unfamiliar people prior to 12 weeks of age, safely at home. The Open Paw Minimal Mental Health Requirements for Puppies suggests puppies meet 100 people during the first month at home. Bring the people to the puppy! Ask every guest to bring a friend or two to handle and train the pup. (Outdoor shoes must remain outside.)
3. Enroll in an off-leash puppy training class to bump-start your puppies play so they develop bite inhibition and learn social savvy around other dogs, and owners can have a hoot learning how to train their puppies off-leash within the play session.
To facilitate breeders, veterinarians, pet stores, trainers and shelters disseminating educational information to prospective and new puppy owners, I have triaged all the essential puppy raising exercises by urgency and importance into three very short, how-to, downloadable picture eBooks and several online courses that are all available for FREE download to anyone from: https://dunbar.info/freepupbooks PLEASE SHARE THESE BOOOKS! — with all your doggy friends, neighbors, customers and clients in your community, urging them to share the books with all their doggy contacts in their communities, across the country and around the world. By all means, please post the books (or links) on social media, or upload them to your own websites for free download. Please help us spread the word.
Our goal is that every puppy gets the education that it needs and deserves. And why do we focus on puppies? Because every shelter dog was once a puppy just begging for an education and the gift of confidence to live in the world of people.
What do dogs want?
I think dogs desperately desire open communication channels between people and dogs. Dogs thrive when their owners are aware of, or know how to evaluate their dog's needs, feelings, and preferences. Additionally, I think dogs especially love being able to understand our needs and feelings, such as how we would like them to act and so, avoid our frustration (and sometimes anger) when they unintentionally 'misbehave' by just acting like a dog, i.e., by displaying their normal, natural doggy behaviors at inappropriate times or in inappropriate settings. Instead let's teach them where to eliminate, what to chew, when and when not to bark, and how to greet people and other dogs.
As always, communication is the solution, and inter-specific communication is most magical. I've been training dogs for over 50 years, yet I still get the same thrill whenever I ask a dog to Come-Sit-Watch... and it does... promptly, willingly, and happily, with a loving gaze and furious wag.
Our New Year's resolution is to make 2025 the most amazingly exciting year in dog training by redoing what we did in 1982 — re-popularizing off-leash puppy socialization classes and lure-reward training and ruthlessly promoting both at the largest veterinary conferences in the world, to get companion dog training back to the way we were.
This time the transition will be much quicker because along the way, we archived most everything we did in the past few decades at DunbarAcademy.com — hundreds of articles, blogs, and iWoofs, and hundreds of hours of seminar and training workshop videos for dog owners and dog professionals and so, training trainers and educating new puppy owners should be much easier, everything you need is in DunbarAcademy.com.