Five Days from Fear to Fun – Classical Counterconditioning.

I decided to spend some of our holiday time working on a whistle recall.  This is when a dog learns to come to the sound of a whistle.  Pamela Dennison has a number of resources on how to teach this skill for anyone who might be interested.

Unlike other whims, I remembered to grab my camera.  When I blew the whistle for the first time, Karma tucked her tail and ran.

I probably should have been a bit more thoughtful in my introduction of the whistle.  However, when you have confident dogs you get accustomed to taking liberties.  Sometimes, Karma bites you in the butt.

Before anyone sends me hate mail, we didn’t plan for this to happen.  We adjusted immediately.  Fear can unexpectedly happen.  I decided to share that one incident because frankly it happens to owners all the time.  It’s the clicker that scares a dog, the dropped cooking pan, the ratting of aluminum foil, fireworks or the whirl of a new kitchen appliance.  Sometimes our dogs get scared.

I thought it would be interesting to put together a time lapse of Karma’s responses as they changed over the course of five days.  It would show the change in her body language and a real time account of what we actually did.

We used classical counterconditioning.  This means that each time the whistle sounded, Karma received a really amazing goodie, mainly turkey.  There were no conditions on her behaviour.  I did reduce the volume and duration of the whistle until she grew more comfortable.  I made sure she was some distance away from me during the first few repetitions.  This made the process easier on her.

The sound of the whistle was initially aversive to her.  Aversive means that it’s something that the dog will seek to escape, avoid or postpone.  At first, you can see that Karma clearly is looking to escape from the sound of the whistle.  By pairing the sound of the whistle with goodies, her emotional response changes over time.

Just because something is aversive for a dog today, it does not mean that it cannot be changed.  Things that used to be aversive can become appetitive and vice versa.  It might have been easier to shelf the whistle recall idea entirely.  Perhaps for some dogs that is an option.  However, as Karma goes to more and more events, I know that eventually someone is going to blow on a whistle.  I do not want her to be startled and frightened if there is something I can do today.

Some people seem to think that classical counterconditioning is hard or doesn’t work. They say it takes a lot of time and effort.  While there are some basic rules to keep in mind, by following the rules, busy people – like myself – really aren’t doing that much work at all.

People fail to realize that long breaks between sessions are beneficial – perfect for the busy dog owner.  In its simplest form, you feed the presence of the trigger.  In this case whistle equals goodies.  You do need to be mindful of things that can block or overshadow the conditioning.  Otherwise, it’s really that simple.  Trigger equals treat.

One thing we did was hide a treat somewhere in the house while Karma was out in the yard.  Later, when she was back inside, I’d blow the whistle and surprise her.  This was done because I was going out of my way to ensure that the whistle predicted goodies.  Not “hand in pocket” or “open fridge door” or “standing in the kitchen.”  It really has to be the trigger that equals the food.  The cleaner you work, the faster the association will happen.

The results in the following video involved the following:

  • Five days of classical counterconditioning
  • Six reps per day approximately (only 2 on the 24th…was busy … it was Christmas eve.)
  • Ten seconds of work per repetition.


Total Training Time
Less than FIVE MINUTES.

I will probably re-visit whistles a few more times.  It pays to finish the job by generalizing the conditioning.  We could work in a variety of locations or work with different types of whistles.  Then we’ll be ready to switch back to our initial plan of a whistle recall.

Yes, I do realize that some dogs have traumatic experience in their past.  Not all problems will go away in five days.  There are dogs with global fears and there are dogs with a history of trauma and abuse.

For the most part, many dogs are just normal dogs that occasionally get scared of one thing or another.  We are the ones who are our dog’s worst enemy. We over think what needs to be done.

We become embroiled in minutia and complexities that are not relevant.  We worry that feeding a fearful dog will reinforce fear.  It doesn’t.  We wait for “good” behaviour instead of just feeding the trigger.  I clearly fed a dog that was afraid and I fed a dog that was not necessarily behaving.  Yet, the fearful response disappeared.  Sometimes it pays to let go of intuition and just trust the science.  Feed the appearance of a well defined trigger.  It works.

Additional resources:

Awesome Dogs Shareables Counterconditioning Collection

Reactive Dogs on Facebook

Fearful Dogs on Facebook

CARE

14 thoughts on “Five Days from Fear to Fun – Classical Counterconditioning.

  1. Hi Yvette!
    Thank you so much for sharing all this valuable information on your blog. I enjoy your view on dog training so much and your knowledge inspires and motivates me! Can I ask you something about CC? My 5-month old puppy Yuki (she actually looks a lot like your beautiful Karma!:)) is scared of other dogs and barks in the apartment whenever she hears a dog barking or even just walking by outside. She also barks at other dogs when we’re on walks. It’s no frustration or play barking – she’s cleary scared and nervous around other dogs, so I’ve been trying to counter-condition her. Now here’s my question: You’re saying that no matter what the dog is doing at that moment, feed her a treat after the trigger. What if Yuki is already barking at the other dog while I’m feeding her the treats? I always try my best to be quick and feed her before she starts barking, but it’s not always doable. Can I still give her the treats when she’s already barking or would that reinforce her for barking? Thanks 🙂

    • Great question. Here’s an example for you. Imagine you feed a cat canned food. Run the can opener, feed food. You buy little cans. You don’t eat canned human food. So can opener almost always means “cat food.” What would you expect to happen? The answer is a cat that runs for the food dish insanely when the can opener runs. The opener reliably predicts the food.

      What if you ate lots of canned food? What would the cat learn? “Sometimes can openers predict cat food, but not so much.” They might saunter. Others might only run for the dish if it’s around dinner time. The rest of the time, the association won’t be there. Something is telling the cat “there are exceptions.”

      Do you worry about when you run the can opener? “Oh no, I can’t run the can opener. The cat is not behaving.” Actually, I don’t think most people even look to see what the cat is doing.

      Can openers predict food. The more consistent that is, the strong the association. So no, you are not reinforcing behaviour. What you’re doing is allowing the dog to get over threshold. Not ideal. But it can happen. If possible, try to find a way to reduce the intensity of the problem while you work on things. White noise might be an option. Giving food over threshold can absolutely still work. It’s just not “nice.” And things take longer.

      If, a quirk happens, the second phase of most training after classical conditioning is regular operant conditioning. So if you do have a dog that tries a happy yip, you can very easily fix that later.

      • Thanks for the reassurance! Sounds logical.
        I’m trying hard to always keep her under threshold, but this has become harder and harder every day. She seems to be just fine, then all of a sudden starts barking and growling. We are strictly avoiding distracting or arousing environments, doing only walks around the block now. I let her sniff around a lot on a long loose leash, BAT style, and do my best to keep her relaxed. However, she seems to be stressed and anxious very often and I don’t know why. For example today, after our walk, she was sitting in the back yard peacefully, chewing on a stick. All of a sudden, she jumped up and started barking nervously at something on the opposite side of the yard. I don’t know what she was barking at. It was raining and there were some sounds that weren’t usually there, like water dripping and leaves moving, but there was no person and no dog. I wasn’t able to refocus her on me so we just went inside. These things happen a lot lately. She also often gets the zoomies after our walks, which I read could be a sign of stress as well. (Or is it just puppy behavior?) And ripping out grass like a maniac, not eating it, just ripping out bunches of grass and spitting them out again. She seems very stressed to me. So I feel like plain DS/CC is not sufficient here. I wouldn’t even know what to CC her to most of the time, because I don’t know what started her barking in the first place. I know I need to boost her confidence and teach her calm behaviours, and to settle down more easily. That it’s important to have a routine, and to not over-stimulate and not under-stimulate her. But it’s all so overwhelming and I am very frustrated. Sometimes I feel like I don’t have enough patience, and then I feel bad because of that, and then I think that my dog can probably feel all of my stress, which makes her even more stressed. That puts a lot of pressure on me and doesn’t make things easier. Sometimes I think maybe I’m too much of a perfectionist and too easily stressed myself to be able to teach a dog how to relax …

        • I can’t comment on what you’ve been doing because BAT has multiple variations and people experiment with it so there are an infinite amount of varieties. Escape from fear is an old and well researched idea. Pavlov’s student started the process and called it (erroneously) Pavlovian fear suppression. It’s not a technique that I use because if a positive association forms, it’s to the warning, not to the trigger.
          I’d say that your best bet is to do really well executed CC/DS. It’s far more powerful that you think. Slugs learn with classical conditioning. It’s suspected to be one of the oldest forms of learning from an evolutionary perspective – necessary to be a functioning organism.
          My suggestion is “give your dog a vacation from training and observe. Write notes.” There is a pattern there OR your dog is sensitizing. Which means they are getting jumpy.
          All dogs with a functioning brain can learn with CC/DS. Step one is to do a functional assessment. What is triggering the problem? What is supporting it? Let your dog’s stress level come down by closing down their world for a little bit so they aren’t so jumpy. CC/DS is possibly one of the most powerful tools out there. Do it well and keep it simple.
          Sensitizing tends to come from the repetitions being too close to one another. Not a long enough break between. It happens when the trigger is leaning on the aversive side during habituation.

          • Hi Yvette, sorry it took me so long to respond. I appreciate your advice very much. I’m absolutely with you regarding the power of CC/DS, I don’t doubt it in the slightest way. What I was trying to say is that I’m overwhelmed by her different triggers and often I’m not sure what to CC her to. I know for sure that she is scared of other dogs, so that’s easy. She sees a dog, treats. She hears barking, treats. (“easy”. ;-)) But when she growls at something in the darkness, what is it that she’s scared of? A noise? Moving leaves? Or if it’s a certain smell that sets her off, like the smell of another dog in the air. I guess in situations where I can’t perceive the trigger before she does I could only watch her body language very closely and feed her treats as soon as I see her getting tense, preferably before she starts barking. Would you say that’s the correct approach? However, when she’s that tense often she’s already over threshold and doesn’t take treats.
            Also, would you say that if she gets treats for a lot of different triggers on a walk, do the treats lose their meaning or power? As in, she starts associating walks with a lot of treats, but not the individual triggers? Sorry if that’s a stupid question. I feel like your answer would be, if my timing is accurate and perfect, she will still associate the individual triggers with the treats.
            Okay, but here’s the problem. There is another factor, which is loose leash walking. We’ve been training that following your video where your son is feeding Kip treats for walking on a loose leash. After about 3 weeks of training she now heels pretty well in the area around our apartment, but with a lot of treats still. (When we leave that area, she starts pulling and forgetting about me and the treats, but that’s another story.) Do you think all the treats for loose leash walking could be in the way of CC? Like:

            Loose leash – “yes”. Treat.
            Loose leash – “yes”. Treat.
            Loose leash – “yes”. Treat.
            A dog – treat. Treat. Treat.
            Moving away from the dog.
            Loose leash – “yes”. Treat.
            A noisy train – Treat. Treat. Treat.

            So many treats for so many different things! Is this still proper CC?
            Thanks so much for your help.

            • Sorry, I just reread the last part of your previous answer. I guess that partly answers the first part of my question. I’m already doing my best to isolate her from most triggers to keep her stress level low. That has helped a lot. For example, staying close to our apartment when we’re on walks, in an area where she already knows most corners. I’m just a little worried that if we only go to the same area every day, next time we need to go to the vet or to another location that’s further away, she won’t be able to deal with the situation because she’s not used to going new places anymore. I know the answer is slow desensitization, taking a few steps at a time, introducing her to new places slowly and gently, and with lots of food. I’m just not sure if I’m doing everything slowly enough, but also not too slowly. You know what I mean?

              • One of the most clever comments I’ve read from someone in a long time. YES it can interfere – the LLW. And yes, too many triggers can start getting muddy.
                Make a list. Hit them in an order that makes sense. Things you can’t manage. Get them off the list.
                It’s possible that at night, she’s edgy. She could be growling at nothing. Like when you jump easily when you hear noises at night after watching horror movies. That’s call sensitization.
                The interference from LLW is called blocking effect.
                Keep it simple. Make a list. If there are too many things on it, split them up. LONG breaks help reduce sensitization. People often work way too hard and do too many reps.

                • Thanks for the clarification. Just one last question. Since our LLW training is apparently getting in the way of CC, I’m now seriously thinking about just putting the LLW training on ice for a few months until she’s more confident outside. I think when she pulls it’s sometimes excitement and lacking impulse control, and sometimes it’s fear. So it wouldn’t make sense to keep on training LLW before tackling her impulse control and fear. But since I still need to go on walks with her, I’m considering just always attaching her leash to the front clip of her harness from now on and eventually when I restart LLW training with her in a few months I could attach it to the back clip. Do you think it would be unwise to postpone LLW training? She’s 6 months old now and probably won’t be easier to train once she’s hit her teenage phase. I’m really hesitant, but I feel like training LLW AND doing CC is just too much for now.

  2. That whistle was aversive to me, too :))) Nice to see her go from avoidance to happy bushy tail 🙂

    I am among those people who think of CC as hard… the reason is that my association to CC is what I used it for in my dogs… traffic (I live in a city), other dogs (again, living in a city I cannot fully control how often or how far away other dogs appear). Which doesn’t mean I wouldn’t use CC next time, but it definitely wasn’t easy and it wasn’t done in 5 days.

    • The whistle is in my mouth. Camera in front of me. So they’re really close to one another. It’s not that loud in “real life.” Just to be really clear on that. I wouldn’t blow any whistle ever that close to a dog’s ear. So yes, it sounds louder than it really is.
      Although I will say that we dealt with car chasing in about 6 planned sessions. Spinning, barking, lunging, screaming. But no, not all dogs will be fixed in 5 days. The better the mechanics, the faster it does go.

      • I meant the sound of it… maybe all dog whistles are like that, I don’t know.
        For us it was fear, not chasing. Fear of everything on wheels (even baby strollers) and fear of the sound of traffic. With a whippet who didn’t value food all that much for the first year of his life. And a first time dog owner who had to learn about thresholds from trial and error 😉

        • Dog whistles are fairly quiet and high pitched.
          And yes, first time out can be HARD. Our Kiki years back was reactive in the car. Kaya took me 4 hours to get out of his crate the first time he came home. Feral pup. Was certified as a therapy dog by 1.
          Really, I know it’s a strange concept that perhaps it’s not that hard to do. But as you get the hang of it, you get this moment where you realize how easy it is. But that comes with time.

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s