One of the traps of good intention, training without aversive punishment and with unquestionable love and affection is to be remiss in teaching that good moments also run alongside moments of “bugger” … that was a disappointment.
Within a few hours of arriving into the world M.O.B. (moments of bugger) become the fabric of life. These pups were enjoying an afternoon snack, when Mother turned upside down. Those titties were not where expected, the scent of them was definitely in the air, but climbing was needed to secure one. The small little sausage bottom left is still experience a MOB whilst the others are tucking in. Just as well she could not see at this age!
Time is an experienced learner. He is set up to learn that I would like him to go around the bin clockwise and then stand on the platform. Watch the video through once to get the idea of what I am teaching (this is running faster than reality).
Here is the first video:
As the bin is moved further away from me to the end of the platform it seems perfectly logical for him to take the short cut and come direct to the platform with the bin on his left.
Here is the second video with his logical choice:
This is not my desired behaviour, I want to eliminate this. I have two options.
Initially on the video I demonstrate withdrawing reinforcement for arriving on the platform when the bin was on his left. No click, no food, I even fake disappointment. Then I set him up to try again by refreshing success for the correct choices: bin on your right. We arrive at the same critical decision point and he makes the same choice: the shortest route to success is with the bin on my left. Again, no click, no food.
This may be called selective reinforcement, but it does not explain to him why he is not getting success. It is NOT an NRM. He completes the behaviour to find there is not success.
There are 2 behaviours:
Both behaviours in the chain, one after the other need to be successful. He is focussed on arriving on the platform, not relating the first behaviour of how he goes around the bin to the change in his success. This is using the consequence (no reward) to attempt to extinguish the undesired behaviour.
BUT, the desired behaviour of arriving on the platform is in danger of becoming extinguished, where as the first behaviour – arriving with the bin on your left, is the undesired behaviour. By withdrawing reinforcement at the end of the chain I am not explaining what IS required. He repeats the same logic, (not really an error at all), again. No evidence of “no reward” having an effect.
Here is the Moment of ….
Instead I use a cue (raising my foot) to block the undesired behaviour and leave him to make a choice. He adjusts what he does and goes around with the bin on his right to gain success on the platform. This is extremely high probability from the previous successes.
This is a chain of two behaviours, two successes. The decision to adjust what he does and go around the desired side needs feedback and arrival on the platform also needs feedback. Both are critical to the success of this chain. I give verbal “well done” on his point of decision and click AND a fuss, and a re-set treat for arrival on the platform.
You could choose to do this the other way around, click at the point of the desired decision and give verbal feedback on arrival. But knowing my dogs they are likely to come to an enquiring stop when hearing the click.
The important points here:
At the point he was prevented from going the logical route I describe as a “moment of bugger”. He is then left to make alternative choices – which he had plenty of time and reinforcement to have learned, it was very fresh in his mind. He made the logical choice only once more.
For me this is ethical prevention of an undesired choice. The undesired aspect is my choice, and I spend a lot of consideration setting up the learning in the direction that is beneficial to all of us.
This is an example of using a MOB. It is an important strategy when the dog’s logical choice is likely to make life harder for us and for them. I want to prevent using extinction as a consequence to undesirable choices.
A clear example is me leaving the dogs in the garden when I go through the front gate. Sometimes the dogs come with me for training, sometimes not, I go to feed the chickens. The conditions look very similar. I always teach the dogs that access through the gate is never from behind my legs, I will go through, turn and face the dogs and invite them to join, or not. On the “or not” occasions they will experience disappointment, a moment of bugger. The closed gate prevents the error. I do not let them come through an then attempt (hah, not likely) to extinguish that behaviour. The gate not opening is a cue (the antecedent) of disappointment.
Notice I use the terms “gate closed”, not “closing”. I do not use the movement of the gate into the dogs as a punishment for trying to get through. Initially when I step through the gate I throw treats behind me. This avoids any discomfort a puppy could experience if they tried to follow me and let me teach the wait behind me whilst gates and doors open. Once through and the gate is shut, I turn to face, then if I want them, either all or a single dog to accompany me they will be called to join
I add a verbal cue to this “not today”. I use it carefully, with calmness. I could have used my verbal cue when Time was about to make his logical choice with the bin, instead I used my foot and hand to block the choice.
We can learn to use a variety of preventative blocks: a flat of hand when the dog is coming towards us. Most of us use this with a puppy just before they make a logical choice (the error) of jumping up at us, a particularly valuable strategy if I am carrying a cup of hot tea. We recognise the preparation to jump with that slight dip of the shoulders and put our hand forwards, either as a visual block, or if within reach a steadying hand to the shoulders. It does not punish the behaviour of jumping or showing affection, desire to greet, just a prevention “not at this moment”.
Arriving at the end of the lead should be just a MOB, not an aversive reaction from the person to pull back or yank. A fence is just an MOB if what you want is on the other side, puppy gates the same, the crate, the same, the opening of the car boot, the same. By careful forethought we can pair these moments of disappointment from an arranged environment to a verbal cue “not today”. Rarely will these behaviours come under the category of “never”.
Rules for MOB