Understand how to use “Differential Reinforcement” rather than “Punishment”.
Differential Reinforcement is a critical and essential skill for Professional Pet Trainers. It is the alternative to using punishment when working with pets. Rather than punish one behavior, we target and build an alternative, more appropriate behavior while withholding reinforcement from the current problematic behavior, thus reducing the likelihood of it being offered.
This targeted (goal) behavior may be an incompatible behavior, an alternative behavior, any other behavior, or a low rate of the existing behavior – see where I am going with this.
So, with a dog who jumps on people, and No, we do not just ignore it!
- First, target the jumping for extinction – withhold the reinforcement while simultaneously.
- Targeting any other behavior for reinforcement, any behavior other than jumping (DRO) – Differential Reinforcement of Other behavior
- Then, the target for reinforcement is four feet on the floor (DRA) – Differential reinforcement of alternative behavior.
- Then target for reinforcement an incompatible behavior – the dog sitting (DRI) Differential Reinforcement of an incompatible behavior.
- Then increase the levels of difficulty
- Then generalize the new, more appropriate behavior to all environments.
Where To Start
You cannot target a more appropriate behavior or decide which Differential Reinforcement protocol you will use to build the new behavior until you really understand the problem and what you want to achieve. You will then need some measurement and baseline data in terms of duration, frequency, or intensity. It’s always important to set yourself up so you can track the success of any intervention you make or lack of success so you can change course. Once you have data and a baseline measurement, you can choose the most appropriate DR protocol.
For Example – A Dog Barking
If barking is an issue for the family, you only need to focus on one dimension to resolve a dog barking. Think about this: Does the dog bark at too many things (frequency), for too long (duration), or too loudly (intensity)?
So, determine what do you want to change or what makes the most sense to focus on. For my dogs, it’s duration. I have no problems with them barking, but I would like them to stop after a few seconds. So I teach them that after a duration of 6 -10 seconds, thank you, and we are all done.
- Do you want to change the duration of the barking?
- Do you want to change the frequency of the barking?
- Do you want to change the intensity of the barking?
Another example. House Training Your Puppy
Think about house training your puppy. You wouldn’t focus on the intensity or duration of the puppy’s urination. You would probably target reducing the frequency, reducing the number of times your puppy pees in your home!!! So you would set up a management system to ensure the puppy goes outside and is reinforced for this while targeting the frequency of inside urination for extinction. It’s no longer happening, reinforcement, the relief to the puppy of going inside, is withheld! (bathroom behaviors are negatively reinforced, there is an unpleasant feeling and then relief felt once the puppy urinates) .
Differential Reinforcement involves targeting a behavior for reinforcement while simultaneously targeting another behavior for extinction, withholding the reinforcement that has established and maintained a behavior. Just using extinction can create frustration, and this can temporarily strengthen the problematic behavior; it also can create fallout and stress. Two things we work to avoid.
Don’t fool yourself; your pet is always doing multiple things at the same time, so when you are working with Differential Reinforcement, you need to know: what am I reinforcing? This is why timing your ‘yes’ or ‘click’ is so essential, or you may end up reinforcing the very thing you do not want.
So when you are reinforcing your targeted behavior, the other behaviors that are going on, non-criterion events, are resulting in no reinforcement for the animal (or are they). There are always concurrent contingencies, that is, multiple behaviors the animal could be performing. In layman’s terms, there is always a “choice,” even if it seems like just one is the only realistic one.
So, any time you are not reinforcing in accordance with your DR protocol, you are extinguishing these other choices. Removing the likelihood the pet will offer them in the future.
- Know what your target problem behavior is!
- Know your goal; that way, if you do not achieve it quickly, you will know what you are reinforcing is inaccurate and will be in a position to track back and get it right.
- Track data so you can tweak your training protocols
- Sometimes you will start with DRO – reinforcing any other behavior offered. Then you may move to an alternative behavior and then to incompatible. Start where you get success to make it fun for both you and your pet.