Book: Teaching with Reinforcement
It is not easy
Author: Kay Laurence
Confusion probably causes more behaviour to break down than a single punisher when working with positive reinforcement. We perceive ourselves as working in an environment where we only want to reinforce behaviour and we avoid aversive or deliberate punishment. But our own lack of self-discipline can cause effective punishment when we fail to communicate clearly what is required, and whether it is correct or not. We default to using words, which dogs find hard to discriminate between, and contradict what we say with how we use our body language.
We stand at the door telling the dog to go outside, but our body language clearly states that we have no intention of going outside. Which is the dog to believe?
We tell the dog to stop barking but our agitation clearly tells the dog there is something amiss.
We tell the dog “good boy” but our disappointment clearly tells the dog we are frustrated, possibly with ourselves, but we “taint” the intended reinforcement.
Connection, compliance and behaviour begins to “bloom” when the communication is clear, without doubt and easily understood.
The variety of different reinforcers and punishers available to us is extensive. The repertoire is often very individual, what is reinforcing for one person is not for another and the same for dogs, and what reinforces a behaviour in one situation does not in another.
Our purpose is to remove uncertainty
Our second step is to learn the tools of using the reinforcers to affect the results we desire.
Reinforcers and punishers are threading their way through all our behaviours. It is the nature of behaviour to change, and the outcomes of these changes are memorised, or we make the same mistakes repeatedly. The outcomes will either maintain or fade the behaviour depending on whether they are reinforcing or punishing.
But behaviour rarely happens in isolation. Several behaviours can happen at the same time and reinforcers and punishers can be attached to the wrong element. The dog may bark at the back gate, and be punished by a spray of water. The association can be made with the location, being at the back gate, the stranger walking by, (and the dog now becomes even more anxious when they hear the approach of strangers), or the barking, or you holding a bowl of water. You can only look for evidence of the effect of the punisher on the behaviour that diminishes.
You think you are reinforcing the paw wave, the dog is in the sit position. When you cue the dog to shake a paw when he is standing there is no response. There is no reinforcement history for that behaviour – the paw wave AND the sit, have been paired and only reinforced together, not in isolation.
We teach the dog to drop (down), this is usually achieved somewhere near our feet. When given the cue around the other side of the sheep the dog returns to your feet to drop. What is reinforced is what will be strengthened. What is reinforced is the entire pattern of behaviours, location, speed, pace, and emotion.
Behaviours connected with the reinforcing behaviours can feel the “wash” of the reinforcer, particularly where the reinforcer is strong. Part of our skill in using reinforcers effectively is to make very sure which behaviour they attach to.
What is reinforced is what will be strengthened.
What is reinforced is the entire pattern of behaviours, location, speed, pace and emotion.
Begin to become aware of what reinforcement you get out of the process of training or enjoying your dog. This may vary from the pleasure of seeing your dog alert and interactive, or it may come from the approval of other people when you demonstrate the final behaviour. It may come from both, but if one part of the teaching process is higher in reinforcement than the other you will become biased and effective at that element at the expense of the other.
We have learners who inadvertently reinforce undesirable behaviours. Either because the dog is mirroring them, their state of anxiety or stress; or they receive more social attention from friends and experts when their dog has a “problem”.
Become aware of your own reinforcer, and manage it.
Exercise : Going forwards
At this stage it is important that we develop observation skills and try to unravel the flow of reinforcers and punishers.
Look for behaviours that are changing:
- Are they the same as they were this time last year, last week?
- Have they become stronger or become weaker?
- Can you detect what consequence is affecting the behaviour?
It is often our nature to see what is getting worse, but remember to also notice what is improving and getting stronger
Learning how to teach with reinforcement is often about becoming observant and making the small, subtle changes rather than picking up a clicker or treats.
It is the change to our view of behaviour that has the most significant impact.