How dogs learn
Operant Conditioning focuses on using either reinforcement or punishment to increase or decrease a behavior. Through this process, an association is formed between the behavior and the consequences for that behavior.
Positive Reinforcement is giving your dog something he likes when he is doing a behavior you want in order to make that behavior more likely to occur. Example: your dog sits and you give him a treat.
Negative Reinforcement is removing something the dog finds unpleasant to make the desired behavior more likely to occur. Example: you pull up on your dog’s leash trying to get him to sit, as soon as he sits you release the pressure.
Positive Punishment is adding something the dog does not like when he is doing something you do not want him to do in order to make the behavior less likely to occur. Example: your dog barks so you spray him with water.
Negative Punishment is removing something the dog likes when he is doing a behavior you do not want him to do in order to make the behavior less likely to occur. Example: you ask the dog to sit, he jumps up instead so you remove his opportunity to get a treat.
Classical Conditioning occurs when a conditioned stimulus is paired with an unconditioned stimulus. In the case of dog training a marker word “yes” is paired with food. The word “yes” previously had no meaning to the dog but after numerous pairings of “yes” to food, the word “yes” now serves as a predictor to food. Classical Conditioning is essentially learning by association.
A sound, a visual cue, an object becomes associated with and predicts a primary reinforce. Example: in marker training a “yes” predicts a reward.
A primary reinforcer is something a dog likes. Example: a treat, play, a walk.
Stages of Learning
- Acquisition/ Learning Phase
Dogs should be taught with the least amount of distraction possible. Do not use corrections during this phase.
After the dog has learned the behavior, start applying it in many different places and in different ways. Example: ask your dog to sit while you are sitting down or not facing the dog; ask your dog to sit while you are in a new environment such as new rooms, your backyard, and other places with minimal distractions. Do not use corrections during this phase.
This is where we improve and fine tune behaviors. We add the 3 D’s of dog training: distance, duration, and distractions. This is the state where the dog should have a firm grasp of the command and be reliable at providing it. At this stage it is acceptable to correct the dog when he does not provide the behavior requested. This will be discussed in much greater detail during the course of your training.
Behaviors must be practiced for the life of the dog or they will disappear.
Rules of marker training
Yes = Reward Marker/Release
Good = Duration/Keep doing what your’e doing
Eh Eh = No Reward Marker
No= Punishment marker
Break=general release/ no reward given
Use “Yes” to mark the behavior the instant it happens. This tells your dog exactly what they did to earn their reward. “Yes” also serves as a release word since after the dog is rewarded they are not expected to hold the behavior.
“Good” provides feedback to the dog to let them know they are doing a great job and will be released and rewarded soon. We can give the dogs rewards after we say “good” to encourage them to continue the behavior.
Once the behavior is on verbal cue (example: when we say “sit” the dog sits) we use “good” to add duration to the behavior and “yes” to release from the behavior.
“Eh Eh” is our No Reward Marker, this is a form of negative punishment. When the dog is doing something we don’t want and we remove the chance to get a reward. Example: we ask our dog to sit, they jump up instead, we say “Eh Eh” and put the treat behind our back.
No is our punishment marker. We use no followed by a correction when a dog fails to perform a known cue or when they are doing a pack behavior we need to correct
Break is used to release a dog at thresholds such as doorways or in and out of cars. You can release a dog to their meal or release a dog out of their crate. This can also be used if a dog doesn’t perform a known behavior well and you just want to start over or transition to another behavior.
Most behaviors should be taught in this order:
- Lure/reward- The first step in luring is to take a food reward and hold it very close to the dog’s nose. Next you move your hand in the direction of your desired behavior as your dog follows it. For example; luring downwards to encourage a down. The instant your dog achieves the desired behavior you mark it by saying “yes” then giving your dog the reward.
- Hand signal/reward- Hand signals are created by using the same motion as your lure, the only difference is now you are not using food. Once your dog is in the desired position mark with “yes” and then reward.
- Verbal cue/ hand signal/reward- Now that your dog can get into position without the use of food we can name that behavior. You would say the cue ex, “sit” pause a split second then give your hand signal, mark with “yes” and reward. During this phase we can focus on fading our hand signal.
- Verbal cue/reward- At this point we give the verbal cue and wait for a couple seconds for the behavior to occur. Once the behavior occurs mark with a “yes” and reward
Shaping. Shaping is rewarding successive approximations towards the behavior you are training. For example if you were shaping “fetch” you could start by rewarding any interest in a ball. Mark and reward for looking, then touching, then putting their mouth on it, then picking it up, then moving with it etc…
Capturing. Capturing a behavior is naming and rewarding a behavior that the dog is naturally doing. For example, when your dog is laying down you would say”down” then mark and reward
- Continuous – reward every time the dog performs the behavior correctly. This is used during the
Acquisition/ Learning Phase through the Generalization Phase
- Variable – rewarding at random. Example: Vegas slot machine. This is used during Generalization Phase and Adding Distractions.
- Differential – A random reinforcement schedule based off of performance. This is the last schedule and is used to fine-tune the dog’s performance. Rewards should be given for excellence.
- Prey drive– Prey drive is any part of a hunting sequence. Chasing, tugging, eating, stalking, biting, shredding, shaking toys. This can be used well in training to motivate dogs to play. This could also be a source of behavior problems like chasing cars…
- Pack drive– Pack drive is the dogs will to find out where they fit in a social setting. All dogs know where they rank in human family settings and dog family settings. When you have a dog that knows they rank the lowest you won’t have dominance issues. When your dog is not given leadership they assume they are the leaders themselves and you can develop a myriad of issues.
- Defense drive– Defense drive is the dogs will to protect himself and stay safe. This presents itself in behaviors such as guarding, leash reactivity, and fear based behaviors.
Knowing drives is important for determining how to motivate our dogs. Rewarding with food or play is in the prey drive. Rewarding with praise or petting is in the pack drive. Rewarding with distance or space is in the defense drive.