No Class ( second installment)
You should have completed last week's homework at this point. If you have not, please click here and begin as soon as possible. Home work is listed at the bottom of the Week Three Class Outline.
*If you have not written your thoughts about Positive Punishment, please do so now and email them to before the next class. What I am looking for is does it work? When? Why? How can you measure if it works? What can you use in place of Positive Punishment?
OK - Let's Get Started!
Positive is Good Right?
In dog training, there are many theories and methods that apply to training. Premack's theory, B.F. Skinner, Pavlov's Conditioning ( AKA 'classical conditioning'), Operant Conditioning, etc. etc. Enough to make your head spin! :)
What we are using in class is primarily :
Operant Conditioning: Skinnerian conditioning – an animal learns that its behavior has consequences. Discriminative stimulus (your command) – Response – Consequence.
There are 4 possibilities: Positive Reinforcement, Negative Reinforcement, Positive Punishment, Negative Punishment
Now, many people ( trainers included) make this harder than it is... so let me simplify. Forget about POSITIVE = GOOD and NEGATIVE = BAD Try THIS instead:
POSITIVE means to ADD 'something' as in "+" NEGATIVE means to REMOVE or take 'something' away as in " - "
REINFORCEMENT means to give support to or ensure the previous
behavior will be repeated. For example: Teresa asks Mandolin to
"sit", as soon as Mandolin 'sits' Teresa rewards the correct
action with a consequence - a treat, praise, etc. This is an example of
__________ REINFORCEMENT. If you said POSITIVE REINFORCEMENT, you are
correct. It is POSITIVE because Teresa was ADDING something. This
behavior - the sit- is likely to be repeated because something "good"
is added when the behavior is performed ( treat.)
Negative reinforcement is less easily understood.... an example would be if your dog is barking and you were to clamp your hand over his muzzle ( which I do NOT endorse, by the way!) and when he stops barking you remove your hand. So you are removing ( negative) your hand when your dog is quiet (reinforcing the 'quiet'.) This behavior is less likely to be repeated because it gets rid of something "bad" ( your hand on their muzzle.)
PUNISHMENT means to provide a consequence for an unwanted behavior so it is less likely to be repeated.
From the previous homework assignment, provide a paragraph demonstrating to me that you understand POSITIVE PUNISHMENT. Be sure to EMAIL me your paragraph! :)
Negative Punishment is best demonstrated by thinking of taking away something when the dog behaves badly. If Putter is barking for attention, and Marion walks away from Putter, ignoring her barking, Marion is demonstrating NEGATIVE punishment. Putter wants Marion's attention, but Marion takes it away ( there's the negative) when Putter is barking. When Putter stops barking, Marion rewards the behavior with POSITIVE Reinforcement by giving Putter praise or just by returning to Putter's side.
Assignment One:
Think about both + and - reinforcement and punishment. When you are working with your dog, whether in practice sessions or just spending time with him, take note of what type of punishment or reinforcement you are providing.
Sometimes you may be providing REINFORCEMENT without realizing that you are doing so. Most commonly is the jumping up situation. Dog jumps up ( why?) we use our hands to push him down, or we might just start petting them without thinking (so much easier than bending down :) In both cases we are REINFORCING the behavior we are complaining about.
OK, so back to your dog. Is your dog having problems with the stay? With allowing a stranger to approach without jumping, lunging, barking? How can you provide punishments or reinforcement to get the result you want.
Assignment Two: Variable Rates of Rewards
Remember we are going to be eliminating food rewards in this class. This is something that requires WORK and Patience. Please complete the handout on rewards that can be used in place of FOOD for your dog and bring it to the next class.
Now, to start FADING the food reward, we need to stop rewarding for EVERY action with a food treat. You should be pairing your food treat with a verbal cue, right? Such as "yes", "good", "ok", etc.
Now from what I have seen all dogs are doing extremely well with the SIT. So it is time to start a Variable Rate (VR) of food treats. Here is an example: Ask for a sit (command), Dog sits ( action), Food Treat ( Reward). Now the next time, you don't give a food treat, you give your verbal marker ('good', 'yes', etc). 3rd time, you give a verbal and a food marker, 4th time, you give a verbal, 5th time a verbal, 6th time a food marker. Now you shouldn't do 6 sits in a row... you should space them out within a minute or two of each other, a good way to do this is while walking, or in between play.
I would like you to start fading food markers with the following commands. **If your dog starts to respond less quickly, add a marker back for a few sessions and then start fading again** Be sure to EMAIL me if you are having any problems with this assignment!
Variable Rate of Rewards for the following commands:
Sit Down Stay (short stays without distraction or distance.)