Dog Forum banner

1 - 18 of 18 Posts

·
Registered
Joined
·
316 Posts
Counter conditioning is a form of classical conditioning, which is about involuntary reactions. In counter conditioning, we're working to make something inherently negative predict something amazingly wonderful. Over time and with many pairings of these two things, a dog can learn that something that used to be unpleasant is now amazingly wonderful, because an association between these two things has been made.

Example: Cheese makes my dog very happy (an involuntary response). Garbage trucks make her somewhat scared. My dog has learned that garbage trucks predict cheese. Therefore, garbage trucks have come to make her feel happy.

Operant conditioning is about voluntary actions. Voluntary behaviors can become more or less likely to reoccur over time, based on their consequences. When most people think of "dog training," they are picturing operant conditioning -- the process of teaching a dog to do some things more frequently, and other things less frequently, by controlling the consequences of chosen behaviors.

Example: My dog sometimes sits when she sees a garbage truck (a voluntary action). If I give her a piece of cheese after she sits, I would expect to see that behavior increase in frequency over time.

Both are learning processes, and can overlap or even occur simultaneously. But there are important differences, and often good reasons to choose one over the other for a particular training situation. Hope that helps!
 

·
Registered
Joined
·
2,418 Posts
A very basic difference is that classical conditioning builds an association between an involuntary response with a neutral stimulus. Other than Pavlov's dogs, Watson's Little Albert experiment is probably the most well known. Albert was allowed to play with a rabbit. While he was playing, Watson would bang together two metal bars scaring him. Soon, Albert cried in fear every time he saw a rabbit (or, reportedly, other small furry animals).

Operant conditioning is the process of changing voluntary behavior by manipulating consequences. If you're given $10 every time you wash dishes, you'll probably start washing dishes more often. Conversely, if you're docked pay at work every time you open FB, you're likely to access it with decreasing frequency. There are a few more scenarios, but I think there's a sticky the explains in more detail.

Counter (classical) conditioning is the process of changing an involuntary reaction to a stimulus. In dog training, counter conditioning is often used when dogs have a negative association with a particular stimulus (e.g., fear of people). By pairing something the dog doesn't like / is fearful of (like people) with something the dog loves (like chicken or tug), the dog should start to develop a more positive response towards the stimulus.

counter (classical) conditioning deals with associations and emotions
operant conditioning deals with observable behaviors

Does that answer your question?
 

·
Registered
Joined
·
2,418 Posts
It's like @SnackRat and I were thinking with the same brain. :) I even typed a few lines about Pavlov being on your shoulder, but deleted it because I thought it was more information than the OP wanted.

One minor quibble is the phrasing of counter conditioning as working with "something inherently negative." The stimulus might be negative for the individual or may have taken on a negative association due to past events, but it's not necessarily inherently negative. I think that using the adjective inherently has the potential to lead some people (no one here, of course) to misinterpret, minimize, or ignore negative responses when the scary thing isn't obviously scary.
 

·
Registered
Joined
·
140 Posts
Can someone explain the difference to me?
I found it hard to get my head around the Learning Theory ( Classical Conditioning and Operant Conditioning ) :eek:
I learnt a lot from reading How Dogs Learn by Burch and Bailey.

This is my take on it and I hope to be corrected if I am wrong .:D

Simply put, Counterconditioning is undoing bad experiences by changing the dog's emotions. If your dog previously liked other dogs but was then attacked and now dislikes them , you counter condition by offering the dog something it loves, like chicken or cocktail sausages, every time you see another dog. Your dog will eventually associate yummy things with the other dogs and start liking them again.
Classical conditioning is to do with emotions and reflexes eg sweating, blinking , startling. Its not to do with voluntary behaviours such as barking ,growling or wagging a tail.

Operant conditioning is learning by consequences. We used this to train dogs to sit, come, stay etc. They learn to behave in such a way as to obtain rewards and avoid punishments. When a dog learns that she receives a treat from you when you call her, it increases the chance that she will come back next time for the treat. If your dog jumps up at you and you yell at her, it decreases the chance of her not jumping up at you to avoid being yelled at. It varies as to what each individual dog finds rewarding or unpleasant.

Sometimes Classical conditioning and Operant conditioning over lap .
 

·
Registered
Joined
·
519 Posts
Discussion Starter #6
Okay, I kind of figured something similar but I was confused due to a Reactive Dogs group I am in. Someone mentioned counter-conditioning with marking and treating for positive interaction with the "scary thing" but an admin had mentioned that clicker work does not exist in counter-conditioning and if you are clicking/rewarding it is then operant conditioning not conter-conditioning so there's where my confusion lies lol
 

·
Registered
Joined
·
519 Posts
Discussion Starter #7
wow, i'm a total ditz and did not even see the comments above dibby lol.

okay so let me take a shot at this...

let's use Nala for example since we're working on getting her used to a dremel right now (she does not particular like me grabbing her paws for grooming, but she has come a long way!) i'll also use cheese as an example since she also loves cheese :p so since Nala already established a bad experience with nail trimming, if I were to give her cheese every time I pull out the clippers... that's counter conditioning?

i have not yet personally used the dremel on her and she has not gotten her nails dremeled since she was about 4 months old (now 12 mo.) but we just got one and i'm trying to get her used to it. what is it it called when I pull out the dremel, mark and treat? here's some more details:
to introduce the dremel anytime I have the dremel out I give her treats. if i'm doing athena's nails and nala comes over, curious and not frightened, she gets more treats. when we're working on just nala i might put the dremel on the floor and give her treats for voluntarily interacting with it. then i might pick it up and turn it on while giving her treats. then i'll turn it off and move it towards her and treats for not backing up. then maybe i'll touch it to a nail and then treat, etc etc.
what is form of conditioning is that?
 

·
Registered
Joined
·
519 Posts
Discussion Starter #8 (Edited)
okay so i just read over snackrat's and cookie's post again and i THINK i got it lmao...

so if i'm just associating a positive feeling towards the clipper/dremel so that she's no longer scared, that's counter-conditioning... but let's say i teach her to sit in my lap when either come out to make the process easier on me then that's operant conditioning?

so what about the LAT game for reactive dogs? i'm teaching athena that when dogs are around she gets food, but at the same time she's learning to look at me after seeing the trigger because she knows she'll get food if she doesn't react and doesn't fixate. i think this is where i get confused because technically i guess i'm teaching her to stay quiet/calm and look to me after she sees her trigger to get her reward for not freaking out BUT isn't this also teaching her that when dogs are around she gets food so seeing dogs is now becoming something she enjoys? is this a situation when the two overlap? sorry if i'm making this more confusing than it has to be, i just really want to understand.
 

·
Registered
Joined
·
2,418 Posts
The admin in the Reactive Dogs group is correct - there is no marking in counter conditioning. You have it right in your examples. Nala sees or hears the dremel, she gets cheese: classical counter conditioning. She doesn't need to do anything, so there is nothing to mark. If you train her to sit in your lap for nail trimming, that is operant conditioning and you can use a marker/clicker because it involves voluntary behavior.

Look at That combines CC and OC. The dog looks at the trigger (voluntary action), handler marks and reinforces that behavior. Counter conditioning is happening because, as Snackrat said, both processes can occur concurrently: you are reinforcing a voluntary action (OC), but the pairing of the trigger (e.g., other dog) with the reinforcer (e.g., cheese) is building a positive association (CC).

There is a saying that Pavlov is always on your shoulder. It means that you are always building associations. My dogs love our training center because it's associated with treats, toys, and fun. Every time we have a training session, the dog is associating food, games, and attention with me. Now, if I were training some other way, the association might not be so positive.

Not to make things more confusing, but clicker training always includes classical conditioning. Pairing the clicker (initially meaningless to the dog) with food (generally important and appealing to the dog) is building a positive association between the two.
 

·
Registered
Joined
·
4,437 Posts
Okay, I kind of figured something similar but I was confused due to a Reactive Dogs group I am in. Someone mentioned counter-conditioning with marking and treating for positive interaction with the "scary thing" but an admin had mentioned that clicker work does not exist in counter-conditioning and if you are clicking/rewarding it is then operant conditioning not conter-conditioning so there's where my confusion lies
lol

Might be a group I know well but maybe not as there are several. When doing desensitization and counter conditioning, you are not marking and rewarding a behavior (operant) you are helping the dog change the way it feels about the trigger. Dog sees trigger=food appears. Nothing to muddy the picture for the dog if you're doing it properly.

When you use a clicker to mark "positive" behavior you're no longer doing D/CC, you're rewarding a behavior and that can become confusing for the dog. Change the way the dog feels about the trigger and then you can move on to operant training to ask a dog for a trained behavior.

Counter conditioning is classical conditioning. The term counter conditioning is used with desensitization and is used to change a dog's emotional response and should always be done far enough away to not provoke the previously learned negative association.
 

·
Registered
Joined
·
140 Posts
wow, i'm a total ditz and did not even see the comments above dibby lol.

okay so let me take a shot at this...

let's use Nala for example since we're working on getting her used to a dremel right now (she does not particular like me grabbing her paws for grooming, but she has come a long way!) i'll also use cheese as an example since she also loves cheese :p so since Nala already established a bad experience with nail trimming, if I were to give her cheese every time I pull out the clippers... that's counter conditioning?
Yes. its CC.
BTW You can work separately on gently holding her paw and just touching her nails so she gets used to that. That's CC. If she starts to give you her paw so that you can touch the nail that's OC .

:to introduce the dremel anytime I have the dremel out I give her treats.
CC

if i'm doing athena's nails and nala comes over, curious and not frightened, she gets more treats.
OC


when we're working on just nala i might put the dremel on the floor and give her treats for voluntarily interacting with it.
OC

then i might pick it up and turn it on while giving her treats.
CC

then i'll turn it off and move it towards her and treats for not backing up.
Erm Not sure

then maybe i'll touch it to a nail and then treat, etc etc.
what is form of conditioning is that?
CC
 

·
Registered
Joined
·
140 Posts
Okay, I kind of figured something similar but I was confused due to a Reactive Dogs group I am in. Someone mentioned counter-conditioning with marking and treating for positive interaction with the "scary thing" but an admin had mentioned that clicker work does not exist in counter-conditioning and if you are clicking/rewarding it is then operant conditioning not conter-conditioning so there's where my confusion lies lol
It does get confusing! The Learning Theory is a scientific theory and many people think its not a perfect one. It causes a lot of disagreement between trainers and behaviourists !

I wouldn't worry too much , if what you do works for your dogs that's the main thing!

ETA I'm pleased to see you have more replies while I've been posting and very good ones they are too. I write thee posts late at night and sometime my brain goes blank so i hope that if anything I have said is incorrect . it will be pointed out .:)
 

·
Registered
Joined
·
140 Posts
In my last post, I meant BF Skinners OC theory, the four quadrants and extinction causing disagreements.

Its useful to remember that behaviour is an outward expression of emotion.

Its also said that if you change the behaviour, the emotions follow.

Emma parsons in her book Click to calm says
"one of the major benefits of clicker training to treat aggression is that it incorporates both classical and operant conditioning principles simultaneously to help us desensitise and "reprogram" our dogs. By clicking and feeding the dog in aversive environment , you not only change the dogs emotional association wityh the environment (CC) but you can pinpoint every correct decision the dog makes when he "operates" in the environment(OC.)

Sarah Whitehead of Clever Dog co , formerly of COAPE founded by John Fisher and Peter Neville , mentions anchoring the emotions using a clicker.
Its part of her Think Dog Advanced course which I haven't done so am unable to explain how that works .

So different views and ideas :confused:
 

·
Registered
Joined
·
519 Posts
Discussion Starter #14 (Edited)
@Grabby @cookieface @dibbythedog

Thank you guys for being so patient with me!

When I pull out the clippers, since Nala usually runs away, I click and treat for her even looking at the clippers. My brain process behind that is I want her I guess "teach" her that she can look at the clippers without being fearful, as apposed to hiding from the clippers the moment they come out. So I click when she shows interest in the clippers to let her know I like when she looks at/touches them and to show her that clippers = yummies.

Am I not doing that right? I just feel like the clicker is gold, they hear the click and they get excited because they did something good and of course will be rewarded for it.

ETA: NVM, I got it. LOL, so since we're in the beginning stages we're only working on the association not an actual action like jumping in my lap that's why I don't need the clicker. I'm jumping the gun here, aren't I? haha. Again, thank you so much. Life is clear again :p
 

·
Registered
Joined
·
140 Posts
Grabby, :rofl:

Mjan, Well done for working it out .
I know how you feel about clickers, I have one in every jacket pocket.

Let us know how you get on.

At the moment , i am using CC at the gate of the park and the area just inside . This is the scene of tension as most the run ins with other dogs have happened here. Because of bushes, you can't see what's coming until you step in through the gate. I'm just handing treats out and throwing them on the grass for Pip and Libby to seek .

My behaviour has been affected too and that affects my dogs. I tend to get tense and jumpy as I approach the gate . I think that's CC :D
 

·
Registered
Joined
·
316 Posts
I think you've worked it out. Don't you love it when it "clicks?!"

I've been getting my dog used to a dremel tool too. I started with lots of classical conditioning, with the idea that I wanted her to feel very happy about the tool. No behavior required, just lots and lots of treats anytime I showed her the tool (and then anytime I turned it on, and then anytime I moved it toward her, and then anytime it touched her toenails). The more she associated the tool with food, she more she started actively offering behaviors to try to get me to give her even more treats...pushing her paws into my hands, lying upside-down and waggling her toenails enticingly, trying to push her toes up against the dremel, trying to put her tongue on the dremel just in case that works (sigh). That's one way that classical conditioning and operant conditioning tend to overlap -- the more confident and cheerful the dog feels, the more likely she is to start offering voluntary behaviors. Of course, building those emotions tends to take a lot of time and repetitions, and is not always as immediately fun (for the human) as operant conditioning. It was probably three months or more before my dog was such an eager participant in nail sessions, but now that she is, I can start rewarding behaviors I prefer.

That's also why I like to make sure I do classical conditioning first, because it lays a foundation of confidence & positive feelings on which I can build confident behaviors. Of course, there are many times when that isn't necessary. If I am teaching a dog to "sit," I typically assume that the sheer fun of getting clicks and treats for something so simple are going to also make the dog feel happier about working with me, happier and more confident about "sitting," and so on. In that case, I don't usually have to worry about emotional undercurrents that may mess up my training plan. But if we're talking about counter conditioning, we're talking about situations where the dog already has a negative response to the stimulus. Those negative emotions can poison the whole training endeavor, so I think it helps to start there.
 

·
Registered
Joined
·
519 Posts
Discussion Starter #18
@SnackRat I feel relieved that it took your pup 3 months to become fully comfortable since Nala, of course being the little sensitive baby that she is lol, is taking a while to come around to the dremel. It's been two weeks so I'm glad she's not the only one who's super slow at this. It's funny because I've never actually used it on her yet and out of nowhere yesterday as I grabbed the dremel to do Athena's nails, Nala took off under the couch. BUT THEN as soon as I actually started doing Athena's nails and it was on, she came out and wanted to stick her nose all over it. She's weird :p
 
1 - 18 of 18 Posts
Top