What Could be Better than treats?: Understanding the most Difficult Training Challenges
You didn't hear it from us, but it's true: sometimes... treats are NOT the best rewards for our dogs! Actually, this is a key insight in being successful with force, pain, and fear-free training methods.
Often starting in adolescence, well-trained puppies who previously did ANYTHING for treats suddenly stop coming when called, pulling on leash, counter-surfing, and greeting humans with explosive enthusiasm. These are some of the most frustrating problems in dog-human history, and it may seem the only solution is to throw out our treats and resort to more coercive methods. What happened?!
Well, unfortunately, sometimes undesirable behaviors are reinforced even though we did not give a treat. This is because external circumstances or the situations we are in trick us into allowing our dogs access to reinforcement in their environments even though they are doing behaviors that we do not want. For example, every time our dog leaps up to greet a person, they really are getting a treat: human attention. Another example: no one in the house ever gave the dog a cookie for leaping up on the counter, and yet he does it -- because he just can't forget when it paid off big on Thanksgiving last year!
We can think of these sorts of behaviors as akin to self-reinforcing behaviors, and they have two important
distinctions from any others:
1. If they are occurring, they are being reinforced. And by this, we mean both possible interpretations:
a. They are being reinforced with each performance, therefore, frequency (and even intensity) are likely to go up.
b. The have been being reinforced in the past. Dogs are not just suddenly disobedient or stubborn: if a behavior is repeated, it's been reinforced somehow until now, whether the human noticed or not.
2. It is costly for a dog to abstain from performing them, as they miss out on an opportunity for reinforcement.
Here, Jinbei easily poses for the photo to get a treat - this is his street, he's bored of it! In contrast, that's a tall order for his visiting terrier friends (who sit perfectly at home). For them, it's hard not to check out all the novel activity and smells.
These are precisely the reasons why self-reinforcing behaviors frustrate so many owners. It is helpful to contrast them with a sit. Dogs do not naturally approach humans head on, orient themselves directly in front of and facing humans, drop their rear to the ground, and remain unmoving while offering eye contact. An untrained domestic dog might more likely approach indirectly, and stay standing to maximize their stature or to prepare for a speedy retreat should it be necessary. Despite this, sit is among the most common interactions between domestic pet dogs and humans. So why do domestic pet dogs perform this unnatural behavior so frequently and with such conviction? The answer seems obvious: we reward this behavior in various contexts (either we ask for it, or the dog has made the connection that sit tends to get them what they want). However, an essential but often overlooked piece of the equation is that the dog has nothing to lose - and a yummy treat to gain - in performing this easy behavior. Here, Piper has perfectly aligned himself to worship at his Love The Dog shrine... totally natural.
Most dogs begin pulling on the leash, greeting humans with excessive enthusiasm, and ignoring commands outside in adolescence because reinforcement in these contexts has not been carefully managed. Desirable alternative behaviors were not identified and reinforced to the exclusion of these less desirable ones from early on, and by adolescence, young dogs have had a few months to develop strong reinforcement histories them. Undesirable behaviors neglected early in life (or neglect of training in general) exacerbates the reinforcement that builds naturally each time these behaviors are rehearsed. Puppies are developing strong reinforcement histories for pulling with each step they take forward to the dog park that there is tension on the lead. Most owners never see the problem developing, as starts small, but like any other behavior, once reinforced it increases in frequency, and often, intensity. A pit bull who is celebrated for timidly putting their paws up on a human’s knee in greeting as a puppy is condemned as a problem dog when that greeting has escalated into a classic pitty body slam six months later.
When you think about it, these undesirable behaviors are sometimes inadvertently trained more frequently than desirable behaviors, and many of the rewards are higher value than treats! (Even our dedicated CEO Luna will abandon her treats to greet her human dad when he arrives home - no matter who gets knocked over and out of her way in the process!) Why should your adolescent dog choose to come to you for a paltry treat when continuing to play in the dog park has proved so much more reinforcing?
Just like when a dog sits in the home and they get a treat, they are getting what they want in all of these problem behavior scenarios. The bad behavior works. Also important is another fact: unlike when the dog sits at home, in these scenarios, dogs are frequently losing something if they instead attempt to perform a behavior more desirable to their humans (not pulling, sitting politely to greet, passing up on dog/human greetings, coming when called, “leave it”, etc). It is not only reinforcing for them to do the undesirable alternative, it is costly for them to not do it.
So starting in puppyhood, and with increasing significance for growing or adult dogs who have reinforcement histories for undesirable behaviors, reinforcement must be managed holistically. In fact, controlling reinforcement is the key to controlling all behavior: if your dog believes the reinforcement they want is contingent on the behavior you desire, they will do it. And if your dog believes that access to the best reinforcers comes through engagement with you, they will work enthusiastically on your team. This is really no different than ensuring your dog receives a treat for sitting, but can be challenging to get the hang of at first, and requires a bit more strategizing.
Owners need to detect potential problem behaviors early, stop to ask themselves “How do I want my dog to behave in this scenario?”, and then ensure that they trained and only reinforce what they want (ideally before they need to contend with a well-established reinforcement history). This is especially the case in high stakes games like access to the dog park. It also requires a more sophisticated understanding of high versus low value reinforcement. There is a hierarchy of reinforcers in any context, and for any individual dog in different places and different stages of their life, it’s not always the treat in your pocket. So when the best reinforcement in a scenario is not in your pocket, you need to figure out how to get it there. In other words, you need to control access to the best reinforcement, and your dog must believe these reinforcers are contingent on good behavior, and that they come through teamwork with you.
On his first boat ride of the year -- with a leash? From Piper the Drahthaar, not even waterfowl are safe, and hunting is intrinsically reinforcing. Fun water retrieves in the past makes - and mismanaged access to them - make the prospect of catching a nearby cormorant or duck much better incentive for leaping overboard than much we have to offer him for hanging around.
The good news is, treats can still really help you in whatever strategy you choose to realize the above, and a science-based, certified professional trainer can show you how! (One good trainer database can be found at http://www.ccpdt.org/dog-owners/certified-dog-trainer-directory/)