I’ve decided to split my remaining Puzzle material into a few more posts – they are just getting too long otherwise. Today’s post only has one video, but it is one I truly love, and could watch over and over again. It makes me laugh because it’s cute and goofy – but if I change my point of view, I can also see it as a metaphor.
Social skills: successful communication
An adult free-ranging dog tries to charm Puzzle, and get her to play. Puzzle isn’t afraid – she could walk away or hide behind me. She doesn’t feel the need to hide or flee, but she clearly says, “No!” by turning her head away and NOT engaging. The other dog works hard, but isn’t intruding in her space. He is being gentle, and self-handicapping by making himself small and rolling on his back. He doesn’t get frustrated or impatient – he just works very hard, and keeps respecting Puzzle’s boundaries.
The reason we know this is good communication – even though it doesn’t go anywhere – is the fact that Puzzle is able to stay put. She was sitting on this step before the other dog got here, and she stays in the place she picked for herself throughout the conversation. She doesn’t feel threatened. Yet she clearly knows that he is communicating with her, and she responds politely and clearly: “No.”
These are great communication skills on both parts. Watch this – maybe more than just once. The next time you need to either set a boundary for yourself (see Puzzle) or respect someone else’s boundaries (see the adult dog), remember this video!
Two lessons for humans
Set your own boundaries kindly. You don’t need to yell, and you don’t need to hide from or stonewall the other person.
Accept the boundaries of others gracefully. No need to get frustrated or annoyed. Just do your best. Dogs don’t generally take things personally – for example, this adult dog won’t be unable to sleep tonight because he’ll obsess over what he should have done differently. He will get up (after the video ends), wag, and move on with his life. Don’t take things personally. Be more like this dog.
The (not so) blank slate: what the puppy brings to the table
The laws of learning apply to all puppies equally. Also, every puppy is different. Both of these things are true: sadly, things are rarely as black and white as we control-loving dog trainers would like them to be.
I was pretty certain my neighbors hadn’t done strategic socialization before the puppies left their nest. However, they likely grew up in a family environment, around young children, cats, ducks, and their dogs (apart from the dam, they have a small male that looks like a Miniature Schnauzer/Chihuahua mix). That’s a good foundation!
As soon as they were ready to explore, on their own time, they started venturing out into the alley with their mother, a little bit braver and further every day. This is one aspect of growing up free-roaming I love: it’s up to the puppies when they are ready to leave the nest, and how far they are willing to go. Their humans just let them be.
Out in the alley, they would meet passers-by and the occasional dog or neighborhood cat. They were also always able to retreat behind the safety of their gate, and had a mom who’d defend them fiercely against passing strange dogs (but not against known neighborhood dogs) until they were between 6 and 7 weeks old, when she intervened to a lesser and lesser degree.
There were five puppies, and this is how I see their baseline temperaments on a scale. Note that my scale only goes from the shiest to the most curious puppy in that litter. It is not a scale of all puppies, or of puppies in general.
The parents’ temperament and stress levels
We also know a little bit about the parents’ temperaments: the mother is neutral/friendly towards all people outside the home. She’ll bark briefly when someone enters her yard. She is neutral/friendly towards known dogs, and slightly suspicious of unknown ones. The father (assuming he is who I believe he is) is confident and mellow around all dogs and all people.
Genetically, this is a nice combination for a free-roamer or a pet dog: mellow and neutral, leaning towards confidence from the father’s side; no exuberance or red flag behaviors in the parents.
I don’t think either one of the parents has a particularly stressful life. They have lots of freedom, plenty of food, and a routine that rarely changes. This should result in a good in-utero experience for the litter. (Mothers who are stressed during the gestation period are more likely to produce pups who are prone to depression, anxiety, and social deficits. This is known to be true for rodents1,2,3 and assumed to also be relevant for other mammalian species such as humans and dogs.)
Two sets of experiences for Puzzle
I wanted Puzzle to have two sets of experiences: one set would prepare her for a potential pet dog life, and the other one would allow her to thrive as a free-roamer and scavenger. The second set was taken care of by the environment she lived in and the freedom she had. I focused on the first set. I wanted her to experience living inside a house, being left alone, being crated, mat work, walking on a leash, being in busy places with lots of people, being in stores, being handled and carried, being dog-neutral and dog-confident as well as people-neutral and people-confident, starting housetraining, getting used to traffic noises and other city sounds, being inside moving vehicles.
Not all of these experiences fall under the category of socialization – some of them are more general pet puppy skills. I also did not get through all of them while I had access to Puzzle. However, I think we did pretty well, given the fact that we only had a few weeks together. The aspects I’m going to focus on in my next two posts are socialization to dogs, and socialization to busy urban spaces/feeling neutral and confident around strange people.
(1) Weinstock, Marta (2016). Prenatal stressors in rodents: Effects on behavior. Neurobiology of Stress, S2352289516300133–. doi:10.1016/j.ynstr.2016.08.004
(2) Cabrera, R.J.; Rodríguez-Echandía, E.L.; Jatuff, A.S.G.; Fóscolo, M. (1999). Effects of prenatal exposure to a mild chronic variable stress on body weight, preweaning mortality and rat behavior. Brazilian Journal of Medical and Biological Research, 32(10), 1229–1237. doi:10.1590/s0100-879×1999001000009
(3) Soares-Cunha, Carina; Coimbra, Bárbara; Borges, Sónia; Domingues, Ana Verónica; Silva, Deolinda; Sousa, Nuno; Rodrigues, Ana João (2018). Mild Prenatal Stress Causes Emotional and Brain Structural Modifications in Rats of Both Sexes. Frontiers in Behavioral Neuroscience, 12, 129–. doi:10.3389/fnbeh.2018.00129
We stay at the very first bowl, and then end the session. Puzzle lets me know she wasn’t ready to approach the live vacuum any further, and I listen. CU is all about communication!
In her second session with the live vacuum, Puzzle is being very brave, and takes me all the way to bowl #4. At that point, she does not make eye contact again. I listen to her, increase the distance, and end the session.
Followed by another short session:
We make it up to bowl #4/5 again:
… and again:
In the next session, you’ll see Puzzle reach the fifth and last bowl for the first time! Stay tuned!
For more dog training tips and videos, join Chrissi’s February class at FDSA: Calling All Dogs!
Now that Puzzle has shown me that she can predict where the next treat will show up in the Superbowls game, it’s time to add the trigger into the pattern. In Puzzle’s case, that trigger is the vacuum. She thinks it’s quite creepy!
When working with fear or anxiety, raising criteria slowly (rather than starting with the trigger at full intensity) is always a good idea. In the case of the vacuum, I’ll start with a dead – i.e. silent – one before asking Puzzle if she wants to approach a roaring, growling live vacuum.
Doing SO well! Puzzle is ready to face the live vacuum in her next session! (And if she isn’t, that’s okay, too: she’ll be able to ask me to stop approaching at any time.)
For more dog training tips and videos, join Chrissi’s February class at FDSA: Calling All Dogs!
In Hadley’s book, quite a number of things are alarming. One of them: new objects in familiar spaces, like the neighbor’s trash bag that hasn’t been sitting out the day before, or a penguin wearing a hat, standing provocatively at a doorway where no one used to stand. (I totally get that. Penguins are not supposed to wear hats; now that’s just weird!)
My favorite way to deal with scary stuff is to make it part of a game. I’ve done this with Phoebe back in the day when she had a random-objects-are-scary phase in her adolescence, and now I’m using the same strategy for Hadley. By means of shaping, I want to give Hadley the experience that he controls the situation, and can turn scary stuff into cookie vending machines by means of choosing to engage with it.
Engaging with scary objects in return for a cookie is entirely his choice, not mine. I’m not luring him closer, and I’m not forcing him to engage with the scary object in any other way. Hadley decided whether he goes all the way up to an object, touches it, or just plays a little LAT from a distance. If he chooses to disengage after a little while, that’s okay, too.
Now that I’ve finally decluttered my camera phone, I got to film today’s encounter with a penguin wearing a hat. We met that weird bird on our way home from a walk in the neighborhood. We frequently walk past this house, and never before has there been a penguin standing in front of it. Obviously, Hadley was concerned. It looked quite devious in its green hat, pretending to be all innocent, just standing there provocatively. It might just have been planning to murder us all, and Hadley was right to point this out to me.
This is what our penguin session looked like:
Note that rather than using strategic points of reinforcement, I’m feeding away from the penguin, so the increase of distance acts as an additional reinforcer (R-). The whole thing took about 5 minutes, including a few breaks whenever either Hadley chose to disengage and do sth. else for a little bit, or when I went to reinforce Phoebe who I had put in a sit-stay. When Hadley offered looking at the penguin or approaching it again after a break, we were back in the game. At 0.30 in the video, you can see from Hadley’s body language that he’s getting too close. I should have clicked sooner, i.e. after fewer steps towards the penguin. He trusts me enough to keep playing, so for the next click, I lower criteria to just a few steps, something he can easily do. Then I gradually increase criteria again. At the end, you see his first bold touch. He’s not worried anymore and recognizes the penguin as the latest cookie-vending machine that has been placed here for his convenience! Engage with it, get a cookie from mum. Sweet!
A few days ago, we met Tini and Nayeli for a walk. Hadley recognized Nayeli after briefly alarm-barking at her from the car, and immediately started playing chase with Phoebe and her! Wow – this is the first time he has played as intensely with a dog who isn’t a family member. Nayeli is simply a great role model, and a wonderful auntie to have as a puppy. I’m sure Hadley will have fun with Tini and her when he vacations with them in January.
We encountered two strange off-leash dogs on our walk. The first one was a tiny, shy puppy. Phoebe, Fanta and Nayeli didn’t care about the tiny dog, but Hadley approached him with a friendly wagging tail! WOW! Best. strange. dog. encounter. ever! I was soooo happy; proud of my puppy and of my training success, and happy that my dogs get to have dogs like Nayeli in their lives.
The three musketeers are having fun near Lusthaus.
Today, we had another very successful outing: we went for a walk today – just Phoebe, Fanta, Hadley and I. Off leash, on the fields.
After a few minutes, two women with a dog slightly bigger than Hadley, also off-leash, crossed our way. We saw them coming from a distance. Phoebe and Fanta walked over to say hallo, and Hadley … looked, wagged, and went back to playing chase with Phoebe! He had only hesitated a moment, than decided that the strange dog wasn’t a threat. He didn’t keep close to me, and didn’t mind walking or running close to the strange dog. The women and I walked together for about fifteen minutes.
Phoebe, Hadley and the first dog we encountered on today’s walk.
A little later, the next challenge: a with an on-leash Spitz about Phoebe’s size came straight at us. I took my dogs on leash, and made way for the Spitz to pass, started feeding treats when Hadley noticed the strange dog and went on feeding until he had passed us. Hadley watched the dog attentively and calmly ate his treats, then quickly switched to offering sits – the strange dog wasn’t important enough to pay attention to! Hah! I am SO happy with how he is developing!
Phoebe, Fanta, Hadley and the two women’s small dog were let off leash again. Another few minutes went by, and we met the next dog: an old, off-leash Maltese who was standing quietly near his even older owner. The Maltese told our group in body language that he was neither a threat not interested in interacting with any of them, and they all curved around him. Hadley followed suit! While curving, he had his tail slightly between his legs and glanced sideways at the Maltese, but followed the other three without hesitation. Woohooo! Witnessed how to deal with dogs like this, and did it himself! Wonderful puppy, and I’m happy my training has helped him become more confident around strange dogs!
Fanta, Hadley and, in the distance, the dog who walked with us for a while. Everyone’s happy doing their own thing. There’s plenty of space for everyone, and no need to feel threatened.
We parted ways with the two ladies and their dog. I played with my camera while Phoebe and Hadley played near the water and Fanta had one of his rare it’s-my-favorite-season runs.
On our way back, we met an off-leash Border Collie; an adult black-and-white female. Phoebe mistook her for Xandro and was quite startled when she realized that Xandro isn’t the only beautiful Border around. I didn’t interfere with Hadley’s behavior because it had been going very well so far. Hadley looked and I could see that this dog was more concerning to him than the others had been. She was more active, and held her busy tail up high. And then she even looked at him directly! Hadley made one tiny bark. I kept walking and called him, he came. She came over, he let her sniff him submissively, and then happily greeted her human. We exchanged a few words while Hadley watched Phoebe and the Border discuss who was going to keep the stick they had found.
We walked on, and passed the old man with the Maltese again. They were still standing at the same spot, chatting with an acquaintance. This time, Hadley curved around the Maltese without hesitation and without putting his tail between his legs. Yeah!
Almost back at the car, we met a woman with a big, on-leash dog resembling an Akita, but slightly smaller. They were walking straight at us. I put my dogs on their leashes, and noticed that the woman deliberately lead her dog on the side of her body that wasn’t facing us and was feeding treats while approaching us. It always makes me smile to see other dog people working with their dogs in similar ways as I do! Also, I’m always happy to encounter polite dog owners who are as keen to avoid on-leash encounters as I am.
We walked a little to the side and let the two of them pass. Hadley requested that I play LAT with him! He looked at the Akita, then back at me. At the Akita again, then back at me! Hah! This is awesome! Thank you, Leslie McDevitt, for coming up with this simple, yet brilliant game. Of course, since he asked me to, I played with Hadley, and he got to earn a few treats for looking, and then for the sits he offered. And on we went, off leash again, back to the car.
I have to say, I am relieved and really, really glad Hadley’s attitude towards strange dogs is slowly relaxing. I am also glad that the strategy I chose for dealing with his issues is turning out to be the right one for him!
Wow – time really does fly. So much has happened since the last time I found a moment to sit down and write a blog post. Where do I begin?
The little rascal has been a pretty easy puppy to take care of. He’s been spending lots of time with me when Tom is at work, and I couldn’t help comparing him to Phoebe. In most regards, Hadley has been less of a challenge than Phoebe when she was his age. Phoebe was an extremely high-energy puppy, and she was very mouthy. Hadley has mostly been relaxed and friendly.
There is, however, one thing that concerns me: Hadley is a rather wary puppy, particularly when it comes to strange dogs. From day 1 onwards, he has been alarmed by strange dogs, even the ones that were 1.5 blocks away. I am worried about this because Hadley spent his puppyhood among all kinds of different dogs – his breeder has more than 10 Border Collies, Norwegian Lundehunds, and a Beauceron. To my knowledge, Hadley has only had good experiences with her dogs. In theory, these positive early socialization experiences should have turned him into a dog who approaches new dogs with curiosity and confidence. However, this is not the kind of puppy he turned out to be: initially, he would avoid other dogs whenever possible, froze/stared and eventually barked when avoidance was not possible, and tried to hide/flee if they came too close. He also took a comparatively long time (read: several meetings over the course of several days) to warm up to new dogs. However, once he considered a dog a friend, he’d play with her like any other happy puppy.
After consulting with friends and colleagues and debating how best to handle a dog-sensitive Border puppy, I came up with the following plan, which I’ve been working on since Hadley has moved in:
Part A – socialization
introduce Hadley to my friends’ friendly adult dogs in various short sessions. Always put up a portable crate and/or familiar blanket for him to retreat to, and make sure the other dogs respect his safety zone. Let him watch and decide for himself whether and when he is ready to initiate interaction. Never force contact. Never overwhelm or flood him.
My idea was that I would provide Hadley with a number of distinctly positive experiences that lead to dog-dog friendships, rather than create lots of neutral dog-dog experiences. I hoped that the more dogs he got to know and make friends with, the easier it would be for him to be around new dogs in the future, and that eventually, he would start considering strange dogs to be interesting rather than scary.
Part B – management and alternative behavior
I would also work on Leslie McDevitt’s Look at That game (LAT). That is to say, I would teach Hadley to earn clicks and treats by means of looking at strange dogs from a distance: I wanted him to start seeing strange dogs as cookie-vending machines rather than potential threats. “Dad, mum, there’s a dog, did you see it? Look, it’s over there! Where’s my cookie?”
LAT makes use of both classical and operant conditioning. One the one hand, a potentially scary stimulus is repeatedly paired with a strong reinforcer (tasty treats), which changes the emotional response to the stimulus. On the other hand, the dog is being empowered as he learns that he can use dogs he spots on the street to make a treat happen. All he has to do is point them out to his humans with a movement of his head.
If strange dogs were too close, I would retreat by means of putting a barrier between ourselves and the trigger, changing sides or doing a U turn.
Part A has been going well. Apart from my own two dogs, I’ve strategically introduced Hadley to 12 dogs by now; some male, some female, some neutered, some intact, some small, some large:
1 Border Collie
4 Miniature Pinschers
1 Irish Setter
1 Golden Retriever
1 American Staffordshire Terrier X
1 Akita mix
1 Sheltie X GSD
1 small Terrier X
He has met all of them several times in safe, short sessions, and made friends with all of them. The first few outings, he would just sit in his safe space and observe from a distance until we went home again. I did not try to convince him to come out, but focused on making sure he felt safe. Apart from that, I did not distract him with food, but let him choose what to do – stay in his safe spot and observe, walk away and do his own thing, or initiate contact with the new dogs. Helene, a friend who shares her life with 7 wonderful dogs, has been a huge help with this. (Thank you, Helene, Xandro, Wasti, Arkani, Schoko, Hexi and Guinness!)
Helene lives just around the corner. So we would meet up at a meadow close by. I would get there first and set up Hadley’s safety zone: a pop-up crate and a blanket in front of it. He could choose to hide in the crate, sit on the familiar blanket, or come all the way out on the meadow. I took one of my own dogs with me so Hadley could see that they were not afraid of the new dog we introduced him to. If he wanted, Hadley could take the crate’s side exit and go explore the forest and shrubbery rather than engaging with the other dogs, who did their own thing out in the field.
The first two times, Helene brought Border Collie Xandro and Miniature Pinscher Wasti, and I brought Fanta and Hadley. Helene and I spent twenty minutes sitting on the blanket and chatting. Hadley stayed in the crate or on the blanket with us, but did not approach either of her dogs. This was okay. It was his choice. After twenty minutes, we left and Hadley went back to sleep at home to sleep off his adventures and maybe do some latent learning.
The third time, Hadley approached Wasti with a cautious wag … and started following him around at a distance. Whenever Wasti turned around, Hadley would hurry back to his safety zone, but soon after, his curiosity took over and he followed Wasti again. He did, however, still keep his distance from Xandro.
The fourth time, Hadley was happy to see Wasti and followed him around more, even if it meant moving further away from his safety zone. His overall confidence had clearly grown, and he even sniffed Xandro’s tail a few times – of course, when Xandro turned to face him, he would retreat like he used to do with Wasti. But from behind, the Border Collie had stopped looking all that scary.
We did numerous sessions like that. Once Hadley had grown comfortable with one dog, we’d introduce another one. The last time Helene and I met, we didn’t need a blanket or crate anymore, and were able to take all 9 dogs for a walk together. Hadley had fun from beginning to end. He mostly played with Phoebe, but did not mind running ahead with her, getting close to Helene’s dogs, and quickly bounced back the two times he didn’t respect Schoko’s personal space and got a reprimand by his new auntie. I’d call this a BIG success – thank you very much for your help, Helene, and a big thank you to your patient, friendly dogs who have already been a big help in raising Hadley!
Hadley is having a good time during a 9-dog outing with Phoebe, Fanta, Xandro, Guinness, Wasti, Arkani, Hexi and Schoko.
Another dog who has been immensely helpful is Olivia, the dog who’s mum runs our local pharmacy. Olivia is a friendly and very patient Dalmatian. We’ve been visiting her several times in the course of the last weeks. At first, we kept Olivia in a back room behind a baby gate, while Hadley could look at her from the far end of the adjoining room. He could choose to walk closer or leave, to just observe Olivia who slowly wagged her tail and looked sideways, or to engage with the pharmacy personnel who were happy to greet him and let him lick their faces. (Meeting people is something that has always made Hadley happy.) The second time we went, Hadley chose to approach the gate and cautiously greet Olivia and lick her lips. The third time, he was able to meet her without a gate, and was happy to dance around her and explore her space. Olivia, the patient girl, gave him all the freedom in the world and happily took my thank-you treats.
Hadley and Olivia – first time without being separated by a baby gate.
Phoebe’s best girlfriend, the Golden Nayeli, has had a very easy time when it came to making friends with Hadley. She and her mum visited us at home and spent an afternoon with us. In his own home, where he feels most confident, and able to watch Phoebe and Nayeli play, Hadley quickly decided that he wanted to join in the fun – and that’s what he did. Thank you, Tini, for helping Hadley make a new friend! Nayeli has already been a great aunt for Phoebe when she was little, and now she’s doing the same thing for Hadley. It takes a village, doesn’t it?
Various other helpers later, Hadley has made great progress! By now, he will cautiously approach new dogs with a wag after only 1 or 2 minutes of observing from a save distance.
However, his initial response is still fear, and unless I carefully set up these situations and manage the initial distance, he will default to freeze/stare or hide/flee.
It was interesting to visit his breeder two weeks ago. His mom, dad and brother were there. Tom let Hadley out of the car. Hadley saw his father and immediately hid under the car. His father lowered his head to look at Hadley, and there was a lightbulb moment of recognition – as soon as he recognized the Border Collie in front of him, Hadley was ready to approach and happily greet his dad. Or at least, that’s what it looked like to me. It’s not that Hadley is afraid of his father – but until he recognized him, he wanted to hide.
Hadley, his parents and his brother Horace got to have a little family reunion when we visited the breeder.
What does this mean? Does he have a genetic predisposition to being on the fearful side? His breeder remarked a while ago, when I commented on Hadley being cautious, that he had always been “the most sensitive of the litter”. Is sensitive a euphemism for something else? I don’t know. And in the end, it does not matter. No matter where a certain behavior stems from, the laws of behavior always apply. And these laws are the foundation of all training. Also, no matter who Hadley was yesterday, is today, or will be tomorrow, the one thing that will always be true is that he’s the world’s most wonderful puppy, and the most perfect dog Tom could have adopted 🙂
But back to Hadley’s dog issues:
Part B has also been going well. I’m always armed with clicker and treats anyway, so I’ve been playing LAT with all the random dogs we met on walks. I like how having an objective (teach Hadley that the LAT game is fun!) changes my attitude towards dog encounters: it makes me happy whenever I see a dog in the distance rather than annoyed that I have to change sides or do a U turn. This always happens when I play LAT with a new dog – Pirate and I also had a lot of fun whenever we went out trigger hunting and LAT adventuring. It became one of our favorite bonding games.
As for Hadley, he is becoming an LAT expert. I’ve started naming the behavior, and the distance we can play at has shrunk. We can now play with (calm) dogs on the other side of the street rather than 1.5 blocks away, and after only a few Look-s, Hadley will now switch to offering a different behavior (usually prolonged eye contact or sit). Definitely a success worth celebrating!
Tom and Hadley also participated in my recall workshop the other day. Hadley had to keep a bit of a distance at first, but soon was able to comfortably work near the other dogs, and was happy to play with them after class. He’s a very brave little puppy!
Tom and Hadley testing the quality of treats. Even though the other participating dogs are nearby, Hadley can relax and concentrate on his task.
The nice thing about writing these things down is that it makes me see the progress. When I don’t keep notes, it’s easy to miss out on the tiny little steps of progress I’ve been making every day or every session. It’s like watching a kid (or a puppy) grow up: you see them every day, and you don’t notice how they get bigger – unless every once in a while, you ask them to stand with their back to a door frame and draw a line where their head is. Taking training notes is like drawing lines on a door frame. It helps me see change.
I’ve made another observation that makes it clearer what often happens to clients who have reactive dogs: when I’m out with Hadley in our neighborhood, we hardly ever have an incident. I’m always ready to change sides, make a U-turn, play LAT … Tom and Hadley, on the other hand, still have those encounters where Hadley starts barking or freezes for a moment or two. That means Hadley still practices reactivity.
I’ve been thinking about why this happens to Hadley and Tom rather than Hadley and me, and come up with the following list of reasons. I think being aware of these might help me better coach clients with reactive dogs:
– Until we’ve trained our eyes and brain to selectively focus on dogs in our environment, we tend to see them too late (aka after our dog has already seen them).
– Until we’ve fine-tuned our observation skills to read the fine print in a dog’s body language, we tend to notice fear only when it is obvious – i.e. when our dog is about to react or has already started reacting.
– Unless we have experienced fear ourselves, or really taken an interest in how it works, it is not obvious to us that a puppy’s dog reactivity is a reason to worry in the first place. We tend to assume it’s just a phase he’ll grow out of, or that it will go away with random exposure to dogs, or that a dog is still capable of learning when in fight-or-flight mode.
– Unless we have experienced fear ourselves, or really taken an interest in how it works, it is not obvious to us that aversives are not a constructive solution for reactivity.
– We tend to forget that dogs learn all the time, not only in the training sessions we specify: we’re likely to forget clicker and treats when we take our dogs out to potty rather than setting up for a training session.
– Putting our dog’s safety and comfort level first, even if it means ignoring/stopping/avoiding/standing up to friendly strangers (and their dogs) is an attitude we have to consciously adopt, and to practice.
I wonder how I can make these pieces of the puzzle more accessible to my clients to get them to this point sooner rather than later. I want to minimize their frustration and maximize the quality of their and their dog’s walk. The more “mistakes” happen, the longer it will take for a reactive dog to get over his fears. The longer it takes for our reactive dog, the longer we will have to actively work on his issues, and the longer it will be until being out and about with our canine companion will be the walk in the park will be the uncomplicated, fun activity we’ve been looking forward to.
Of course, this is not to say that Hadley and I don’t run into problems on our walks, too. Walking a reactive dog is hard. It requires both background knowledge, concentration, the desire to be our dog’s advocates, and a number of skills we need to practice: observation skills, timing of the click, and speed (as little time as possible should pass between click and reward). We need to prepare before we go out (clicker, treats, mindset), and keep in mind that like children, our dogs learn every minute – not just when we want to train. Walking a reactive dog is not a walk in the park, it requires your full attention. At least for me, it still requires my full attention. When I don’t pay attention, I often run in a situation I become aware of too late. While walking around my neighborhood has been categorized by playing LAT and hardly any reactive incidents for me, going new places is harder because I don’t know when and where to expect the next strange dog. The other day, Hadley and I were hanging out at a park. He was on leash, and since it was a sunny Sunday and a number of people were out walking their dogs, I ceased the opportunity to play LAT from a safe distance near my car, always ready to retreat behind it, should it be necessary to get another barrier between us and a strange dog. After a while, a woman with her French Bulldog on a flexi lead passed us. Hadley was off the road at a little distance, and on a short leash. It should have been pretty obvious that I was interacting with/training my dog rather than seeking social encounters. The Bulldog came closer, and the woman let it run on the flexi … I politely asked her to stop her dog from coming closer, since my dog was afraid. But what did she do? Let the Bulldog keep running towards us rather than stopping her flexi, telling me, “Well, he has to get used to other dogs at some point, doesn’t he?”
Hadley barked before I had a chance to retreat behind the car. Encounters like this really annoy me. It’s NOT up to you, stranger, to decide when, how and what dogs my dogs are meeting up close. And it is never okay to let your dog run up to a dog on leash without asking. Dogs are on leash for a reason: maybe my dog is scared, or maybe he’s on a leash in order to keep your dog safe from his teeth, or maybe he has flees that I don’t want him to pass on to yours! ALWAYS ask before letting your dog great a strange dog on leash.
Anyways – time to post this update, which is, in fact, already a few weeks old – I just haven’t found the time to finish it yet.
Endlich mal wieder hab ich was zu reviewen, das ich richtig, richtig toll fand, und damit einen Grund, meinen Blog wiederzubeleben! Am Samstag hat Nicole Pfaller zum Thema “unterschiedlich[e] Methoden zur Verbesserung von Ängstlichkeit, Reaktivität und Aggressionsverhalten beim Hund” referiert. Ich glaube, dasselbe Seminar gibt es in absehbarer Zeit zwar nicht mehr, aber Nicole ist demnächst bei Sarina und Kenne von den Doglovers Graz zu Gast, um über BAT zu erzählen – das wird sich sicher teilweise mit dem heutigen Seminar decken. Wer also den Samstag verpasst hat, sollte sich den 5.9. freihalten!
Je mehr Vortragende ich höre, desto mehr wird mir bewusst, wie schwierig es ist, die perfekte Mischung aus Grundlagenwissen (vs. wie viel man voraussetzen kann), weiterführender Information, illustrativen Beispielen/Anekdoten und praktichen Übungen zu finden. Natürlich ist diese perfekte Mischung auch für jede*n Einzelne*n im Publikum anders sein, weil jede*r ein anderes Vorwissen und andere Erwartungen mitbringt. Man kann bei sowas wohl immer nur für sich selbst sprechen. Für mich selbst kann ich jedenfalls sagen, dass Nicole am Samstag die perfekte Mischung getroffen hat. Mir war keine Minute langweilig, wenn Nicole bereits Bekanntes in ihren Worten erzählte, einen neuen Blickwinkel auf lerntheoretische Grundlagen warf, den einen oder anderen mir noch fehlenden Baustein in mein Hintergrundwissen einfügte …
Auch die Vorstellung der einzelnen Methoden, die auf die lerntheoretischen Grundlagen folgte, fand ich ausgesprochen spannend. Das, was man gemeinhin unter den Labels “Angst”, “Aggression”, “Reaktivität” etc. versteht, gehört für mich zu den spannendsten Verhaltensweisen, und ein wissenschaftlich und ethisch fundierter Umgang damit zu den wichtigsten Werkzeugen, die man in seine Werkzeugkiste packen sollte, wenn man mit Tieren arbeitet. Nachdem ich das selbst noch nicht so lange mache, freue ich mich jedes Mal, wenn Trainer*innen, die ich bewundere, ihr Wissen mit mir teilen.
Nicole hat 6 All-Around-Ansätze herausgegriffen und kurz umrissen, jeweils ein oder mehrere Videos zur praktischen Anwendung gezeigt sowie die lerntheoretische Hintergründe beleuchtet. Gerade der letzte Punkt hat mir ausgesprochen gut gefallen. Wenn ich weiß, warum das, was ich mache, wirkt, kann ich es viel bewusster einsetzen und für eine bestimmte Trainingsstrategie argumentieren. Und die Basis für das Warum, die findet sich nun mal in der Lerntheorie – und die ist an sich ja schon ausgesprochen spannend; ich könnte stundenlang darüber hören oder lesen, ohne je müde zu werden.
Nach der Analyse der Lerngesetze teilte Nicole uns jeweils ihre Einschätzung der Vor- und Nachteile der einzelnen Methoden mit. Auch diesen Punkt fand ich sehr spannend – nicht zuletzt darum, weil er der Tendenz, nur eine einzige Methode für gut und richtig zu befinden, die uns leider unter Trainer*innen immer wieder begegnet, so wunderschön entgegengesetzt ist. Zu jeder der vorgestellten Methoden lassen sich Vor- und Nachteile finden, und welche ich anwende, hängt von den Umständen (Hund im Tierheim? Hund einer Privatperson? Etc.) und von den jeweiligen zwei- und vierbeinigen Klient*innen ab. Nicole fand an jeder Methode objektive Vor- und Nachteile (was nicht heißt, dass sie selbst alle Methoden anwenden würde) und betonte, dass es in der Praxis häufig zu Mischformen kommt. Tatsächlich ist das fast immer der Fall – lupenrein sind Methoden höchstens in der Theorie, und selbst dann basieren sie oft auf denselben Lerngesetzen. Es muss nicht darum gehen, die einzelnen Methoden zu ranken und zu vergleichen – sie dürfen einander durchaus ergänzen, und auch ein situationsbezogener fliegender Wechsel kann, ja soll sogar stattfinden. Und wenn wir trotz dem Dschungel an Akronymen auch noch wissen, was wir da eigentlich machen, statt einfach nur draufloszutun, sind unsere Erfolgschancen größer, unsere Erklärungen verständlicher und unsere Umsetzung fehlerfreier.
Die Ansätze, die Nicole vorstellte, waren konkret:
Click and Retreat (Ian Dunbar und Suzanne Clothier)
Der Mensch geht auf den Hund zu, wirft ein Leckerli und entfernt sich sogleich wieder. Alternativ kann das Leckerli hinter den herankommenden Hund geworfen werden, sodass es der Hund ist, der sich entfernen kann.
Lerngesetzen im Hintergrund: klassische Gegenkonditionierung, DRO und R-.
– Ein guter Ansatz für Hunde, die bereits über der Reizschwelle sind. Wenn mir ein bereits aufgeregt kläffender Hund begegnet, kann ich ihm ruhig ein Leckerli zuwerfen, bevor ich mich entferne.
– Click & Retreat gibt dem Hund die Möglichkeit, sich zurückzuziehen – es gibt dem Hund Kontrolle.
– Nur interessant bei Angst/Aggression gegenüber Menschen.
– Aufgrund mangelnder Rückzugsmöglichkeiten für Mensch und Hund schwierig in kleinen Räumen umzusetzen (z.B. Tierheim).
LAT (Look at That!, Leslie McDevitt)
Was soll ich sagen … LAT ist eins meiner Lieblingsspiele. Überhaupt bin ich ein großer Control-Unleashed-Fan und finde, dass sich fast alle von Leslies CU-Spielen nicht nur für Agilityhunde, sondern für jeden Hund eignen. Besonders das Puppy Program ist eins der tollsten Welpenbücher, die ich kenne. Also freue ich mich auch jedes Mal, wenn jemand von LAT redet! Bei LAT wird der Hund dafür geclickt, dass er den Trigger ansieht. Später wird dafür ein Signal eingeführt, und noch später clickt der Mensch dafür, dass sich der Hund zurück zu ihm orientiert.
Die Lerngesetze dahinter sind wiederum die klassische Gegenkonditionierung sowie die systematische Desensibilisierung.
– Emotionale Reaktionen in schwierigen Situationen werden ins Positive verändert.
– Kann in unterschiedlichsten Situationen verwendet werden (z.B. Angst vor Hunden, Menschen, unbekannten Objekten …)
– Muss unter der Reizschwelle geübt werden, um zu funktionieren.
TACT (Touch Associated Clicker Training; Julie Robitaille & Emma Parsons)
Auch diesen Teil und die dazu gezeigten Videos fand ich sehr interessant. Von TACT hatte ich bereits gehört, mich aber noch nie damit auseinandergesetzt. Das werde ich jetzt bestimmt nachholen! Der Ansatz klingt toll, und das Workbook gefällt mir nach einem kurzen Durchblättern. Ein ganz kurz angeschnittener Aspekt, der im Video vorkam, klang besonders spannend für mich: der “secret handshake with strangers”. Ich vermute, dabei handelt es sich darum, dass dem Hund ein Kinn-Touch auf die ausgestreckte Hand beigebracht wird. Wenn ihm ein Fremder die ausgestreckte Handfläche anbietet, darf der Hund dann entscheiden, ob er sein Kinn drauflegen will (oder auch nicht), wofür er von seinem Menschen geclickt und belohnt wird. — Eine spannende Idee! Damit wird der Fremde teil einer vertrauten Struktur, und damit weniger unheimlich. Zugleich bleibt die Entscheidung zur Kontaktaufnahme beim Hund. Auch die Idee, dass der Hund bei TACT sogar an Berührungen gewöhnt wird, gefällt mir gut. Damit geht diese Methode einen Schritt weiter als viele andere.
Nicole meinte, TACT könne auf unterschiedlichste Hunde maßgeschneidert werden, verbindet Massage mit Clickertraining und enthält viele Grundlagenfähigkeiten, die im Alltag nützlich sind. Die Methode soll auch ausgezeichnet durchdacht und in vielen kleinen Schritten systematisch aufgebaut sein, wobei Elemente von LAT, Targeting, Mattentraining und Handtouch vorkommen. Wow – sehr spannend. Ich werde wohl bald mal wieder bei Tawzer bestellen müssen …
– Das Verhaltensrepertoire des Hundes wird erweitert.
– Strukturierter Aufbau. (Ich steh ja auf durchdachte Struktur im Training; Wischiwaschi-Methoden sind so gar nicht meins.)
– Viele Wiederholungen.
– Kann nur für unerwünschtes Verhalten gegenüber Menschen eingesetzt werden.
– Erfordert viel Management im Alltag.
Click to Calm (Emma Parsons)
ist eine der wenigen vorgestellten Methoden, bei denen der Hund nicht notwendigerweise unter der Reizschwelle bleiben muss. DRL wird angewendet, um den Hund von der unerwünschten Reaktion wegzushapen. Diese Methode hält Nicole für wenig sinnvoll und wendet sie auch selbst nicht an.
– Kann bei Aggression gegenüber Hunden und Menschen eingesetzt werden.
– Wirkt bei einem Hund, der ein guter Kandidat dafür ist, unter Umständen schnell, sofern es richtig aufgebaut wird.
– Trotz allem wird das unerwünschte Verhalten verstärkt.
– Unerwünschte Verhaltensketten können entstehen.
Click to Calm habe ich vor Jahren gelesen und erinnere mich nicht an die Details. Ich weiß aber noch, dass mich etwas an dem Buch gestört hatte – es waren aber nicht die Dinge gewesen, die Nicole aufgezeigt hatte. Ein Blick ins Buch, und ich erinnere mich wieder. Es waren Absätze wie der folgende, die mich störten:
(Location 450; Click to Calm. Healing the Aggressive Dog. Parsons, Emma. Waltham: Sunshine Books, 2005. Kindle E-Book.)
Der zitierte Absatz spielt auf längst widerlegte Trainingsprinzipien an, was dem Buch einen unangenehmen Beigeschmack verleiht. Trotzdem enthält es auch Ideen – konkret eine -, die ich gern empfehle: Oft hat der Hund einen potentiellen Stressor noch gar nicht entdeckt, wenn die Reaktion seines Menschen ihn bereits in Alarmbereitschaft versetzt. Wenn es dir nicht gelingt, deine eigenen Stressreaktionen (verkrampfte Körperhaltung, beschleunigter Schritt, schnelles Atmen, plötzliches Luftholen, unwillkürliches Spannen der Leine etc.) zu unterdrücken, lehre deinen Hund in einer entspannten Situationen, dass diese deine Stresssignale Gutes für ihn bedeuten, indem du sie z.B. mit Leckerlis verknüpfst.
Auch die am Ende des Buches angeschnittene Idee, Beschwichtigungssignale unter Signal zu setzen, d.h. sie einem Hund, der Probleme im Umgang mit anderen Hunden hat, quasi wie eine Fremdsprache zu lehren, fand ich beim Lesen damals faszinierend. Um zu sagen, ob ich sie nützlich oder doch eher unangebracht finde, müsste ich mich aber erst näher damit beschäftigen.
CAT (Constructional Aggression Treatment; Kellie Snider, Jesus Rosales-Ruiz)
Auch über diesen Abschnitt habe ich mich besonders gefreut: Ich kannte bisher nur diese Zusammenfassung zum Thema CAT und das Werbe-Video von Tawzer. In dem Video hätte ich, wie auch Nicole in ihrem Vortrag befand, den Hund verstärkt, sobald der Trigger auftaucht, statt zu warten, bis er bellt und dann wieder aufhört – die schriftliche Zusammenfassung fand ich aber sehr spannend und hatte darum schon länger in Betracht gezogen, mir die DVDs zu schenken: Besonders gefiel mir die Idee, dass CAT unter Umständen schneller wirken könnte als die weniger intrusiven Ansätze, mit denen ich bereits vertraut bin. Nach Nicoles Einschätzung bin ich jetzt aber davon abgekommen und werde mir wohl eher die TACT-Serie bestellen.
Und so sieht CAT in der Praxis aus: Der Klient*innenhund bleibt auf einer Position, aber der Trigger bewegt sich: Der Trigger erscheint. Bellt der Hund, passiert nichts, der Trigger bleibt. Sobald der Hund aufhört, zu bellen, verschwindet der Trigger.
Der lerntheoretische Ansatz dahinter ist also negative Verstärkung von angepasstem/ruhigem Verhalten.
Nicole meinte, sie würde niemandem empfehlen, CAT anzuwenden – es sei denn, es geht nicht andres. So könne es in Tierheimsituationen also durchaus sinnvoll sein. Das leuchtet mir so auch ein.
– Wenn gut gemacht, kann es zu guten, schnellen Resultaten führen.
– Nicht geeignet für Otto-Normalhundehalter.
– Muss in verschiedenen Kontexten wiederholt werden.
BAT 2.0 (Behavior Adjustment Training; Grisha Stewart)
BAT ist auch eine Philosophie ganz nach meinem Geschmack: Sie enthält jede Menge Elemente, die im Alltag nützlich sind, und lehrt zugleich, auf den Hund und seine Körpersprache zu achten, um ihm jeweils die größte Kontrolle zuzugestehen, mit der er in einer bestimmten Situation zurechtkommt. Ich fand schon die 1.0-Version gut, und 2.0 gefällt mir ebenso. Die Strandanalogie ist Grisha ebenfalls sehr gut gelungen.
Wie dem auch sei; ich war gespannt auf Nicoles Analyse, die mir dann auch sehr gut gefallen hat:
Lerngesetze: Gestaltung der Rahmenbedingungen, Desensibilisierung, Generalisation durch natürliche Verstärkung (Erkunden der Rahmenbedingungen, Kontrolle), Signaldiskriminierung, R+, R-.
– Der Hund – nicht der Mensch! – hat Kontrolle über die Trainingssituation. Das ist empowering!
– Vor dem Training am unerwünschten Verhalten werden wertvolle Kenntnisse für den Alltag aufgebaut (Leinentechnik, Körpersprache, Survival Skills für den Alltag wie Mark & Move, Find it etc.)
– Die Hunde beginnen wieder, mehr zu kommunizieren. (Ein Riesenplus!!)
– Braucht viele Wiederholungen.
– Schwer verständlich für manche Halter*innen: Es ist gar nicht so leicht, den Hund nicht ständig anzusprechen, sondern ihm einfach nur zu folgen.
Ich mochte auch, dass Nicoles Vortrag kurze Videos beinhaltete – die Länge war jeweils gut gewählt und vermittelte einen kleinen Einblick in die entsprechende Technik, ohne jedoch langatmig zu werden, und machte Lust, selbstständig weiter zu recherchieren, wenn man eine gewisse Technik noch nicht kennt. Nicole wählte teils Videos der jeweiligen Trainer*innen (Grisha, Jesus Rosales-Ruiz & Kellie Snider), eigene Videos aus ihrem Trainingsalltag mit eigenen Hunden oder Kund*innenhunden, aber auch andere Videos z.B. Donna Hill, Sarah Owings oder Jennie Murphy. Gerade dieser eklektische Ansatz hat mir gut gefallen, und auch, dass die vorgestellten Videos nicht immer “perfekt” waren. Schließlich ist auch der Alltag nicht perfekt, und wenn wir uns in der Theorie noch so genau überlegen, wie unser Training ablaufen soll, ist unser Timing dann doch manchmal ungenau, die Leine zu straff, oder wir erkennen erst im Nachhinein, dass wir vielleicht früher vom Trigger hätten abdrehen sollen. Jeder Hund und jede Situation sind anders. Auch orientierten sich die einzelnen Trainingseinheiten in den Videos nicht immer “lupenrein” an einer einzigen Technik, sondern mischten verschiedenes, was die Realität sehr gut widerspiegelt.
Eine weitere Auflockerung wurde durch zwei praktische Übungen erreicht: Einerseits gab’s eine Übung zu Grishas Leinentechnik, andererseits ein Clickerspiel. Ich hatte jeweils tolle Partnerinnen, viel zu lachen und jede Menge Spaß.
Die Leinentechnikübung zeigte sehr gut, wie sanft die Einwirkung beim Slow Stop sein kann, um am Hundeende der Leine doch gefühlt zu werden. Nicole hatte für jedes Team eine Leine dabei, zeigte die Technik kurz vor, erklärte sie wirklich verständnisvoll (Grishas eigene Erklärung auf der DVD hatte ich etwas verwirrender empfunden) und gab uns dann die Möglichkeit, das Ganze in 2er-Teams auszuprobieren. Genug Zeit, um mehrmals die Rollen zu tauschen, aber nicht so viel Zeit, dass uns langweilig hätte werden können.
Die zweite Übung machte gleich noch mehr Spaß. In 3er-Teams gab es jeweils eine*n Schüler*in, eine*n Lehrer*in und eine*n Beobachter*in. Die Schüler*innen wurden aus dem Raum geschickt, während die Lehrer*innen und Beobachter*innen instruiert wurden. Ich war Schülerin, und nachdem ich wusste, dass Nicole kürzlich mit Jesus Rosales-Ruiz und Mary Hunter PORTL gespielt hatte, war ich der festen Überzeugung, dass eine komplexe Verhaltenskette auf uns zukäme.
Schließlich durften wir wieder reinkommen und fanden auf unsren Plätzen mehrere Gegenstände vor:
Wir wurden in 3 10-Verstärker-Durchgängen von unseren Lehrer*innen geclickt und sollten nach dem 10. Verstärker jeweils aufschreiben, wofür wir meinten, verstärkt worden zu sein, und wie wir uns dabei gefühlt hatten – eine gut durchdachte und ausgesprochen spannende Übung, wie sich herausstellen sollte. Ich kam nach einem anregenden Gespräch über Shelties und JAD-Dogs zurück, freute mich auf die Übung, sah die Gegenstände auf meinem Platz und war bereit, damit zu interagieren. Sobald das Startsignal gegeben wurde, hob ich mit der rechten Hand den Fuchs hoch, wurde geclickt und probierte, ob es auch für das Anheben der Taschenlampe einen Click gäbe. Ja! Und was war mit dem Plastikzahn? Ebenfalls! (“Hah, ich bin so gut!”, denkt sich Chrissi und freut sich an ihrer hohen Verstärkungsrate.) Ich probierte die weiteren Gegenstände durch und begann nach den ersten 4 oder 5 Clicks, mich etwas verunsichert zu fühlen, weil ich kein System erkennen konnte. Ich erhielt eine hohe Verstärkungsrate über die ersten 10 Verstärker hinweg. Anfangs fühlte ich mich gut (yey, hohe Verstärkungsrate!), dann unsicher (alles wird geclickt? Ich erkenne kein System dahinter!) Beim zweiten 10-Verstärker-Durchgang variierte ich die Reihenfolge der Gegenstände, um rauszufinden, ob es um eine bestimmte Reihenfolge ging. Nein! Wiederum war die Verstärkungsrate hoch und ich wurde für alles geclickt, obwohl ich diesmal auch die linke Hand einsetzte und den Zahn bewusst mit der Unterseite nach oben zurücklegte. Nach einem kurzen Hochgefühl aufgrund der hohen Verstärkungsrate stieg meine Irritation weiter an, weil ich immer noch kein System erkannte. Im dritten 10-Verstärker-Durchgang gab es plötzlich keinen einzigen Click mehr, auch nicht für das, was bisher funktioniert hatte. Häh? Was bitte sollte das?! Ich war ratlos und versuchte, die einzelnen Gegenstände höher anzuheben bzw. alle gleichzeitig anzuheben – das war wohl mein Extinction Burst! – und dann wurde die Session nach 30 Sekunden beendet. Ich war mehr ratlos als frustriert, weil ich den Eindruck hatte, nicht ich als Lernende sei “schuld” daran, dass ich das System nicht erkannt hatte, sondern meine Lehrerin – ganz offensichtlich hatte sie vergessen, den Schwierigkeitsgrad zu steigern, was sie schon im Laufe der ersten 10-Verstärker-Session hätte tun sollen, weil ich ja offensichtlich so toll gewesen war und alles richtig gemacht hatte, und dann, in der dritten Session, hatte sie den Schwierigkeitsgrad plötzlich viel zu schnell gesteigert, sodass ich keine Chance mehr hatte, erfolgreich zu sein. Gutes Shaping sieht anders aus, dachte ich mir, während ich sie ausfragte, was sie denn nun eigentlich gewollt hätte. Immer noch war ich der festen Überzeugung, dass ich hätte geshapt werden sollen, und ich wollte jetzt verdammt nochmal bitte endlich wissen, was das Zielverhalten gewesen war! Erst, als meine Lehrerin und die Beobachterin mir immer noch nicht sagen wollten, worum es gegangen war, wuchs meine Frustration. Hey, ich hatte mitgespielt, hatte mich bemüht – ich hatte mir die Lösung redlich verdient, verdammt!
Endlich wurde des Rätsels Lösung verraten: Es war überhaupt nicht um Shaping gegangen, sondern darum, tatsächlich in den ersten beiden Durchgängen für alles zu clicken und im letzten für gar nichts mehr. Das hätte ich so gar nicht erwartet; nie wäre ich darauf gekommen. Ich finde es ausgesprochen spannend, dass ich dieses nicht vorhandene System nicht durchschaut hatte, ja nicht mal im Traum darauf gekommen wäre, dass es kein System dahinter geben könnte! Und fast noch spannender fand ich, wie emotional involviert ich war und wie groß mein Wunsch, endlich die Antwort zu erfahren. (Shaping is, after all, all about surfing the extinction burst …!)
Auch das Feedback der Beobachterin fand ich spannend. Sie meinte, ich sei neugierig, aufgeregt und nervös gewesen, als ich in den Raum gekommen sei, und hätte bereits einen der Gegenstände (das Papier-Ding mit dem Flugzeug drauf) angefasst, bevor die Session überhaupt losging. Dann hätte ich sofort durch Aufheben mit den Gegenständen interagiert – vehementer, als sie selbst das getan hätte. Sie konnte sowohl meine Freude über die hohe Verstärkungsrate als auch meine Ratlosigkeit beobachten. Beim ersten Durchgang schaute ich noch auf die Gegenstände, beim zweiten dann zwischen Gegenständen und Lehrerin hin und her (Gibt mir ihre Körpersprache einen besseren Hinweis als der Click?), und beim dritten Durchgang schaute ich nur noch die Lehrerin an. Bereits nach fünf Sekunden ohne Click machte ich ein enttäuschtes Geräusch und hob die Gegenstände höher an, bzw. mehrere zusammen. Meine Beobachterin fand meine deutliche Enttäuschung nach bereits 5 Sekunden bemerkenswert.
Was mich besonders amüsiert: Ich sehe jede Menge Parallelen dazu, was ich in meinem Pudel beobachte, wenn ich sie zu selten clicke: Ihr “What-the-fuck-give-me-a-hint-mum! Any-hint-will-do!” ist ein forderndes Schnappen in die Luft bzw. Klappern mit den Zähnen. Und genau wie ich ist sie mit Feuereifer bei der Sache, wenn die Verstärkungsrate hoch ist, wirkt aber ebenfalls irritiert bzw. hört schließlich auf, zu experimentieren, wenn ich zu lange bei denselben Kriterien bleibe. Erst nach diesem Spiel wird mir jetzt ganz deutlich klar, was ich vor ein paar Monaten in Phoebe beobachtet habe, während ich Sue Ailsbys Shaping-Kurs besuchte. Ich nehme wohl an, ich löse manchmal ganz ähnliche Gefühle in meinem weißen Flauschtier aus wie meine Lehrerin am Samstag in mir. Ich bin meinem Hund wirklich ausgesprochen ähnlich: hochmotiviert, aber mit geringer Frustrationstoleranz!
Die Implikationen dieses Spiels sind faszinierend:
Wird zu lange dasselbe verstärkt, führt das zu Verwirrung, und wird nichts verstärkt, resultiert das in Frustration. Zweiteres war mir bewusst – ersteres nicht wirklich! Auch kommt das, was wir zu verstärken glauben, oft gar nicht so beim Lerner an. In einer anderen Gruppe war die Schülerin zum Beispiel der festen Überzeugung, bereits im ersten Durchgang würde nur bestimmtes Verhalten ihrerseits verstärkt. Extrem spannend. Ich glaube auch, wenn man so etwas selbst ausprobiert, ist die Wahrscheinlichkeit, es auch beim Hundetraining im Hinterkopf zu behalten, größer, als wenn man sich damit nur in der Theorie auseinandersetzt. Das war ein ausgesprochen gut organisiertes und kurzweiliges Spiel – danke, Nicole!
Anfang und Ende des Vortrags bildeten einen schönen philosophischen Rahmen für den Mittelteil. Nicole hatte uns zu Beginn daran erinnert, dass agonistisches Verhalten ein ganz natürlicher Teil des Ausdrucksverhaltens unserer vierbeinigen Freunde ist und wir im Grunde froh sind, dass unser Hund in der Lage ist, uns mitzuteilen, wenn ihm etwas zu viel wird. Ein wichtiger Gedanke, den man nur zu leicht vergisst!
Nach dem vermeintlichen Shaping-Spiel griff Nicole diesen rosa Faden wieder auf, indem sie uns ins Gedächtnis rief: “Der Organismus der Lerner hat immer recht” – er kann mit seiner Lernerfahrung in jeder Situation nur jeweils so reagieren, wie er das eben tut. Folglich kann es gar “kein ‘abnormales’ Verhalten [geben].”
Das ist nicht nur wissenschaftlich gesehen wichtig, sondern auch ethisch gesehen: Solange wir im Gedächtnis behalten, dass ein Tier immer genau das Verhalten zeigt, das es in einer Situation zeigen kann bzw. muss, gibt es keinen Grund, uns über das Tier zu ärgern – allerhöchstens über uns selbst, weil wir ihm nicht verständlich genug erklärt haben, was wir von ihm wollen.
Auch die beiden anderen Rahmen, die Nicole zu Beginn ihres Vortrags gezogen hatte, schlossen sich am Ende wieder: Erst gab es eine kurze Zusammenfassung, die sich auf die Gemeinsamkeiten der unterschiedlichen Methoden konzentrierte und unter anderem betonte,
– dass das Ziel jeder einzelnen Methode sei, die Lebensqualität der Lerner*innen zu verbessern,
– dass sie alle nicht nur am Problem trainieren, sondern auch wertvolle Alltags-Fertigkeiten aufbauen,
– mit gestellten Set-Ups arbeiten
– und mit zwei Ausnahmen (Click to Calm und CAT) Wert darauf legen, unter der Reizschwelle zu bleiben,
– dass sie alle eine gewisse Zeit in Anspruch nehmen
– und, was ebenso wichtig ist, dass die verschiedenen Ansätze in der Praxis verschwimmen.
Auch diese Schlussbemerkung gefiel mir sehr, weil sie wiederum dem Methodenkrieg, der sich – zumindest im deutschsprachigen Raum – immer wieder abzeichnet, entgegensteht und zeigt, dass es nicht nur vertretbar, sondern sogar wünschenswert ist, sich umfassend fortzubilden und aus verschiedensten Ansätzen für die Praxis eben das herauszuholen, was für einen speziellen Fall am besten passt.
Die letzte Klammer, die Nicole am Ende schloss, ist die allerschönste: Ganz am Anfang hatte sie bereits erwähnt, dass eines der Ziele ihres Vortrags “die Verbreitung tierschutzgerechten Trainings auf Basis wissenschaftlicher Konzepte” sei. Hach, ja! Genau darum sollte es doch in allem gehen, was wir machen! Nicht darum, das eigene Wissen geheimzuhalten, nicht darum, sich zu ärgern, wenn jemand eine unserer Ideen umsetzt, sondern darum, den gemeinsamen Pool an Wissen und ethischen Trainingskonzepten zu vergrößern, indem wir Wissen weitergeben und teilen und damit – wie auch die einzelnen vorgestellten Methoden – die Qualität unseres Trainings erhöhen. Ich mag diesen Gedanken wirklich gerne. Ich glaube, dass unsere Gesellschaft umso freier und humaner wird, je mehr Zugang jede*r Einzelne zu Bildung hat und je größer unser kollektiver Wissensschatz wird – nicht nur, aber auch im Hundetraining. Wissen wird nicht weniger, wenn man es teilt, sondern mehr. Praktische Erfahrungen muss ohnehin jeder für sich selbst sammeln.
Noch ein wunderschönes Nebenbei-Statement von Nicole würde ich gern erwähnen, weil mir auch das sehr gut gefällt und ebenfalls einem Trend entgegensteht: dem Unterschätzen der Hundehalter*innen. Ist das ein Trend im deutschsprachigen bzw. europäischen Raum, oder ist es ein allgemeiner? Zumindest in den Online-Communities, denen ich angehöre, habe ich mitunter den Eindruck, dass der englischsprachige bzw. US-amerikansiche Raum respektvoller mit Hundehalter*innen umgeht als der unsrige. Es scheint, dass manche Techniken darum abgelehnt werden, weil Trainer*innen davon ausgehen, dass Hundehalter*innen diese Techniken nicht verstehen oder falsch umsetzen würden. Für mich impliziert eine solche Einstellung, dass der oder die durchschnittliche Hundehalter*in “dumm” sei – jedenfalls dümmer als die oder der Trainer*in, die die entsprechende Methode schließlich auch verstanden hat.
Nicole begegnete einem entsprechenden Kommentar aus dem Publikum, indem sie überzeugt erklärte, dass ihrer Erfahrung nach Hundehalter*innen sehr wohl in der Lage seien, das Kleingedruckte der Körpersprache lesen zu lernen. Das finde ich schön, und ich teile ihre Meinung da auf jeden Fall. Ich gehe davon aus, dass die meisten dazu in der Lage sind, all das zu lernen, was ich selbst weiß; schließlich bin ich nicht klüger oder irgendwie “besser” als meine Mitmenschen, sondern habe höchstens ein kleines bisschen mehr Erfahrung oder Wissen in Bezug auf Fachgebiet A, während mein Mitmensch wiederum mehr Erfahrung oder Wissen in Bezug auf Fachgebiet B mitbringt. Und genauso, wie ich davon ausgehe, dass ich eine Expertin in Fachgebiet B werden könnte, wenn ich das wollte oder es nötig werden sollte, gehe ich auch davon aus, dass jeder meiner Mitmenschen ein*e Expert*in in Fachgebiet A werden kann, wenn er will oder muss. Genau wie unsere Hunde sind wir Menschen nämlich richtig gut darin, uns anzupassen und Neues zu lernen, sofern wir auf eine entsprechende Verstärkungsgeschichte zurückblicken. Und genau wie es nicht die Schuld des Organismus Hund ist, wenn er ein Trainingsziel nicht erreicht, sondern wir es verständlicher oder schlicht noch einmal erklären sollten, ist es nicht die Schuld des Organismus Mitmensch, wenn er eine Trainingsmethode nicht versteht – vielleicht sollten wir uns einfach nochmal gemütlich zusammensetzen, zuhören, auf Fragen eingehen und Unklarheiten beseitigen, ohne uns angegriffen zu fühlen.
Auch dieses respektvolle Eingehen auf Fragen und abweichende Meinungen – und auch das ist etwas, das ich umso mehr zu schätzen weiß, je mehr Vortragende mir begegnen – ist Nicole am Samstag ganz ausgezeichnet gelungen. Sie hat das Publikum zur Mitarbeit aufgefordert und sich offen auf Fragen eingelassen. Schön finde ich das – so schön. Denn, und das vergessen wir leider auch viel zu leicht: Nicht nur unsere Hunde, sondern auch unsere Mitmenschen profitieren von Empowerment.
Variety is a reinforcer. When it comes to dogs, this holds true for toys, smells, environments, and also food. I try to offer a variety of the above to my dogs. There are usually several toys out in the living room, but I will rotate the selection of available toys every couple of days. That is to say, for the last days, for example, there have been 3 plush toys, a squeaky shoe and a rubber ring out in the living room. I’ll probably rotate toys tonight, exchange the plush toys for different ones, take away the rubber ring and put out a nylabone or a squeaky hamburger instead.
As for food, my dogs sometimes get frozen Kongs with canned dog food, sometimes kibble which may be either offered in bowls or in puzzle feeders, sometimes a variety of raw meats with a variety of fruit and vegetables, with or without joyghurt and butter milk, with a variety of oils, … They’ll sometimes get ostrich bones to chew on, sometimes beef bones, sometimes rawhyde, sometimes all sorts of dried animal parts from pigs’ ears to bulls’ pizzles. And the treats we’re working with are as diverse as I can come up with; there’s cheese, dried fish, hot dogs, store-bought dog treats, home-baked dog treats, bread crumbs, liver paté, salmon paté, …
Today’s doggie breakfast
Also, of course, our outings take us to different places, to the city, to the forest, to the meadows, to the dog park, to the lake, to the shopping mall …
The reinforcing quality of variety has been confirmed in formal studies, and it has been shown that “[a]nimals with variety in their lives are healthier and happier, just like people” (Schneider, 26).
Control as a reinforcer
However, it has been found that control (over variety) is an even stronger reinforcer than variety itself: getting to choose from a variety of options is more reinforcing than a variety we cannot control. Schneider (28) quotes a study with nocturnal deer mice. Being nocturnal, these mice prefer darkness over light. As would be expected, when they learned to operate a light switch in an experiment, they would choose to turn off the light. However – and this is the interesting part -, when the light was automatically turned off every 30 minutes, the mice would turn it back on! Control (being able to operate the light switch and manipulate their own environment) trumped their preference for darkness.
Deer mouse (peromyscus maniculatus)
What does that mean for ourselves, and for dog training? Well, for me, it means that I will sometimes choose a flavor of chocolate or ice cream that isn’t my favorite one, simply because I like variation and making little choices. It means I like going to restaurants, because ordering is fun. Also, it means that if I had children, I’d give them as many choices as they could handle: which sweater do you want to wear? What do you want for breakfast?
When it comes to my dogs, I have observed that especially Phoebe loves choosing her own toy from the toy box. Even though she knows what toys are in there and she gets all of them on a regular basis, she has the best time if I just present her with the open box and she can pick one. I’ve always done this, but never had an explanation why choosing would be more fun than simply being presented with a different toy every day. Schneider’s book brings it all together.
All my dogs also get to pick their chewy treats every once in a while: I’ve got a variety of different ones in a box, and I’ll offer the box to them and let them pick one.
Schneider, Susan M. (2012): The Science of Consequences. How they affect genes, change the brain, and impact our world.
When you’re interested in training animals, when you work with animals with behavior problems, or when you suffer from behavior problems yourself, sooner or later, you’ll find that life is messy. Now there are two kinds of training-and-behavior people. The ones who seem to say: life is messy. Avoid the mess (Turid Rugaas, Anne Lill Kvam etc.), and the ones who say: life is messy. Clean it up. (Behavioral analysists.) I, for one, am clearly a “Clean it up” kind of person: I want to pinpoint the mess, see what precedes it and what it results in, and then I want to organize it in whichever way I please. In other words: I love behavioral science. I want things to be explainable, quantifiable, mesurable, and I want to be able to change behavior rather than avoid the conditions that make it occur. For me, science equals control (over myself and my environment). Control equals empowerment. And I want to be empowered! (And I want my dogs and students to be empowered, too.)
Enter Susan Friedman. Susan Frieman who says just that: we can use behavioral science to empower our companion animals! Hell, yeah, I’m sold! I’m sold because of Susan’s philosophy of empowerment that frames her talk. I’m sold because her appreciation of her audience is as tangible as appreciation can be, her smile is real, and her ear is open to our every question. Also, I’m sold because it’s hard to organize a great seminar, and it’s hard to give a great seminar. Susan Friedman and Happy-Fellow managed to do just that. And this is how.
Attending seminars with renowned speakers flying across an ocean in order to speak in Austria is expensive. It always is. Passion is what makes us happily pay the seminar fee. However, if we pay a lot of money, we expect something in return – not just from the speaker, but we also expect the general set-up to adhere to certain standards. We expect an environment that makes listening and learning fun and easy. We expect a spacious seminar room, a sufficient number of sockets for everyone who wants to plug in a computer, and a comfortable temperature. We expect lunch breaks with good food and an opportunity to connect with likeminded people. We expect free fresh water. We expect free W-LAN so we can tweet and post about the awesomeness of the speaker on Facebook in real-time.
While it’s always expensive for the organizer to invite a renowned speaker, and it’s connected with months of work. It’s resource- and time-consuming to organize a conference or a seminar, and to organize it well. If you organize such an event professionally, pay your speaker a fair salary, and keep the fee affordable for the audience, you’re practically a volunteer. You can’t host a well-organized event like Susan Friedman’s seminar for financial gain, because the hours and hours of work you’ll put into preparing, planning and setting up, and the amount of money it costs to rent a decent seminar room and pay for a delicious coffe and lunch buffet for your audience doesn’t leave much for yourself. Especially if you think of all the little things such as goodies for seminar bags or ruffle rewards and offer early-bird registration fees. I’m happy to report that the Seminar organized by Happy-Fellow‘s Nicole M. Pfaller and Gernot Sadovsky exceeded my expectations.
And now, let’s get to the content!
“Like gravity, the laws of behavior apply to all living animals, whether or not we recognize them at work. These are natural processes in the wild, in captivce facilities, and in our homes. We didn’t invent them, we just harness them!” (Susan Friedman)
Those of you who know me know that I’m addicted to writing seminar reviews. I feel like everyone should know how awesome certain speakers are! So here goes. These are some of the areas Susan discussed that I found most interesting.
– Operationalize your labels!
The problem with labels is that they give us information about how the speaker feels about the animal rather than how the animal behaves. In order to successfully change behavior, we first have to try and get rid of labels. “We can’t teach animals what to be but we can teach them what to do and when.” (Slides part 1, p. 10; my emphasis). I love how how Susan fine-tuned our awareness of our use of labels and insisted we operationalize them in order to come up with observable data: don’t teach “friendly”. Teach “recall” and “relaxed body”! Don’t talk about “reactivity,” but operationalize, i.e. describe what can be observed: does your dog lunge? Bark? Etc. “Lunging” and “barking” are descriptions of observable behavior that can be changed. “Aggressive” is a label – and may well be a self-fulfilling prophecy (4).
We can operationalize concepts like “aggression,” “fear” etc. on a physiological, neurological, and overt behavioral level. The latter one is most interesting for designing a behavior modification program, since these are the reactions we can observe (e.g. freeze, fight or flight, duck, panting …) And yes, of course we can modify these reactions: “Although the basis of fear is innate, the antecedent stimuli that trigger it can be acquired through respondent learning.” (slides part 2, p. 1)
Susan’s wonderful pop culture example for labels was Gloria’s encounter with the “skittish” horse. Those of you who are watching Modern Family will know!
– A-B-C assessments!
Oh, yes! I love functional assessments.
– Common misconceptions:
Misconception: behavioral science based trainers don’t believe that animals have emotions.
Rectification: behavioral science does not make such a claim. We just know that we cannot measure emotions. We can never know the experience inside an animal (not even if this animal is a fellow human being). Therefore, we rather focuse on behavior: behavior is “[w]hat an animal does, given certain conditions that can be observed”. (slides part 1, p. 13).
Misconception: there is a dichotomy between nature vs. nurture. Either something is in our nature, or it is based on our environment.
Rectification: nope! It is our nature to learn; i.e. nature and nurture go hand in hand. E.g. our experiences activate or do not activate certain genes.
Misconception: lots of the behaviors you train are unnatural behaviors!
Rectification: there is no such thing as an unnatural behavior. If it were unnatural, it would not be possible for an environment or experience to enable such behavior. Animals learn lots of things in the wild, and they learn lots of things in captivity. These things may be different, but none of them are more or less “natural.” Remember: nature and nurture go hand in hand.
Misconception: behavioral science is mechanistic, or simply a “carrot on a stick” approach!
Recrification: the “[b]ehavioral model has a high degree of scientific rigor, and social significance but a low profile in society at large” (slides part 1, p. 6). Our culture makes it hard for us to see behavior as a tool for a purpose (like, for example, our eyes are tools for seeing). Behavior is an evolved tool to affect the environment in ways that are valubale for us.
Why, however, do we tend to disregard the behavioral model as mechanistic or simplistic? Susan suggests it is because we tend to be biologically deterministic: we tend to assume that we are our genes and believe the behavior that goes with them is just, as Susan has it, “spurting out of us”. However, we are not our genes: our genes are quiet until the environment pulls them (i.e. pulls behavior our genes make us capable of). Our genes are “turned off or on”, depending on the environment.
One example commonly cited for the genetic base of our behavior used to be the fact that cowbirds growing up in a sparrows’ nest speak cowbird, not sparrow, in the end of the day. However, studies showed that when a young cowbird in the nest makes a cowbird sound, the female cowbirds within range will answer and therefore reinforce the cowbird sound. (It is a reinforcer on the neurochemical level of analysis.) Our level of analysis is not the neurochemical level, but the level of the environment pulling behavior. What takes place with the young cowbirds is a beautiful example of operant learning. It is not mechanical or simplistic – it is effective, useful, and fascinating (if you ask me, anyways).
Misconception: “… but positive reinforcement doesn’t work!”
Rectification: this is an oxymoron. It is impossible for positive reinforcement to not work: if it doesn’t increase the frequency of the behavior, by definition, it isn’t R+!
– The law of effect: we behave today based on the consequences our behavior had yesterday.
Once we understand the relation between what we do and the outcome, we can change observable consequences by changing conditions. Behvior is a function of its consequences.
– R+, or: putting money on the relationship account
If I share a history of positive reinforcement with an animal, she is more likely to comply than if I don’t share any history with this animal: every time I use positive reinforcement, I put money into the relationship account. Relationship, of course, is a label. Let’s have it mean: when an animal chooses to approach rather than remain distant.
This is to say: if I use lots of R+, I get an animal who will be much more likely to comply than if I just remain neutral, “just letting the animal be” rather than putting money on the relationship account. This is one of the reason I love R+: it creates a deeper relationship and an more active and vivid dialogue between two species than any other training method.
– “We need to empower the animals to be participants in our conversation.”
I love Susan’s philosophy. It reflects the way I see clicker training: I want my dog to train me as much as I train her. It’s not necessarily about teaching perfect behaviors – the clicker becomes a mediator, a tool that bridges the language gap between two species (dog and human): I click when the dog does something I want. The dog can elicit the click (and treat) by means of figuring out what will make me click.
Susan emphasized that especially if a behavior is not under perfect stimulus control, there is a dialogue going on. Two-way communication! And that dialogue is one of the most fascinating, rewarding aspects of training!
What we do – says Susan – is not “Clockwork Orange”. Rather, we want to influence one another. If people watch me and say that my dog is training me rather than the other way round (we’ve all been there, haven’t we?), there’s nothing wrong with that: of course we teach our dogs or our children that they can “manipulate” us. After all, that’s what conversation, what dialogue is about!
I like that Susan brought up this point, since I have had disagreements with trainers who reject the clicker on the basis of its potential for causing addiction. An example commonly cited for the supposed perils of of the clicker is the client who manages to reinforce behaviors, but doesn’t manage to put them under complete stimulus control. It is argued that this addiction to the click stresses the animal a lot by means of putting constant performance pressure on him.
I can follow the argumentation, but I don’t agree. Of course, you can misuse any tool (including a clicker). However, I think in the majority of cases, even if a client doesn’t manage to put perfect stimulus control on her dog’s behaviors, the clicker empowers her dog and facilitates two-way communication between the species. The client will end up with a dog who offers behavior in certain situations (under certain circumstances where offering behavior has been reinforced in the past). In other situations (e.g. client falling asleep on the couch, client in bathtub (unless, that is, said client has read Peggy Tillman’s book!), her dog is unlikely to offer behaviors because they have not been reinforced under these conditions in the past. The dog will not have lost down time and periods of relaxation. Rather, he will have gained periods of conversation that can be mutually initiated.
Furthermore, there is another powerful argument against the “addiction” theory: if every click is followed by a reinforcer, your clicker is a coke machine rather than a slot machine. And while we get addicted to putting money in slot machines, I’ve yet to meet someone addicted to feeding the coke machine.
– Every click should be backed up by a reinforcer.
Clicking should be like a soda machine, not like a slot machine. “If [the] click doesn’t reliably predict [the] back up reinforcer, then it really becomes a bridge that leads to nowhere.” (cf. slides part 2, p. 19) If it is not backed up with a well established reinforcer, the animal will find a more reliable marker – one that is 100% reliable that the treat will come (e.g. your hand reaching into your treat bag). The strength of the click as a secondary reinforcer depends on its reliability as a predictor of the treat!
I always treat when I click. However, some good clicker literature suggests that it’s easiest to reinforce every click, but that it is also possible to only treat after some clicks, though this might be am approach suggested for expert rather than novice trainers.Backing up each click with a reinforcer doesn’t equal treating every single time I give a cue. It just means that if I click, then there will reliably be a tangible back-up reinforcer (food or toy). If I say “Good girl!” instead, there won’t necessarily be a back up reinforcer. If I say “Yes!”, what follows will often be a Premack reward (go get that pigeon; say hi to that dog!).
I really liked Susan’s illustration of why it makes more sense (and is not just easier) to always back up the click with a reinforcer:
(part 2, p. 22)
The coke machine operates on a continuous reinforcement schedule (just like my click): this also makes it most reliable in predicting my actions: whenever I am thirsty, I will reliably go to the coke machine. It’s not stressful; I am 100% sure that if and when I put in a coin, I’ll get a soft drink (which reinforces my future behavior the next time I’ll be thirsty and encounter a coke machine.) In case the coke machine is broken, I’ll probably throw a tantrum or kick it, because I expected it to work, but then I’ll walk away rather than throwing in more and more coins to no avail.
If I back up every click with a reinforcer, my dog will not be stressed by clicking and will not have to throw a tantrum. But if and when he’s hungry, he’ll most reliably work for treats, and when he’s wide awake, he’ll most reliably work for tennis balls. Score!
Slot machines work differently: they operate on an intermittent reinforcement schedule. That is to say, they are very unreliable predictors of reinforecers. If I throw in a coin, I don’t necessarily expect it to give me a reinforcer. If it doesn’t reinforce me, I won’t throw a tantrum, but I will throw in another coin. And another. And another. Slot machines build persistence. They can be more frustrating and stressful than coke machines, and they do not reliably mean that whenever I’m hungry/thirsty, I’ll throw money in a slot machine.
In animal training, what we generally need is quick learning and strong behavior – not persistence. Continuous reinforcement builds very STRONG behavior (whenever I’m thirsty, I WILL throw money into the coke machine) and facilitate FAST learning. This is what we want for most training goals. Continuous reinforcement does not build PERSISTENCE, but persistence is quite unnecessary for most training goals anyways. So: if you click, treat! (2)
– Empowerment and control
Remember how I (and, well, Susan, of course) said we need to empower animals to be particionats in our conversations? Then I gave you an example of the clicker as a facilitator of conversations. However, we haven’t yet defined what “empowerment” means: It means that the animal has control over its environment. More precisely, we are talking about the learner being able to control its own outcome – i.e., to use behavior effectively. Control is central to behavioral health, and (maybe we should even say: because) it is a primary reinforcer.(1) This is, by the way, also the explanation for the fact that punishment is reinforcing to the punisher: she experiences control over her environment and is reinforced by this experience.
A lack of control over one’s own outcomes leads to a lack of resilience. If we (or other animals) lack resilience, we’re going to have a hard time in this world! We are born with resilience, but that resilience is a function of how much control we have over our outcomes. This doesn’t mean that every animal must have control over every single outcome – this is simply not possible. Rather, it means that the “ratio of control:no control should be very high”.
– Control via enriched environments
The notion of control relates very nicely to the effectiveness of enriched environments. Susan Friedman finally gave me the scientific explanation I’ve been missing in Anne Lill Kvam’s course. Anne Lill keeps insisting on enriched environments as part of nearly every (?) problem behavior intervention program we define as trainers. Sometimes, she even suggests regular exposure to enriched environments as the sole solution to various problems clients might have with their dogs. I heard what she said, but I never understood why exactly enriched environments would be so effective.
Last weekend, Susan gave me the scientific explanation I need in order to embrace Anne Lill’s training approach and recognize its scientific validity. “Empowered animals are brave learners”, as Sarah Owings has it. That is to say, the more an animal experiences being able to control its environment, the bolder it will become towards new stimuli. “Bolder” means that the animal will curiously approach a novel object rather than engage in distance-increasing behaviors. If you want your young parrots to grow up to be bold adults, you’ll let them grow up in enriched environments, i.e. with lots of stimulus diversity (i.e. change the cage environment every day). Birds are not wary by nature, but wary by experience! If your bird grows up with little or no stimulus diversity, it will grow up to be wary.
The same holds true for raising puppies. If you raise a puppy according to Ian Dunbar’s advice, you’re on the best way to have a bomb-proof adult dog, since Dunbar’s focus is on the greatest possible stimulus diversity for your puppy.
This is why enriched environments may also help (adult as well as young) dogs to experience control over their own outcomes via interaction with the environment. Remember, control is a primary reinforcer. Getting lots of this primary reinforcer of control leads to a bolder animal. (An object in the environment is only considered enriching, btw., if the animal is interested in interacting with it. Otherwise, it is purely ornamental.)
– Resilience: other contributing factors
Having a rich bank account from R+ training also adds to the animal’s resilience. This resilience is useful in the few situations (e.g. emergency) where we have to force an animal. For example, this may happen at the vet’s. If we have built up resilience, the animal will “bounce back” quickly after a negative [negative used in its everyday sense] experience. Because the animal is empowered the rest of the time, it is okay to withdraw a little from that bank account when it can’t be avoided. It’s not about empowerment 100% of the time. Life on earth is messy and not always empowering, and that’s okay. It’s about the ratio of positive reinforcement/empowerment:punishment/force. There always has to be a lot more weight on the R+/empowerment side of that scale.
It’s okay to be relaxed about controlling the environment! You don’t need to keep it free from negative [everyday use of the word] stimuli. Never being exposed to even mild aversive stimuli might even shatter the subject: the person may lack the experiences that lead to resilience.
Susan’s thoughts on that matter (and her telling of an anecdote involving her daughter Marnie and a roll of toilet paper!) were prompted by Christine Schragel’s question. Christine had an experience with one of her dogs that strengthens the assumption that it may be counterproductive to shelter your puppy from all and even the mildest aversive stimuli.
However, there are renowned trainers out there who teach that, indeed, especially when growing up, a dog should – ideally – never be exposed to even mild aversive stimuli. That is to say, you should strive to control the environment as much as possible. There are more and less radical views on that matter out there, but I would venture that trainers like Turid Rugaas and Anne Lill Kvam would disagree with Susan’s suggestion to relax about controlling the environment, especially when it comes to puppies. For example, Anne Lill suggests in her trainer education that the first one or two weeks your puppy is with you, she should not go out to meet other dogs or people, but be given time to adapt to his new home. In Austria, most puppies move into their new homes when they are 8 or 9 weeks old. This would mean that they not leave your house and garden until they are 10 or 11 weeks old!
Other renowned trainers such as Ian Dunbar suggest a radically different approach: “Your puppy must socialize with at least 100 different people before she is three months old. That’s just twenty-five people a week, or four a day.” (Dunbar, Before & After, p. 207) etc.
Personally, I side with Dunbar’s and Susan’s approach, because science shows that resilience grows with stimulus diversity. And I believe that even the most versatile environmental enrichment doesn’t provide as much stimulus diversity as meeting and interacting with unfamiliar people and objects “out in the real world” (in addition to the own house and garden) does. However, if Turid Rugaas, Anne Lill Kvam can point me to a study and can show me data that proves that her approach is more successful in building resilience, I will be happy to change my mind.
One could argue that it is unncessary to build up the maximum amount of resilience that could be reached by lots of stimulus diversity in the puppy’s home and out in the world: one might suggest that as long as the animal’s environment is controlled by the handler all its life, the animal would never need the strong resilience it didn’t build up as a puppy. I assume that you could keep a dog this way for all his life – you could make to always control his environment, i.e. keep it free from even mildly aversive stimuli like honking cars etc. (Which will in itself be reinforcing to you, because control is a primary reinforcer.)
However, for me, this argument is not a practical one. Personally, I believe that every animal (human as well as non-human) should “be able to enjoy as much freedom as she can handle,” as Leonard Cecil nicely put it a while ago (Facebook status, June 17, 2014). “Freedom” includes being off-leash on walks – on walks in the city and on walks in the forest. It also includes the possibility to accompany me when I travel, or go to work. It means being okay on subways, and being okay in restaurants and other busy places, and being okay at friends’ places, and in hotel rooms. (I am not saying I am forcing my animals to join me – rather, I want to give them the option to join me sometimes, and not other times, when they would rather stay home. And yes, we communicate: sometimes, they will follow me to the door with wagging tails, or lead the way and motivate me to venture out with them when I hadn’t even planned it. Other times, they will stay on the couch instead. And that’s a choice I respect.) It means it’s okay for them to be dogsat by friends, even if this means being put in unfamiliar situations and meeting new people. It means being okay in shopping centers, being okay in busy dog parks, and being okay when meeting other dogs or other animals. In order for all these things to be okay, my dogs need resilience. And if I get a puppy, I will always try to build resilience to give her the greatest possible freedom to join me on my ventures, and to be off leash in as many places as possible. I’ve tried to give Phoebe lots of stimulus diversity as a puppy, but I would try to give my next puppy even more.What do others think? Do you agree or disagree with me? I would be interested in your opinions on building resilience, and on the greatest possible freedom your dog can handle. Let me know in the comments!
Contrafreeloading refers to the obervation that when given the choice between “eating for free” and “working for their food” (i.e. using their behavior effectively), animals tend to choose the food that requires effort over the free food. This is why clicker training is so much fun!
This theory is not new to me. However, Susan presented it in such an enticing way that I was compelled to try a little contrafreeloading experiment myself.
My Standard Poodle Phoebe knows puzzle feeders well, and as far as I can tell, she enjoys them (yes, I know that “enjoy” is a label rather than a behavior description. But let’s say, that’s okay for now). However, I have never offered Phoebe a choice between food in a bowl and food in a puzzle feeder – I usually serve her either the one or the other.
In my experiment, both the food bowl and the puzzle feeder hold the same amount of the same food. I expected Phoebe to choose the puzzle feeder – however, surprisingly to me, she chose the free food first and only started on the puzzle feeder once she had finished the food in the bowl. I repeated the experiment the next day in a different room and with the puzzle feeder on the left and the food bowl (a different food bowl) on the right, but once again, she finished the free food first and engaged in the puzzle feeder afterwards.
I repeated the same experiment with my foster Galga Luz. She has has only been with me for a short while, and has only recently been introduced to puzzle feeders. While she seems to enjoy them a lot, I expected her to choose the free food over the puzzle feeder, since she has little experience with it and tends to eat very fast. This is exactly what she did: she finished the food in the bowl first and engaged in the Wobbler afterwards.
It’s interesting that my dogs chose to eat for free! I’ll be sure to think about this some more and devise a follow-up experiment!
– An ethical guideline: least intrusive, effective alternative.
Why? Because effectiveness alone is not enough. And because science agrees with the least intrusive, effective alternative (cf. slides part 1, p. 42).
– Which strategy do I employ in order to change an animal’s behavior?
A hierarchie of interventions, from least intrusive and most socially acceptable (1) to most intrusive and least socially acceptable (6):
(slides part 1, p. 47)
The way I understand it, this graphic reflects the continuum from least to most intrusive for the average animal. However, as Susan repeatedly pointed out, it is important to remember that it is always the study of one: we are not dealing with an entire species, we are not dealing with the average dog. We are dealing with the individual, with the animal in front of us. So while she suggests to always start attempting to revise behavior with the least intrusive approach and then, if it doesn’t work, move down the pyramid, the order of the hierarchy may be slightly different for any individual animal.
So we [people bound by the “least intrusive, effective” ethics] start at the least intrusive level and move down the hierarchy if the (correctly applied) approaches don’t work. Does that mean if level 1, 2 and 3 fail to work, we will move into more intrusive territory? Are there instances where, if we apply the “least intrusive, effective alternative” ethics, we will use extinction/negative reinforcement or even positive punishment? The simple answer is: yes. However, first of all we will review our less intrusive approach carefully, and get feedback from others. Only if and when we are sure that nothing else works and that we applied the less intrusive approaches expertly will we consider using positive punishment. Over the course of our training experience, this will happen extremely rarely or not at all, since most problem behaviors can effectively be changed with level 1, 2 or 3. Positive reinforcement is an extremely powerful tool, and almost everything can be solved with it. For example:
– Why I agree with the “least intrusive, effective alternative” ethics
For me personally, the “least intrusive, most effective” ethics resonates more with me than the “I will ONLY use R+” credo of some colleagues. A “least intrusive, effective approach” allows for full use of all the tools in the tool box, if necessary. Therefore, this approach is able to change a bigger range of problem behaviors than the “I will ONLY use R+” approach.
Another important message to take home from Susan is that there is no obvious relationship between extremeness of a problem behavior and the strategy that succeeds in changing it. A more severe problem behavior does not require a more intrusive approach (which is basically what Cesar Milan wrongly claims as justification for his use of P+). Very severe problem behaviors can be changed by means of R+! However, the more severe the problem behavior, the more in-depth should be its assessment.
– Punishment is a bitch!
If P+ isn’t working, it hasn’t been punishment. Punishment is the mirror image of reinforcement: just as, by definition, a consequence is only reinforcement if it increases the frequency of a behavior, it is only punishment if it decreases the frequency of a behavior. If you think you have been positively punishing your child/spouse/dog/parrot etc., but their behavior hasn’t become less frequent, you have not really been punishing them. Simply because a consequence produces escape behavior (is perceived as uncomfortable) does not make it punishment. It is only called punishment if it weakens the behavior that came before. If it doesn’t weaken the behavior, but still produces escape behavior (e.g. taking a step back), what will happen is adaption: the animal gets used to the stimulus and will slowly tolerate a higher and higher stimulus intensity.
Susan described an experiment where rats were put in a cage with a feeder. Whenever they fed, they received a mild shock. The shock was uncomfortable engouh to produce escape behavior (the rats jumped back), but not strong enough to weaken the behavior itself (they still returned for more food). Hence, it was, by the scientific definition, not punishment.
The intensity of the shock was slowly increased, and eventually, the rats were able to tolerate a rather strong shock.
Another group of rats was placed into a cage and immediately shocked with the strong shock the very first time they went to the feeder. Without gradual increase, there was no opportunity for the rats to adapt to the intensity of the shock. This was punishment: it weakened the behavior. The rats did not go back to the feeder; they would starve to death rather than getting more food.
This is something anyone working with P+ has to bear in mind. And this is the reason why parents who start talking loudly when unhappy with their child will often end up screaming at their children sooner or later – all to no avail, since the children have adapted.
As Susan says she likes to point out to her graduate students: her use of positive reinforcement is a choice, not a weakness. If she wants to, she can punish with the best of them. It’s useful to bear that in mind yourself: if you ever want to really punish your spouse, for example, do it right and prevent adaption! 😉
– Motivating (establishing) motivations:
We’re in operant territory here. We can “change the frequency of a behavior by temporarily altering the effectiveness of the consequence”. (slides part 2, p. 8). It’s ALWAYS possible to change the strength of a reinforcer.
Yey! This is great news! Even when a reinforcer is extremely strong – say, intrinsically motivating, such as getting to chase someone or something – we can temporarily weaken its strength by means of letting the animal have a good run before training. We can temporarily change the strength of a reinforcer by moving the animal forther right or left (depending on whether we want to decrease or increase the strength of the reinforcer) on the continuum of
hunger ………………………………………… satiation
fatigue ………………………………………… rested
activity deficit ……………………………… activity excess
For example, for me, chocolate is a strong reinforcer. However, when I’m satiated, it’s considerably less strong. On the other hand, a rare treat like Lindt or Zotter chocolate is even more reinforcing for me than an everyday treat like Milka.
Also, finally, I’ve got a scientific source for the fact that small and frequent reinforcers build behavior more quickly than large, occasional ones! Susan mentioned Schneider (1973), and Todorov, Hanna, & Bittercourt de Sa’ (1973). Will have to look into these in more detail.
– Cues (discriminative stimuli)
Give stronger reinforcers rather than bigger/louder cues! Cues can be very small – they need to be clear, but not big. The stronger the reinforcer, the stronger the cue will become. “To get positive reinforcers, learners often enthusiastically exceed the minimum effort necessary to gain them.” (slides part 2, p. 14) For R-, on the other hand, animals will often only do the minimum effort necessary for escape.
– Antecedent arrangement to make behavior more likely
I loved Susan’s example: Watch the video “The piano staricase” on http://www.thefuntheory.com. The antecedents (the stairs) are tunred into a giant piano to make it more likely that people will take the stairs rather than the escalator. (The other videos on this website are also lots of fun to watch, by the way!) Antecedent arrangements are quite common in everyday life, actually. At least in my life. Usually, when I want to seduce or entice someone, I’ll carefully arrange my apartment in a way that might make it more likely that they fall for me. I’ll select certain kinds of music, clean my apartment, maybe get my visitor’s favorite snack – I might casually mention that I just happen to have made Guglhupf the other night -, and I might even move certain books or DVDs to a place where they can be easily seen by my visitor, and hide others.
– Non-contingent reinforcement
if you freely give the reinforcer the animal used to get by means of crying/biting etc., the frequency of the problem behavior will decrease. (3) I found a nice outline of NCR online that shows how NCR can be used to change attention-seeking problem behaviors in children.
– The cultural fog, or: a philosophy of teaching
Susan shared her philosophy of teaching with us, and I am very grateful. She has found perfect metaphors for concepts I’ve been struggling to get a grip on. Listening to her, it was obvious that for Susan – unlike for a surprisingly high number of “positive” trainers! – ethics doesn’t only apply to non-human animals. It is equally important to her to treat her human clients with the utmost respect.
She reminded us that she finds it helpful to think of clients in the following way: “The client is me, just with less information.” This is not only a useful way of looking at it if we want to build healthy and productive relationships with our clients, it is also true. Clients aren’t “stupid” or “ignorant”; they really just are us with less information.
Simply burying them under an avalanche of new information, however, is not the way to go. If we want to convince someone of the effectiveness of R+ and the unnecessariness of P+, we are not only asking her to change the way she treats her dog, but the way she sees the world. Approach your clients gently, because you are asking a lot! Keeping this in mind keeps us compassionate teachers.
What makes it so hard to accept new information is what Susan calls cultural fog. Cultural fog refers to the conventional wisdom that can become an obstacle to learning. “… despite new information to the contrary, conventional wisdom has a property analoguous to inertia that opposes the introduction of contrary belief.” (slides part 1, p. 2)
Cesar Milan (who we should be thankful to for being such a convenient negative example to so many things) uses cultural knowledge rather than scientific knowledge: he operates within the cultural fog, which is why his followship is so big. He caters to beliefs already held within the cultural fog and refines them.
No one is immune to the cultural fog. While I might be “out of the cultural fog” when it comes to canine behavior modification, I am sure there are other areas of life where I’m comftably holding my position in the cultural fog without even being aware of it. So it’s not okay to “blame” others for not being able to see clearly: it’s hard to see through the fog. Just try driving over the Wechsel in November, and you will know what I mean.
I enjoyed Susan’s cultural fog metaphor not only because it applies to clients with dogs, but also because it applies to the human students in the adult German classes I teach for immigrants in Austria. Feminist that I am, I regularly discuss gay equality, adoption rights etc. in class. I usually come out as a gay woman when there is a homophobic comment, when someone criticizes the pride parade or complains about the life ball poster …
My wish is to lead them out of the cultural fog. I ask them whether it’s okay if we discuss this topic a little further, and if they agree, I tell them stories. Personal stories: stories about falling in love with a woman. Political stories: stories about the 40 differences that still exist between domestic partnerships and marriages in Austria. Stories about countries where being gay gets you sentenced to death. I try not to preach, but just share. I don’t take it personally if they disagree, and respect their opinions, their cultural background: they are just me with less information.
I encourage them to ask questions; political ones, life-style related ones, and personal ones. I ask questions myself: I ask who can think of reasons why a pride parade may be considered necessary or useful by the gay community. I ask them what we (the gay community) might want to archieve with public visibility. I want them to come to the realization that not having to care what others think is a privilege rather than a choice. I try not to tell them things, but to listen to their opinions, validate them, and challenge them in ways that let them practice their German – and maybe, just maybe, take a tiny little step out of the cultural fog. Maybe now that I have Susan’s metaphor, I can lead them out of the fog even more effectively? We’ll see! I’m already looking forward to the next time the topic crops up in class.
As for clients with dogs, that’s a piece of cake for me. If you deal well with something as personal as gay rights in the cultural fog, it’s easy to deal with dog training clients in the cultural fog. If you happen to be gay and a trainer, I encourage you to come out to conservative strangers more often – it’ll make your encounters with clients just seem so much easier.
– Don’t underestimate your clients!
I also liked that Susan encourages us to not underestimate our clients. I completely agree: Of course, our clients are smart enough to undestand and implement behavior modification programs if we just explain them well. Even (or especially) children can be great teachers for their animals if we just manage to be decent teachers for them. (Compare the fantastic video with Noah’s parrot below!) We are no more intelligent than our clients; it’s just that our respective expertise lies in different areas.
I believe that, if a client or student doesn’t manage to do well, it’s because we have failed them as a teacher, not because they have failed us as a student. It may be easier to complain about our clients than to take a critical look at our teaching approach, but the latter is just so much more effective – and so much more rewarding! After all, happy clients make for positive feedback, and positive feedback is a strong reinforcer for us! (This is why I love teaching: I’m addicted to the positive feedback! And if there sometimes is critical feedback – and there is – well, I love a good challenge as well. I will work even harder on my relationship with that particular student and try to implement his wishes in the classroom.)
– A-*B-C assessments and case studies
I particularly enjoyed the last part of day 3, when we employed what we had learned and did A(ntecedent) – *B(ehavior) – C(consequence) analyses ourselves. First we practiced with videos with different animals, then we paired up and did an A-B-C assesment and found ideas how to modify antecedents and consequences in order to modify behavior for a case one of us dealt with. Christine Schragel and I worked on Nala’s case, one of the dogs she works with at Wiener Tierschutzhaus. Trying to describe Nala’s behavior without using labels, and identifying the antecedents and consequences to her problem behavior was quite challenging! We came up with a number of ideas Christine will try to implement, and I’m looking forward to hearing about Nala’s progress! I keep my fingers crossed for her!
When doing an A-*B-C assessment, Susan suggests we always start with the *B, and then fill in the respective A and C. We would first do A-B-C for the problem behavior. Then, we would think about what we want the animal to do instead of the problem behavior, and how we could arrange antecedents and consequences to make this replacement behavior more likely to occur than the problem behavior. However, in addition, we would also preserve the reinforcer for the problem behavior (the original C): this is a reinforcer valued by the animal, and Susan suggests we always also come up with an acceptable alternative way for the animal to earn the reinforcer that was originally gained by the problem behavior. It seems a very good idea to preserve all reinforcers!
… and of course, there was just SO MUCH MORE I could talk about. But then I’d never finish this post. So even though it’s hard to stop, I leave you with a final thought: if you get the opportunity to attend one of Susan’s seminars, do so. It’s a wonderful opportunity, it’s a lot of fun, and even if you have worked with animals for a long time, there’s still going to be lots you can learn from Susan.
Wheee … and that’s all of us posing with our “certificates of excellence” on day 3:
(1) Examples for control as a primary reinforcer, as cited by Susan: Watson, 1967, 1971 (human babies); Joffe, Rawson, & Muliak, 1973 (rats); Mineka & Henderson, 1985 (Rhesus macaques).
(2) Susan mentions 5 sources for the growing consensus on this approach among experts: Fernandez (2001), Martin (2011), Ramirez (1999), Bailey (personal communication with Susan, 2011), Pryor (personal communication with Susan, 2011).