Desensitization & Counter Conditioning Concepts
Learn about terminology, training formula, and troubleshooting to help your reactive dog by using counter conditioning techniques.
Vocabulary
-
Learning is a change in behavior based on past experience.
Learning is always happening regardless if anyone is actively teaching.
-
The gradient refers to changes in the intensity of a stimulus
Typically the stimulus goes from mild to strongest i.e. a sound gets louder and louder, petting gets harder and faster
May refer to distance, proximity, volume, duration, pressure, etc.
-
The point that, when exceeded, produces an effect or response, typically an overreaction to the stimulus.
Sometimes referred to as a “choice point.”
May refer to distance, proximity, volume, duration, pressure, etc.
-
A consequence that increases the probability of a behavior being repeated.
Primary Reinforcers:
Food, Water, Play, Praise
Secondary Reinforcers:
Markers: Clicker, “Yes,” “Free”
Anything that predicts a reinforcement
It must be reinforcing to the dog. You may think they like a piece of food or rough pets but it may not be rewarding enough to them to reinforce a behavior.
-
A consequence that decreases the probability of a behavior being repeated.
It is generally aversive or unpleasant, but it shouldn’t be harmful.
It must be punishing to that dog. They need to find it unpleasant enough to not want to repeat a behavior that causes it.
Punishment is rarely used or needed, but if it’s used it must be:
Immediate
Consistent
Effective
-
The specific reinforcement that matters most to the dog in that situation.
If a child is approaching a dog the reward for the dog may be space and distance from the child.
A hungry retriever may want food!
An attention starved Labrador may want praise and attention.
A Basset Hound in the woods may want the opportunity to sniff.
These are helpful to use when they can be identified and controlled.
-
Avoidance blocks exposure to a stimulus.
Avoidance removes the opportunity to rehearse problematic responses to a stimulus - this is important!
Maintain distance from people or dogs on walks
Confine to another room prior to visitors coming inside your home
White noise machine to mask sounds that trigger them
-
Low intensity exposure to a stimulus, followed by a gradual increase in stimulus intensity (stimulus gradient).
Intensity of stimulus must be below the dog’s threshold.
Often paired with Counter Conditioning.
Different from “habituation.”
-
Counter Conditioning is the process of changing a dog’s emotional or behavioral response to a specific stimulus.
It requires active training of some type, you need to work on it a lot.
Scary Thing + Treats = Joy
Counter Conditioning is a form of classical conditioning. They are basically the same thing. Counter conditioning is used when the dog already has an emotional response to a stimulus.
-
Classical Conditioning forms an association between two stimuli.
The sound of a can of wet food being opened initially means nothing to the dog, but soon they make the connection that the sound means they’ll be getting food.
This is what we are using when we teach dogs markers like “Yes”.
Classical conditioning is happening all the time naturally.
-
Operant Conditioning forms an association between a behavior and a consequence.
It requires a voluntary action by the dog. They must choose to sit for example and the consequence is they will receive a reward (positive reinforcement).
They may charge at the door when it’s opened only to have it shut before they can get out. The bolting behavior caused what they wanted (outside) to go away to punish that behavior (negative punishment).
Options for Operant Behaviors that are Helpful
Identify
Stimulus
What is causing the problem behavior?
Problem
What is the problem behavior? Fear? Anxiety? Frustration?
Goal
Would you like them to be comfortable with the stimulus, interact with it, move away from it, etc?
Avoid
Avoid encounters with the identified stimulus triggers until you are ready to work on them in a controlled setting.
Set up controlled exposure to the stimulus
Make sure you are working below threshold
Create conditioned emotional response to the stimulus using counter conditioning
Reinforce specific behavior responses like disengaging, looking at you, sitting, etc.
Repeat
Continue step 3 while slowly increasing the stimulus intensity over time.
Formula
Additional Applications
Graduated departures for separation anxiety
Aggressive reaction to passersby
Reactivity to dogs on walk
Greeting family members
How would you use the formula to work on these problems?
Identify the stimulus, the behavior and what you’d rather they do
Being left alone or walking away
Proximity to people
Proximity to dogs
People entering house
Avoid exposure to the stimulus
Avoid leaving the dog alone unless you are able to work in a controlled setting
Avoid getting too close to people
Avoid getting too close to other dogs
Avoid having the dog around when people enter the house
Set up controlled exposure to stimulus
Take a step away from your dog and then reward them
Have a helper walk perpendicular to your dog at a distance where the dog is under threshold. Give dog treats while they are moving and for other good behaviors like disengagement.
Have a helper and their dog at a distance where your dog can handle seeing them. Reward them for seeing the dog and looking away from the dog.
Get your dog used to all the triggers that lead up to someone coming over and reward them for staying calm. Knocking, door opening, etc.
Repeat while increasing the intensity, proximity, etc.
-
“Real life” training is more likely to result in the dog going over threshold and becoming reactive.
You need many repetitions to solidify the response you want and often real world situations don’t allow for this.
Less skilled handlers need things done slowly so they can focus on their steps and the dog, real life scenarios usually don’t allow for this.
-
It can be tempting to use food as a “bribe” for desired behavior.
You may need to start with a food lure to teach the behavior but then adjust the timing so food comes after the behavior.
Food should be given at moment dog notices the stimulus (Classical Conditioning). Or in response to a specific behavior (Operant Conditioning).
If the food is given too soon then it isn’t contingent on the presence of the stimulus or the specific behavior.
If the food is given too late it may not have the desired effect or may reinforce an unintended behavior.
-
A behavior that is practiced is more likely to be repeated.
Avoidance is especially important for emotionally driven behaviors (fear, anxiety, etc.)
If the undesired behavior “works” there is no incentive to learn something new. If they bark at the mailman and the mailman leaves every time, then barking works so why do anything different?
-
This is dependent on handler’s ability to read and interpret Body Language
You need to be able to identify the signals that are predictive of a reaction from your dog so you can be proactive.
Proactive training requires lots of practice
-
“He knows it now, so I don’t need to continue to reward, right?” WRONG! At least at first.
Your dog has probably practiced the reactive behavior thousands of times, so they will need thousands of positive reinforcement counter/classical conditioning of the stimulus to balance the scales.
Must practice to the point where the desired behavior becomes the default response to the stimulus.
It may require continued reinforcement (treats) to maintain the behavior.
-
Handler may have trouble thinking ahead, and then acting accordingly.
Handler may have a natural tendency to react to things after they’ve happened rather than think ahead.
It can be more difficult for very reactive dogs and reactive owners!
-
Desensitization
Requires active training, below threshold, and awareness of reinforcement patterns
Habituation
May happen naturally for some dogs and stimuli, but not all. Don’t rely on it.
A dog is less likely to habituate to a stimulus where an unwanted behavior has already developed.
-
Handler may project their understanding of the desired behavior onto the dog, but the dog doesn’t know it yet.
The handler may have a lack of attention to the stimulus gradient or intensity and expose the dog to a level they aren’t ready for yet.
Handler may be embarrassed or impatient and so they forgo avoiding or working below threshold.
-
Consider the number of practiced repetitions in relation to:
Duration of the problem
Emotional intensity of response
Handler’s ability to reinforce emotional state or behavior pattern
You are unlikely to have enough opportunities to practice in “real life” situations, set-ups are usually required!
-
What happens when you get a reward (chocolate, money) when you notice a spider, but the spider continues to linger or get closer?
It’s helpful to set-up sessions so there is a release of “pressure” (move down the stimulus gradient) after reward or correct behavior.