They’ve made a text version of that handy video with the cute graphics:
If In Doubt, Add Some Space
Your
puppy doesn’t have to be right in the middle of something to have a
positive socialisation experience. If you’re ever worried that a
situation may be too much for your puppy, move further away and give
them a chance to acclimatise.
A
good example of this is socialising puppies to traffic. For many dogs,
standing right next to a busy road with all the large, noisy cars can be
very frightening. Avoid busy roads at first, starting somewhere like a
park where you can walk along away from the road. As your puppy’s
confidence improves, you can try coming closer and closer
What Should I Socialise To?
There are six main categories of things that you should socialise your puppy to:
This looks like a really good intro, but it doesn’t touch on a couple of important things, so I’ll add them.
Small amounts of very positive socialization are much better for dogs than lots of marginal experiences. It’s tempting to try to expose your puppy to every possible thing as fast as you can, but that risks setting them up for unpleasant experiences, situations you can’t control, or just straight up overloading them. Pick a couple things and do them right. This is especially crucial for dog-dog socialization when you’ve got a breed that is genetically prone to reactivity or aggression because all of their experiences absolutely have to be positive ones. @molosseraptor has some great posts on picking the right dogs to socialize your pup with.
Fear periods change socialization rules. Dogs have two fear periods – one predictably at 8-10 weeks and one later in adolescence, somewhere after about six months of age. These periods are when dogs would be gaining independence and it becomes super important for survival to learn what is dangerous and what isn’t. During these periods dogs are prone to single-event learning, which means you have to be really careful not to set your dogs up for bad experiences because they might shape their behavior for the rest of their lives. Fear periods are a good time to back off on active socialization attempts and work on keeping your puppy happy and comfortably engaged with things it is already used to. (Here’s a link to a great article on fear periods).
The goal of socializing an under socialized and fearful dog should be to help them feel as comfortable as possible in their environment so they can live the most fulfilling life possible. First, use management to help your dog feel comfortable. This might include getting up extra early to go for walks when fewer people are around, or covering the windows so your dog can’t get spooked or aroused by people passing by outside.
Once you’ve used as much management as is practical, assess your dog’s life and your goals for them, and work slowly to reach those goals. Many people want to turn their fearful dogs into dog park dogs, social butterflies, but that’s really impractical for many dogs even if they’re well-socialized. Practical goals for most dogs are learning to walk past other dogs on the sidewalk without panicking, being calm around cars, and ignoring children running past on the street.
The best way to start is with a training class, especially one for reactive dogs. It will give you a controlled environment to work with your dog around the things they find scary. Feedback from an instructor can be invaluable early on in reactivity training and socialization, and it can help prevent you from making mistakes early on and setting your dog back farther. Remember that socialization is about allowing your dog to have good experiences, not necessarily having them interact directly with people and other dogs. Socialization isn’t necessarily “social”.
If you’re going to train solo for whatever reason, I recommend the exercises in Control Unleashed by Leslie McDevitt. Her books can be difficult to understand, so I recommend picking up one of the DVDs based on her books. You can even go on youtube and look for demos of the exercises. The Look at That game (LAT) is especially useful for counter-conditioning. Most of the other exercises are used for management and minimizing risk in your daily life. They may not be absolutely necessary depending on your dog’s behaviour and your current management strategy. You can also look through my #LAT tag for examples and instructions on how I teach LAT.
Please let me know if you have any specific questions once you’ve looked through that. Good luck!
As a contrast to the previous gifset, I wanted to make one with the classic video by Dr. Sophia Yin showing counter conditioning in action. This is a dog that had been displaying aggression severely enough to be up for euthanasia. The stimulus prompting aggression in this video is having his face blown on. While we don’t hear anything about the dog’s history, it’s pretty easy to assume that this is fear-related, as shoving your face at a dog’s face is pretty aggressive body language, a lot of smaller dogs have fear-related aggression due to their boundaries being ignored, and I don’t see any resource-guarding behavior.
You can’t draw a complete parallel, but there are a lot of similarities between this video of an aggressive dog and the video of the aggressive horse. This dog seems to be making a big aggressive display and then retreating, instead of continuing the attack with the intent of causing serious injury. The horse had its movement restricted to the round pen, and this dog has its movement restricted by a leash. Both are unhappy and dangerous animals.
Dr. Yin resolves the aggression by pairing the provocative stimulus (blowing on the dog’s face) with food. After only a few brief sessions and a bit of time, the dog no longer exhibits aggression when prompted. He doesn’t enjoy the stimulus (he still moves his head back and away, and there’s a bit of lip licking) but having his face blown on no longer provokes aggression. Instead you can see eagerness for the treatment and what looks like enjoyment of the exercise (tail wagging, what looks almost like a play bow or an attempt to get a reward with a behavior he was taught, ears forward, open relaxed mouth, looking up at her face). His emotional reaction and outward behavioral response are dramatically different.
I don’t present this as an example of why counter conditioning with food is a preferential miracle cure (dogs are a lot more likely to exhibit aggressive body language, so the horse probably had way more of a backlog of fear, whereas this guy’s fear could be worked around relatively quickly. I also wouldn’t ever recommend anyone tackle aggressive body language straight up with a leash restraining the dog, and definitely not by blowing into the dog’s face, where it’s so easy to get bit) BUT this shows a similar scenario, similar aggression, and a different protocol for resolving the problem that doesn’t involve the use of an aversive stimulus to work around aggression.
I remember watching this video in around 2010 and being amazed that this “counter-conditioning” was such a powerful technique. It was one of the videos that made me up my training game big time, and learning about CC was a massive help in socialising and rehabilitating Breeze.
I did think she was totally mad and was about to lose her nose! Definitely not a dog or situation for novice trainers, but a really useful video about hugely helpful technique.
I’m definitely into this, but I do have a question- why does this work, as opposed to making him display his aggressive behavior more? It seems like this could also been seen as Dog displays aggression->give treat-> dog acts aggressive to receive treat. What would the difference be between this counter conditioning and training a dog to be more aggressive?
The difference there is the difference between operant and classical conditioning. Operant conditioning is training as we usually think of, which is controlling conscious behavior. Classical conditioning addresses involuntary reactions, like Pavlov’s drooling dogs.
In this situation, Dr. Yin isn’t trying to change the dog’s behavior, she’s trying to change the dog’s emotions. Once the emotional state has changed, the behavior goes away by itself.
I also get that this works, but I don’t understand your explanation of why you don’t just accidentally capture the aggressive behaviour? The dog doesn’t know your counter conditioning not operant conditioning.
This is something I struggle with teaching mango not to bark at the window, I definitely taught her to bark once and look at the human when trying to counter condition. We are doing better now by treating pre-bark and giving an alternative behaviour if she’s already barking, but it would be interesting to know more if you have the time?
Maybe another way to put it:
“Dog displays aggression->give treat-> dog acts aggressive to receive treat.”
So why doesn’t this teach the dog to be aggressive?
Because the dog is not thinking operant-like about the aggressive behavior. He’s not thinking “oohh I behaved aggressively and I get a reward!!” He’s feeling scared of something, but that something is always followed by a reward, so he learns it’s not scary. The aggressive response associated with his fear goes away too.
It’s the same reason why you can give your dog a reward at every thunderclap, and it doesn’t teach them to be scared of thunder- the fear response wasn’t operant. You’re not rewarding their fear, you’re associating the stimulus, thunder, with good things.
Dogs don’t think “oh, I think I’ll experience some fear now” it’s not something they (or you for that matter) can /will/ themselves to feel.
Plus fear is so aversive you wouldn’t even if you could. Counter-conditioning does “risk” accidentally creating fake on-purpose behaviors. A la the story of the barking dog from Reaching the Animal Mind who was given treats for barking, but when he tried to bark on purpose, it was more of a uncertain high pitched yelp. That said, those are easy to stop because the behavior is no longer fueled by the underlying emotion.
We humans have a profession that’s all about rewarding people for pretending to have all kinds of emotions: actors. And they still need to practice at it for years. Think about that- you can have a fight with your mom or a heated thanksgiving dinner debate and it’s genuine emotion- but a team of actors staging the same thing have to really work at making it seem real. And someone who’s portraying a criminal doesn’t then become one after the director says “cut!”…. Fake on-purpose behaviors are just not the same, for the same reason- you can’t will yourself to feel emotions on cue. Maybe method actors would disagree lol but that’s the gist of it.
OK, that makes sense, trying to train a dog when they’re over threshold gets you nowhere, and that’s what you would be doing if you were trying to capture aggressive behaviour . And we got rid of ‘bark look at human’ behaviour but just ignoring it, which was a lot easier than getting rid of barking (still working on that, but we make progress).
Thanks
The above explanation works for the layperson, but it doesn’t have a lot of scientific basis. The reality is that we’re only just coming to understand reactions and ‘feelings’ in animals, particularly because of the efforts of some incredible trainers and dogs who are letting us get a glimpse into the conscious mind of a dog via MRI machines. It’s what we THINK might be happening, but realistically all we have to go on is hormone interactions and other foundations in behavioural science.
This is, of course, one of the main challenges of training dogs and interacting with dog trainers. Most people owning a dog do not have any sort of training in applied ethology. Most dog trainers have no idea what applied ethology is. But it’s important, and it helps us to understand consistencies as well as inconsistencies.
To make an explanation with a bit more evidence behind it, we’re performing a chemical override using differing hormones that are released as a response to eating, over the stress hormone cortisol that is heightened when experiencing ‘fear-inducing’ stimuli. There’s also usually a motivation present for the dog to eat, so the dog may have been deprived a meal in the morning (not unusual, many trainers withhold regular meals and choose instead to feed them as rewards, but by doing so you create a motivation in the animal to perform consummatory behaviours, like eating. It’s a perk that motivation increases for the reward.) and is driven to perform a consummatory behaviour to satisfy that motivation.
It has also been shown that if you create a motivation in an animal, and don’t allow them to perform consummatory behaviours to decrease their state of motivation, the existing high state of motivation can actually cause stress. Performing consummatory behaviours to decrease motivation lowers stress. So realistically, there are many small factors coming into play here, many of them related to hormones within the body, that are contributing to associating certain stimuli with good things.
This is a multifaceted issue, but from a physiological and behavioural standpoint, these are just some of the many factors contributing to altering an animal’s perception.
You can indeed capture the wrong behaviour (this is because there are multiple factors working in tandem, you can choose to appeal to classical or operant conditioning while training, but it doesn’t mean you will completely turn the other off, we’ve learned dogs are making inferences in ways many young children are, they are constantly absorbing information.) which is why narrowing the criteria is important. You cease to reward for just anything other than, for instance, biting your face off. You shape a new behaviour as an alternative. Many trainers will just reward ‘not barking’, but you’ll also notice that people will instead reward not barking to start and then will start to reward when the dog lays down, when the dog goes to their crate or lays in their bed. They shape ‘not barking’ into a new behaviour, increasing the criteria from ‘don’t bark’ to ‘don’t bark and turn away from the window’ to ‘don’t bark and turn away from the window then walk to your bed’ and finally ‘don’t bark and turn away from the window and lay down in your bed’, versus leaving the dog to their own devices where they might pick a new behaviour that isn’t ideal. Especially if we’re talking motivation that was previously satisfied by the undesired behaviour, and they are now having to perform displacement behaviours to try to satisfy the motivational state.
You’ll see the idea of teaching an alternative behaviour or task a lot, and you can probably think of a couple you’ve taught your own dog even accidentally. My dog had an issue jumping up, so I realized I was teaching him a replacement behaviour as I continuously asked him to get toys for the visitor instead, as when he retrieves a toy and comes to offer it he keeps all four feet on the ground. I often remind him, but without intensive training or focus on it, he has started to go looking for a toy after being told not to jump up. He is making the connection himself, even without training, that if he has all of this pent up excitement…to use it productively.
For laypeople who follow this blog – I highly do not suggest trying to recondition aggressive behavior by yourself. Yin was an amazing and skilled professional (rest in peace) and you’re way better off getting someone with training to help you than trying to work with aggression yourself. It’s not worth the risk.
Fucking THANK you for this post!! Ive been waiting for the “dominant alpha” theory to die out. It gets me so heated i swear!!!
It’s so ridiculous that people insist on applying an incorrect theory about wolves to dogs, and then try to apply it to humans too
Seriously that last bit, tho. Even if you want to apply animal dominance hierarchy crap to humans, we’re primates. PRIMATES.
Yes this. This is why the concept of dominance in dogs is basically a pile of crap.
Also I haven’t got time to go into it but people’s understanding of ‘dominance’ as it relates to animals is pretty flawed. It’s actually quite variable so it’s things like
A is dominant over B for food but C is dominant over A B is dominant over C and A for the spot by the window A is dominant over B and C for the squeaky bone
Whenever I hear a zookeeper talk about dominance or god forbid use the word alpha sincerely I cringe
Lol people still believe in alpha theory? How retro.
Everyone believes in alpha theory because a) Cesar and b) science doesn’t trickle down and I hate it.
I have to be a pedant here but this infographic is getting training terms wrong again. Striking a dog or putting their nose in poop is positive punishment, because you’re adding something to try to decrease the behavior happening again. People get positive punishment confused with negative reinforcement all the time (seriously it’s one of the hardest things for trainers to define sometimes), but they’re not the same. (Negative reinforcement involves removing something to make a behavior happen more frequently – it’s often called pressure and release training, and can be used with positive punishment but is not always).
What @starrypawz is describing in their reblog is called situational social dominance, which basically says ‘different dogs will stand up for different things in different scenarios, depending on how much they value the thing and what the dynamics of the group regarding the thing are.’
For me, the ideal dog and handler relationship dynamic is a partnership. You, the trainer, are the senior partner, and your job is to train the rookie, have his back, and teach him everything you know, so one day he can have your back too.
Dogs live in the now. If you say, “Hey Brutus, come here!” Then yell at him for the pee spot you found on the floor, he is going to think, “Yikes, I shouldn’t come when I am called I guess.” They cannot associate punishment with things that happened in the past, only things in the moment. Even if you’re pretty sure he gets it because he looks super guilty, he is just responding to the tone of your voice.
This though
That’s why you stick their noses in the pee spot for a whiff too so they know their pee in that spot is making you upset. They will know it’s their pee right away because of their super sniffers, don’t worry. And it’s not as cruel as it sounds–dogs react to the nose before ears or eyes–so it’s like pointing to a juice spill because little Timmy was a brat. The dogs will understand because it’s all happening in the now.
Not sure if this is serious or not, but for those who feel this way please understand that a dog is.. Simple minded. So while you want to deliver the message that peeing in the house is not okay, they are learning that you do not approve of their pee in general. This will create a nervous dog who will hide their eliminations from you in fear, and may even hold their bladder.
The only effective way to provide bilateral feedback is to catch them in the act, then praise them outside. Scolding them hours later by shoving their face in their pee will leave your reasoning up to their interpretation. Do you not like when he smells his pee? Because you’re making him smell it, but then you’re scolding him. Do you not like when he poops? Because you are scolding him for pooping, so next time he will try not to poop.
They cannot differentiate when it is hours later. So while they do understand your anger towards their elimination, they are left to just assume that eliminations make you angry in general. Don’t be a bully.
Jesus I hate that people still do this. It’s very well demonstrated this does not work. It’s on you to set your dog up to succeed, not fail, and sometimes you just have to deal with mistakes.
I want to add that it’s not that dogs are simpleminded per se – they’re very capable of understanding schedules and noticing people’s selective attention for instance – but we have no way of communicating to them a frame of reference in time to which our punishment applies.
We don’t even know how dogs interpret time. We know they can remember previous things that happen but not if they have concepts for concrete blocks like ‘yesterday’ or ‘an hour ago’, so there’s no way of letting them know you’re referencing a previous action or what one it is. They mostly learn in the moment because that’s the only conceptual timeline we can communicate with them in.
Thank you to everyone who provided informative input on this because I wouldn’t have even known where to start with addressing it.
In case you haven’t heard, CM recently released an episode of Cesar911 in which he knowingly let a Boston Terrier who had a habit of killing pigs loose with three animals and then goaded him into attacking them. That’s animal abuse, straight up. It’s getting massive attention and CM might actually get canceled. The original video is set to private now, but here’s a reproduction and a breakdown. This video contains animal abuse and injury. Watch at your own discretion.
This is the petition to get Nat Geo to take CM off the air and cancel his show. And it’s getting massive attention. As I’m posting it, it’s at 8.7k signatures and it got 2.5k just in the first 24 hours it was online.
Sign it and share it – share the actual petition on facebook. We could actually get the abusive asshole canceled. I’ve been waiting to post about it because I didn’t want to spread false hope in case it flamed out, but this is for real. Everyone is finally calling CM out on his abusive, incorrect ways.
Certainly the best learning outcome would be for National Geographic to take a stand for dogs, pigs, and other animals and remove Cesar Millan from the air until he reforms his act.”
Cesar Millan and anyone using his “training” methods perpetuate outdated, disproven ideas about dogs and their behavior and make things that much more difficult for people trying to improve learning experiences for dogs and make training more effective and ethical.
Dog training should be about understanding how dogs actually learn and how they process information, and improving the relationship between humans and their companions. Not about intimidating and confusing dogs into compliance for fear of punishment. There’s nothing cooperative about that. Why would you want that over a healthy relationship with your pet?
I have been asked to do a behavior assessment and to share some much-needed education with an owner who feels entirely overwhelmed by her poor pup. Her son purchased Bella from a “wolfdog” breeder in Aloha, Oregon, who told him she was “70% wolf”; the breeder explained to the young man that he needed to dominate the dog by rolling her onto her back and biting her ear or nose if she did anything wrong, “because that’s how wolves do it”.
In truth, Bella does not appear to have any wolf in her, let alone the 70% her breeder claimed she was. She attacks “without warning” because she has been heavily conditioned through force to avoid warning behaviors such as growling or baring her teeth.
Following my assessment, the plan is to find Bella a suitable behaviorist who can help undo the damage her first owner has done. If she recovers, a proper forever home will be found for her and she will start a new life with folks who will not misrepresent or mistreat her.
Recently on Facebook a picture came up in my feed regarding wolfdogs. Looking through the comments, it seemed everyone and their grandfather owned a ‘wolfdog’. Some I took the time to look through their pictures of. Didn’t see any wolfdogs, but I did see photos of people doing things like having 2 grown men hold down a muzzled husky/malamute/other(?) mix because, “Nail trimming is hard when you deal with wolves.” Also read plenty of comments about how you need to “dominate” your ‘wolfdog’ and “it’s super important to make sure they know you’re the boss.” There were people excusing their poorly trained dog’s behaviors on “well it’s because they’re 96% wolf” and justifying their brutal treatment of their mixed breed dogs because “that’s how wolves are treated in the wild by their alphas.” There was enough misrepresentation and misinformation about canine behavior and training to make you want to turn off your computer for a week.
THIS, ladies and gentlemen, is why people need to knock it off with continuing to spread this kind of BS.
PiraticOctopus 