DunbarAcademy.com

We all want to decrease misbehavior and non-compliance, but aversive punishment is unnecessary, and even when effective, insufficient, (we still need to get the dog back on track). For example, an aversive stimulus might get a dog to stop bouncing, barking, and pulling, but is it sitting? Let's just troubleshoot "Sit" in that scenario. Or, an aversive stimulus may stop a dog from looking at or chasing a rabbit or squirrel, but is it bringing back the dumbbell or pheasant? Let's just troubleshoot a dog to Take, Hold, and Fetch any object in any scenario.

Concentrate on teaching and proofing dogs to do what you would like them to do and then, there is little opportunity for them to do what you don't want them to do. The quickest and easiest way to get a dog to act in a specific manner is to verbally cue the dog. I have always stated that: For resolving behavior problems and non-compliance, lure-reward training is easier, quicker, and more effective than aversive punishment. Let me explain why.

Lure-Reward Training is Easier
Lure-Reward Training is comparatively simple, especially in terms of feedback. It is next to effortless and requires minimal brain power to say, "Thank you" or "Go play". On the other hand, using aversive stimuli effectively as punishment requires adherence to six stringent criteria and a considerable mental and physical skillset.

Additionally, lure-reward training becomes easier as training progresses, whereas punishment-training becomes more difficult. The more we reward, desirable behaviors increase in frequency, making them easier to reinforce, thus creating a positive feedback loop. Eventually, desired behavior squeezes out misbehavior, which makes reward-based training the best way to reduce and eliminate undesired behaviors and hence, the need for punishment.

Effective punishment on the other hand, decreases the frequency of behaviors, making it more difficult to complete the program. To eliminate undesirable behavior entirely, dogs must be "caught in the act", and punished in every instance,

Lure-Reward Training is Quicker
To reduce and eliminate misbehavior effectively using punishment, the trainer must punish the dog in every instance, until desirable behavior becomes the last behavior standing. But let's consider: In how many different locations might a Yorkie chose to pee in your bedroom alone? On the floor, on the carpet, in a closet, in a sock-and-underwear drawer, in a dirty clothes basket, in slippers, under the bed, on the bed, on a pillow, under the covers, and so on. And how many times in each place? that's a lot of times in a lot of places — an infinite number of times and places.

Therefore, for punishment to be effective, the trainer needs to be present to punish the dog for an infinite number of improvisations, which of course requires an infinite amount of time.

However, when teaching a Yorkie where to pee, what to chew, where to lie down, or when to shush — there is ONLY ONE RIGHT WAY, which requires only a finite amount of time to teach. Since finite is shorter than infinite, lure-reward training is quicker than punishment-training.

Lure-Reward Training is More Effective
For punishment training to be effective, the trainer must adhere to the six criteria, which is next to impossible outside of a laboratory setting, wherein computers train captive animals. Most important the owner must immediately punish the dog for every transgression. Now let's say they punished our little Yorkie every time he peed in the house, what exactly have they taught the dog? Not to pee in the house? Or not to pee when the owner is present? Well, the dog must pee and so, will likely pee when the owner is absent. And since his little bulging bladder is likely full to bursting, the Yorkie will likely pee the moment the owner leaves the house.

The single biggest crack in the punishment agenda is that a single inconsistency on behalf of the owner and the dog will learn those situations when it can act like a dog and not be punished. In a sense, many owners unintentionally train their dogs to eliminate, bark, chew, dig, and hyperactivate only at times when they are absent — owner-absent misbehavior. Some people call this separation anxiety. I think in many instances, a more apt and descriptive term would be separation celebration.

Dogs also learn that they won't be punished if the owner is functionally absent, for example when cooking, holding a baby, or when in the shower. I had one case wherein, on hearing the shower running, the dog would head straight to the kitchen and jump through the flip-top lid of the trash can for a smorgasbord of smelly delights.
Additionally, dogs learn when their owners are mentally absent, for example, when checking texts, reading, watching the telly, or simply day dreaming. I can remember one evening when two psychologists were sitting on a couch, one reading a book on cognition and the other on short-term therapy, while I sat in an armchair across from them, thumbing through a Malamute breed book, while Omaha, my first Malamute, calmly, methodically, yet brutally shredded the end of the couch. Three smart human brains all on a different planet, and one engaged canine brain directing destruction.
It's just so much more effective, to offer a food-stuffed chewtoy, so that every piece of extricated food reinforces (autoshapes) what the dog is doing, i.e., lying down, not barking, and chewing the chewtoy instead of the couch. Or, to teach a dog to pee and poop on cue, and then empty them out every hour on the hour and especially, last thing before leaving the house.

The Icing on the Reward-Training Cake
The six stringent criteria for effective punishment are either unnecessary, inadvisable, or embedded within the process of lure-reward training.

Instruction is the essence of lure-reward training. Teaching dogs the meaning of words, enables us to offer clear instructions and ongoing, corrective guidance. Dogs commit no 'crimes' as we teach them. Single-word instructions can be used to teach dogs to cease and desist and instruct them what to do instead. And poof! The undesirable behavior is history. So easy, so simple, and so amazingly effective.

Also, the immediacy and consistency of feedback are not critical issues. In fact, I always emphasize, the longer we delay giving a reward, the greater the number and the longer the duration of reinforced responses. Similarly, inconsistency increases the power of reinforcement, anticipation, and motivation, and hence compliance. When reward-training, inconsistency a welcomed virtue. The gift of reward-training is that consistency is neither required nor advised, which is a boon because people are not famous for being consistent.

Given a stellar trainer, punishment training can be effective, but lure-reward training is always so much more effective, especially for eliminating noncompliance and misbehavior, and getting the dog back on the straight and narrow. Lure-reward training is absolutely the way to go.

Excerpted from a transcript of the Science-Based Dog Training (with Feeling) seminar at Dunbar Academy.com.