Did you know that Krista Macpherson, a researcher from the University of Western Ontario, conducts canine cognition research at When Hounds Fly? Many of her test subjects are current and past students! Here is an update she sent out to past test participants, including a deck she presented at a conference with interim findings back in March.
Want to volunteer your dog and your time? Please email firstname.lastname@example.org and mention if you are (or aren’t) a student (current or past) at school. She is always looking for volunteers to help.
First of all, my apologies for not following up with you all sooner…I keep telling myself “I’ll talk about results once I have collected a bit more data”…but of course I’m always collecting a bit more data, so I might as well talk about results now!
Most of the people on this email list have participated in one (or both) of two experiments…the first was a timing experiment, and the second was a counting experiment…I’ll speak to the progress of each study below:
In this experiment, we had one or two “Manners Minder” machines spit out a a treat on a set interval (either every 30 seconds, or every 1 minute). With this study, we were interested in determining whether over time, the dogs would come to anticipate the arrival of food from the machines, and thus approach the machine as the time interval approached. This ended up being problematic, because most of the dogs easily figured out that the ideal solution was simply to lay beside the machine the entire time and wait for the snacks! While this was a good strategy on the dogs’ part, it made it impossible to measure timing in terms of proximity to the machine.
Luckily, the psychology department at Western has a great team of woodworkers and electricians to build experimental materials…and they have revamped the Manners Minder machine for me. It is now bolted to a plank of wood, and has a large button hooked up to it. So the dogs will still be able to get the food at a set interval…they’ll just have to hit a button first. This means that instead of measuring timing in terms of the dogs’ proximity to the machine, we can measure timing in terms of the number of button presses that occur as the interval approaches. I’m going to start pilot testing this new apparatus in the next couple of weeks!
With the counting experiment, dogs watched as the experimenter (me) dropped different quantities of food into each of two bowls. If dogs can count, then they should reliably choose the bowl with the larger quantity of food. What we found is that when the food ratio involves “food versus no food” (which in this experiment, was done using 3 versus 0 pieces of food , and 1 versus 0 pieces of food) the dogs are extremely good at the task (in fact, they rarely choose the wrong bowl. In any other combination (e.g. 1vs3, 4vs1, 2vs3, 3vs4, 2vs1) the dogs really struggle with this task, and to get right answer only about 60-65% of the time (and since there are only two bowls, chance performance is 50% on this task).
It is interesting that dogs struggle on the counting task, because pretty much any other species tested on this task has performed quite well at it. A couple reasons that they may not be good at this task are as follows:
1. As we have selectively bred dogs to attend to us (herd our sheep, guard our homes etc.), we might have “bred out” other characteristics that were more important for survival in the wild, like numerical discrimination.
2. Dogs are dumb (I personally am not a fan of this explanation!).
Another possibility is that this procedure, for whatever reason, was not “meaningful” to the dog. In talking to another researcher who was working on a similar project, is occurred to me that I have always had the experimenter drop the food into the bowl, while the owner held the dog (but did not otherwise interact with the dog). What if the owner was the one to drop the food into the bowls? Would the dog pay more attention to the task?
Long story short, I changed the procedure so that the owner was the one dropping the food into the bowls–and preliminary results show that in many cases this made a big difference to the dogs. We need to test this more formally, but these initial findings are really exciting!
I presented the counting data (minus the new “owner dropping the food” version of the experiment) at a conference in Florida in March, and it was very well received. I have attached a copy of the slides that I presented, for any of you who may be interested.
I am currently looking for some more dogs to do more counting studies with–I need naive dogs, who have never done this task before. If you have a dog or know someone who has a dog who would like to participate, please let me know! I will be running sessions at When Hounds Fly on April 30th/May 1st, as well as May 14th/15th, between 2pm and 7pm. It takes about an hour to complete the experiment.
As always, many thanks for all of your help with these projects!
Want to see the slide presentation? Click on any of the images below:
Session 1: Efficient Training in Action (Cecilie Køste)
This lab really focused on smart reinforcement placement. BUT, another key aspect of good training is everything that happens before you start training.
Her mantra is THINK – PLAN – DO. Thinking is “What is the behavior to be trained?” Plan is “What is the training plan, what is my criteria, where will reinforcement be delivered, what could go wrong, and what will I do?” When you are training… no thinking, no planning. That must occur before training begins. If the training session is falling apart, you must stop, put your dog at station, and go back to thinking and planning.
The exercises in this lab were:
Cup Game (Clicker Mechanics)
She had all the students warm up with a basic clicker trainer exercise – the cup game. How many treats can you put into a styrofoam cup while maintaining perfect form (feeding hand is still till click, no extra movements of the body, no speaking). I haven’t done this myself since my first KPA workshop over a year and a half ago.
Go to Target
This is the simple “Go to Cone” game we even do at Puppy Class at When Hounds Fly. The best way to use reinforcement here is to click, deliver the treat to the dog’s mouth, let them nibble it while you lure their body/head back towards the cone. When the dog has finished eating and is released, the cone is now right in their face for another repetition. Brilliant!
All the dogs here had finish positions already trained, ranging from kind of wide and sloppy requiring a big hand cue, to verbal only finish cues, to really snazzy jump, rotate mid air, and land in finish Obedience finishes. Efficient training means when you click and treat, let the dog nibble on the treat while the handler moves around and faces the dog in Front position, tucked really, really close (like Front). When the dog is finished nibbling, the dog is now ready to offer Finish again. Round and around you go…
One tip that was mentioned about any of these obedience behaviors is the dog should get it right and perfect on the first repetition – otherwise you have a dog that learns to do a sloppy finish, then wiggle their butt in to get closer, c/t. Oops… Petey has that. Michele Pouliot calls this a “two part finish”.
Long Down – Feed in position (room service) – bring the treat to the dog to reinforce position.
Sit – Feed above head for a more tucked sit.
Stand – Feed towards chest
Reverse Lure – While a dog is in position, tease the dog with food in your hand, almost like you are trying to lure them into another position. If the dog maintains position, c/t.
“Many Downs” – How many downs can you get in a minute? C/T and toss the food to reset quickly, or after a down, handler moves away to get the dog up, so you can cue another down.
Rollover – Feed over the shoulder to encourage more rolling.
Hold Cloth – Shaping a hold of a hotel towel.
A common theme throughout the sessions I have taken – everything you do in training should be a conscious decision, and certainly I have learned to be very conscious about where I reinforce and why.
Session 2: Crosstrain! (Michele Pouliot)
This seminar should just have been renamed “Platform Training”. Platform training is relatively new. Michele had a DVD for sale at ClickerExpo that was flying off the shelves and kept on teasing me since it was played on a monitor in the hallway.
Platforms are so new in training that there are next to no videos on it on YouTube. I just found a handful… this one is the use of platforms to teach position/distance for Obedience:
The use of props in training is not new. We’ve used walls/edges to limit a dog’s option when trying to train a straight heel or a straight finish. We’ve used mats for position. Many are now teaching heel/finish using perches (or paint cans in my case for Petey). Platforms make it easy for dogs to delineate space. It makes it easy for us to understand criteria, since if the dog is on the platform, and it’s the right size, by definition, the dog is straight, and in the right position. Michele showed clips from her new video that shows platforms being used to train front, finish, weaves, go outs, tight spins, stay, and response from a distance work (like paw from a distance, while perched on a platform).
I intend to teach the 18 basic behaviors from Cecilie’s Top OTCh lecture on Day 1 to Petey again, but this time using platforms. I’m going to build my own platforms using leftover flooring from When Hounds Fly.
Lastly, she talked about other useful tools for training such as chutes, ledges, walls, x-pens, etc. to train behaviors. I was delighted to see all this, because Julie is doing the same stuff with our students in our tricks class right now. I love it whenever the methods we use at When Hounds Fly are validated by the world’s top clicker trainers. 🙂
Also, check out her “Step Up to Platform Training” DVD here. She starts training her litter of puppies at 4.5 weeks old. Wow!
Kay Laurence is “one of the world’s top clicker trainers” according to Karen Pryor. All weekend long, people were raving about her lectures so I decided to catch her on the very last track and session for the conference. I am glad I did! This topic was appropriately “light” in technicality but brilliant in terms of expanding my horizons of training. Specifically, Kay has very ingenious ways to incorporate playing with your dog as training.
Through play, we can teach dogs self control, body awareness, and movements such as backing up, side stepping, etc., all of which are needed for many different dog sports.
One important thing to think about (which I often don’t) is the safety of the game. She is adamant that dogs should never run around on laminate flooring, as it’s an easy way for them to pull a muscle, or worse. I guess it’s true -try running around laminate flooring in your socks and see how long it takes to pull a muscle. Her training facility is CARPETED and ripped up and replaced regularly.
She has a pretty large library of videos on YouTube so I’ll just share these videos for you to watch and enjoy. Have fun playing with your dogs!!!
Games with a Sausage (Catch the Mouse)
Lastly, I thought I would share a few things from the Saturday (Day 2) evening dinner featuring Patricia McConnell. She lectured on the topic of animal cognition and how much animals think. I didn’t really take notes but she shared various research on different examples of studies done and also cited Ken Ramirez’s (Shedd Aquarium) work on teaching dogs to understand concepts like Big vs. Small, Mimicry, and of course dogs like Rico and Chaser (object recognition with both noun and verb). She showed some footage that I’ve found on YouTube for you to check out. They’re interesting and fun.
She also shared a few other recent studies about dogs understanding our pointing gestures better than chimpanzees, and also a study that showed domestication does not necessarily mean a dog’s problem solving ability is reduced (which is commonly believed to be true). Anyways I’m not really a scientist so I’ll just ask Krista (from UWO) next time I see her.
Julie, Mirkka, Emily and I had to leave at 4 to catch our flight home, so we missed the closing remarks by Karen and Patricia. Evidentally it was so moving there were 400 dog trainers in tears at the end. Oh well, next year I’ll plan to stick around! In the meantime, I have hundreds of slides and notes to review. Hope you enjoyed the blog series!
I’m exhausted! I feel like I’m back in school… also feeling a bit inadequate, but also really fired up about enhancing my own training skills as soon as I get back home. Cecilie said during her lecture today “You could shape that way forever and be perfectly fine, but doing it this way will make your training more efficient.” Who doesn’t want to be more efficient? So here’s a summary of what I did and what I learned.
Session 1: Shaping Procedures for the Agility Trainer (Eva Bertilsson and Emelie Johnson Vegh)
I’m not an agility trainer and I don’t know the first thing about agility. This is exactly why I took their track.
The most important concept they shared today is the concept that shaping is not just Behavior, Click, Reward, repeat. There is a TON of behavior that occurs between the taking of the reward and the next cue. What happens here MATTERS. If you don’t pay attention to what occurs after the dog eats their treat and the next behavior, you end up building a lot of “garbage behaviors” in your training.
A perfect example I have from my own training is with Petey. He often throws in extra behaviors between reps – a common one I see is he spins clockwise between behaviors. Another common example I see with many dogs is during a Watch Me exercise – after the dog eats their treat, they often start scanning the room and looking at other dogs.
In both cases, the eating of the treat has become a cue to do an extra behavior. If we are trying to build focus in the Watch Me, and you continue to allow your dog to eat, then scan, then cue Watch me, then click, eat, and scan, you are actually reinforcing scanning/looking away.
Good training eliminates the garbage that happens between repetitions. You can manipulate the environment to prevent those extra behaviors, change criteria/increase rate of reinforcement, or work on reinforcement delivery (position) to prevent garbage behaviors from occuring.
Garbage behaviors are also contextual. In obedience, a great behavior to get after eating a treat is for the dog to look back at the handler, since we are trying to build extreme focus on the handler. But in agility, we are trying to have the dog continue along the intended path of the course (criteria would be nose ahead towards line).
Session 2: Shaping Procedures for the Agility Trainer in Action
This was my first Learning Lab for ClickerExpo. Since I am without a dog, I attended as an observer.
The primary exercises for the dogs in this lab was to implement the “Aim for It” procedure – which in a nutshell, is nose ahead towards intended path. These exercises were fantastic, however, what really blew my mind was the exercises that preceeded Aim for It.
Eva and Emelie actually spent a good thirty minutes simply working on the protocols that lead UP to beginning training. This is what happens before you start training with your dog. It is clear that what happens in between training sessions is key to making each training session impactful, an also getting maximum intensity and the right attitude out of the dog.
Training starts with a dog at their “Station”. The station can be a mat or a crate. When a dog is at their station, the dog is off-duty and the handler is free to think, plan, and prepare. Once the trainer is ready, the dog is moved from their station to their work area through a “Transport”.
Transport, as they define it, is not just walking the dog casually on leash (or off) to their working area. When transporting a dog, they assert that the handler must be fully engaged and contacting their dog at all times. That can mean dog is being lured/lead with food from your hand (and making full contact/munching away), lead with a collar grab, or lead via a tug toy.
As soon as the trainer disengages with your dog (food hand is taken away, hand goes off collar, or toy is taken away), training starts, and you c/t the first behavior that meets criteria (for Aim for It, it’s nose ahead). Once the trainer is done the session, he immediately go back to transporting (by using your last food reward as a lure back to the station, or tugging back, or collar grabbing and leading back).
This protocol takes “pay attention to your dog” to a whole new level for me and I feel quite guilty for leaving my dogs “dangling” after the end of a training session now.
Session 3: Efficient Training – Making Progress Quickly (Cecilie Køste)
This session had a lot of parallels to Eva and Emelie’s. A lot of it covered basic good training form (no talking while training, hands by the side (not in the bait bag), etc. She also spent a lot of time sharing videos of her colleagues training, and they do also have very regimented stations – the dog was sent to his crate whenever the trainer needed a break, or time to setup the training environment (moving props, reloading treats, packing tug toys, etc.)
The biggest takeaway I got from her presentation was placement of reward. As I mentioned earlier, up to this point I have been pretty relaxed about treat placement. Primarily, I used treat placement as a way to reset the dog to maximize the # of behavior repetitions in a training session (i.e. click for watch, toss treat for to floor to reset, or click for down, toss treat to side to reset, or click for mat, toss treat away so the dog has to travel back to the mat). She summarized the four different strategies for treat placement as follows:
Reinforce in position (treat in down for down stay – room service as she calls it)
Reinforce to reset position (to expedite the next repetition)
Reinforce for Direction Sliding (treat ahead of the dog away from you in heel to counter-act the tendency for a dog to become banana shaped in heeling)
Reinforce to next behavior in behavior chain (toss toy over jump, if final behavior in chain is a jump
Session 4: Click to Calm Unleashed with Emma Parsons
Last lab of the day – I was an observer in Emma Parsons’ Click to Calm Unleashed lab. She ran the lab walking through the different exercises she teaches in her Click to Calm class for reactive dogs. What’s amazing about them is they don’t actually require a ton of space, and one thing that was really surprising is how hard she pushes her students… making trainers with their space sensitive dogs work in extremely close proximity by being strategic and careful about entrances, exits, and maintaining high rate of reinforcements (or sometimes just shoveling food into the dog’s mouth). With more research and planning hopefully I can incorporate some of these exercises for classes at When Hounds Fly.
The evening ended off with dinner with a lecture by Patricia McConnell – I’ll write more about that tomorrow.
Just on break between my last session and dinner so I thought I’d blog a bit about Day 1 lectures and workshops!
Session 1: Ken Ramirez, Aggression: Treatment and Context
This presentation categorized and summarized many of the aggression treatment protocols commonly in use in science-based training. Classical Counter-Conditioning, Constructional Aggression Treatment, Incompatible Behaviors (Watch me), Look at That, Click to Calm, etc. were all covered quickly, and Ken presented his opinions, pros, and cons of each. Most of the content of this presentation was review for me (since dog aggression is an area I spend a lot of time studying and working on in the field), however, here are a few interesting points I picked up:
Ramirez asserts that training Incompatible Behaviors (Watch-Me) is an excellent tool to prevent the rehearsal of aggression and to keep animals safe. However, he clearly states that these approaches do not solve the aggression problem by itself. This is something I figured out by accident, which is why I incorporate Behavior Adjustment Training now in the treatment of dog reactivity.
Ramirez showed a video of Kellie Snider in a CAT exercise with a Doberman that is aggressive to strangers that visit their home and has a bite history. The video showed clips of a 38 minute protocol where initially, the dog would bark and lunge at the sight of Kellie (over threshold), but only at the precise moment the Doberman relaxed would she leave. If the dog barked as she was leaving, she would return. After 50 repetitions (roughly), she was able to greet the dog and feed the dog treats, and shortly thereafter she was able to pet the dog on the head. The video itself was quite amazing in that a) she is incredibly brave – the dog was unmuzzled, has a bite history, and was only restrained on leash and being held by a pre-teen boy and b) this dog had never allowed a stranger to come into their home in many years.
Ramirez did not discuss BAT (Behavior Adjustment Training) as he is not familiar enough with it at this time. A bit disappointing, since it is my favorite protocol for reactivity now.
Session 2: Michele Pouliot – Anticipation is Making Me Great!
Michele is a world champion canine freestyle competitor – which is exactly why I took this track (I know very little about Canine Freestyle). The topic of this presentation was how to use a dog’s anticipation to your advantage as a training tool.
If you have ever had a dog anticipate your cue and break a start line stay, or impatiently begin offering behaviors prior to the cue, here are some takeaways (Petey is a chronic jump-the-gun dog, and I’m delighted that she frames this “problem” into an advantage)
Anticipation when harnessed creates energetic and immediate (low latency) behavior
Most errors of dogs starting before cued are handler errors – extraneous cues the handler gives unconsciously – video taping yourself is the best way to troubleshoot this
Never correct (even mildly) a dog that anticipates and starts before the cue – you will kill their enthusiasm
Create a “Start-Ready” dog by creating a cue that tells the dog “the first cue is coming” so they know to get energized, focus, and wait for the first “real” cue
Use a “wait” cue and reinforce it either with a primary reinforcer (c/t) or cueing the actual behavior.
Session 3: Cecilie Køste – Top OTCh Skills for Top Obedience
Cecilie is world class competitive obedience trainer from Norway. This is another area I really don’t know much about, so I was thrilled to have a chance to hear her present in person (tomorrow I am attending a hands on workshop with her on the same topic).
A key concept in their training method is to teach a set of 18 “basic skills” that have no cues, then assemble them finished behaviors fluent with very high criteria, then backchain them into performance behaviors, and then assemble routines through backchaining. The analogy she used is “Letters to Words to Sentences”.
Since I’m working on getting Petey ready for Rally-O Trials, I’ll blog more about this separately but here are a few neat points I wanted to share tonight:
They are BIG on Doggy-Zen proofing. All behaviors in the basic set must be performed with food scattered on the floor or in bowls – once the dog has performed the behavior, a release cue is given to go to the food (or toy)
A new term… “reverse luring” – where lures are presented and the dog must ignore the lure and maintain position (a variation of doggy-zen)
No cues are given to basic behaviors – so each session is compartmentalized to focus on one of the 18 basic behaviors. How does the dog know which behavior is being worked on? The dog offers all 18 and you just c/t the one you are working on in a session.
Two videos that were quite impressive was a dumbell hold while hot dogs were placed on the dumbell, and a dumbell retrieve where the dumbell is thrown right next to a bowl of food (zen to the nth degree!)
Time for dinner! Stay tuned for Day Two tomorrow. Comment below if you have questions!
On February 20th, When Hounds Fly hosted a party for a litter of 7 week old Icelandic Sheepdogs from Sulhundur Icelandic Sheepdogs. Besides being a ton of fun for all the guests, this was an incredibly efficient way to have these young pups meet over a dozen people and experience a ton of new things! Guests included the future families of the puppies, owners of pups from past litters, and a few friends and colleagues of ours.
Remember, puppy socialization is the single most important thing you can do, and you don’t have much time to do it. Socialize and train early! (And check out the videos and photos below).
There is a little daschund in our neighbourhood that I see periodically on our walks. The little daschund must weigh all of 8 pounds soaking wet. Her owners seem to really care about her because she gets walked regularly, even in this cold weather. I also believe this little daschund was probably a rescue, because it’s been the same size since I first saw her. Unfortunately I have never talked to their owners because this little daschund is leash reactive (she barks, lunges, and acts aggressively when dogs get close by).
A short while after the daschund arrived in the neighbourhood, I noticed that they began walking the dog on a pinch collar (also known as a prong collar). For those of you who (fortunately) do not know what a pinch collar is, here is a picture of one:
A pinch collar is a tool of punishment. If the dog pulls, the metal ends jab into the dog’s neck causing pain, which causes pulling to stop. If a leash correction (leash tug) is applied firmly, the metal ends jab into the dog’s neck also causing pain. Most dogs will walk gingerly and carefully when you put a pinch collar on them to avoid pain, so they are often used by dog owners who are too impatient to teach loose leash walking.
The owners of the daschund, while on walks, would apply a leash correction every time the dog barked, lunged, or reacted to nearby dogs. Very quickly (within a couple of short weeks), the daschund’s outbursts stopped. She could walk right by me while I was walking Petey and it was as if the little daschund didn’t even see him. On the surface, I’m sure the owners of the daschund were delighted with the results.
A few months after that, I would see the daschund being walked in the neighbourhood, and thankfully, the owners had switched to using a harness. We’d run into each other while walking our dogs; I did my best to give the little girl enough space, and despite the removal of the pinch collar, she still was able to walk by without outburst. I remember being impressed that she was still suppressing her outbursts despite the absence of the pinch collar (this was summer). But I knew all was not fixed with that dog. When I see quiet and still dogs that are reactive or fearful, I think back to this German proverb:
“The silent dog is the first to bite.”
Fast forward now to January. Petey and I were out for our afternoon walk and we see the little daschund again. I actually didn’t recognize her at first since she was all bootied and coated up. As Petey and I walk by, she started barking, lunging, and pulling towards us. Because the threat of punishment is now a long gone memory, her old behaviors have returned.
The moral of the story is – the heart rules the head. Fearful, aggressive, and reactive behaviors are rarely driven by conscious decisions – they are driven by emotions. When behaviors are driven by emotion, the only way to change the behavior is to change what’s in the dog’s heart. The pinch collar never helped the little daschund learn to be confident – in fact, she looked quite depressed while she wore it. It never helped her learn to feel comfortable around dogs – in fact, the sign of a dog coming meant the risk of leash correction was imminent. All the dog learned to do was stay still and bottle her feelings up.
I can understand the allure of a “quick fix”. I wish I could help my clients with reactive dogs walk by other dogs in close quarters and be cool with it in a matter of a couple of minutes. I can’t – but a punishment trainer can throw on a pinch collar and create an illusion of a “fix”. But it’s just temporary suppression – not long lasting change. My clients, however, day by day, week by week, are slowly changing what’s in their dog’s hearts. And once you have changed what is in a dog’s heart, their head will follow.
The second big thing I (re)learned from Kathy Sdao’s seminar is the importance of the sequencing and contingency required for effective conditioning (or counter conditioning).
In plain English – if you’re trying to get a dog to love something, or get a dog over their fear of something, you can’t screw this up!
Everyone knows about Pavlov and his dogs right? We would be very wise to pay close attention to the following diagram:
In Diagram 3, the bell proceeds the food. That’s how conditioning works – the stimuli is followed by the food, so that in Diagram 4, the bell causes salivation.
If the food were to be delivered before the bell, no conditioning would occur.
Sounds simple right? The emotional qualities and feelings the dog gets with food bleeds backwards into the bell.
Here’s where this simple sequence gets mangled in practical dog training and behavior modification.
Ruining Food as a Reinforcer
What if you were trying to get your dog used to being left home alone as part of separation anxiety work, so you begin by getting out a very special food dispensing toy, take it to the kitchen counter, and take a block of cheese and begin dicing it up and prepping the toy. The dog sees you do this and gets excited. Then, you give the dog the food toy and then proceed to put your shoes on and go out the door.
By going in this order, the food dispensing toy, block of cheese, dicing and prep at the kitchen counter will very quickly become a predictor of something really scary happening – you leaving home. With enough repetitions, your dog might start exhibiting signs of fear and anxiety at the mere sight of the food dispensing toy! That’s because the sequence is all wrong – food, then departure. One of the participants at the seminar remarked that her dog started hiding in fear whenever she brought out a can of dog food and began opening it – because that always meant she was leaving. For this dog, instead of conditioning good feelings around a departure, bad feelings became conditioned around cans of food!
A better way to handle this would be to put your coat and shoes on, get your wallet, keys, and cell phone, and then prepare the food.
Poisoning Cues and Behaviors
If you have a fearful or reactive dog, you can also easily ruin cues and behaviors. If you’ve done work around managing and treating reactivity, you are likely familiar with the “Watch me” protocol, which is to train the dog to look at you when the dog sees the trigger that bothers them (other dogs, strange people, etc.)
Let’s say you are working with a reactive dog and off in the distance, you see a dog approaching but your dog hasn’t spotted it yet. If you cue a behavior (like “Watch me”) and the dog performs it, and gets their click then treat, you may think you have dodged an outburst. But if the dog performs that behavior and then immediately after that, sees the dog, you can easily poison “Watch me” to mean that a dog is coming! Very quickly, you’ll see a dog start panicking and scanning when you cue “Watch me” because you got the sequence wrong.
The only way to handle this is to ensure that the dog sees the trigger first BEFORE you cue “Watch me”. Otherwise, if dogs follow “Watch me”, the fear and anxiety of dogs will bleed backwards into “Watch me” and ruin your cue.
On September 18th and 19th I made the six hour drive back to Endicott, NY, to attend a weekend long seminar hosted by Clicking with Canines (the school where I completed the Karen Pryor Academy). This was my first time hearing Kathy Sdao speak in person and with over sixteen hours of lecture, I left with so many new ideas that I’ll have months of blogging content in the hopper to draw from now.
One of the biggest topics of the seminar was the concept of conditioning new reinforcers. Big revelation right there for me. We spend a lot of time conditioning new emotions towards stimuli (i.e. getting a dog over fear) and training new behaviors. But, how often do we go out of our way to create new things that we can use to reward our dogs?
At When Hounds Fly, I spend a lot of time encouraging students to look for reinforcers they can use to train other than food. Food has limitations – dogs get full, you can’t always have food with you, and sometimes its not practical to use food (i.e. in evaluations and performances where food is not allowed). Creating new reinforcers is all about using classical conditioning to transfer the value of one reinforcer (i.e. food) to another (something that is not food?)
Over the weekend we saw many clips of Kathy from her University of Hawaii days working with dolphins. One reinforcer they conditioned was a hand clap. In one clip, a dolphin performed a behavior on cue and the handler didn’t feed it a fish, but instead, did a big open air hand clap. The dolphin saw the handclap and became absolutely giddy with joy! How awesome is that?
In thinking about how we can apply this with dogs – what if we could condition our dogs to react with joy when we do the following things:
– Praise the dog with a hearty “Good boy!” (assuming your dog is currently not that praise motivated)
– Give the dog a thumbs up
– Wink at the dog
– Transfer the value of food (from a food motivated dog) into a tug toy (for the same dog that is not toy motivated?)
The process of creating a new conditioned reinforcer is relatively easy! In fact, if you’re reading this, you’ve probably already conditioned the following reinforcers:
– Bringing out the leash (your dog wags his tail and gets giddy)
– Clicking (your dog reacts with joy and looks for their treat)
– Walking towards the entrance of When Hounds Fly (your dog starts going crazy looking for the green door)
How did you condition a leash, a click, or coming to school to produce such strong emotions? Simple – every time you showed your dog any of those 3 things, something wonderful happened immediately after. The leash is always followed with a walk. The click is always followed by a treat. Going through our entrance is always followed by a ton of food-based training.
My personal training goal is to now condition my dogs to have additional reinforcers. I’m trying to train Petey to become tug toy crazy. All I’m going to do now is prepare his dinner (in a bowl) and hide it somewhere like a bookshelf or counter. Then, I’m going to bring out the tug toy and tug with Petey for 15 seconds and then quickly bring down the food bowl and surprise Petey with dinner. Just by doing that, I’m going to attempt to transfer the emotional value of dinner to the tug toy.
Similarly, for both Duke and Petey, I’m going to try to condition a “thumbs up” as a conditioned reinforcer. All I’m going to do is when I click for a behavior I like, I’m going to give a thumbs up first and then feed my dogs. The emotional value of the food reward should bleed backwards into the thumbs up, and what I’ll be looking for is tailwags and a smile when I give my dogs a thumbs up.
So, my challenge to you is – you’ve trained a lot of useful and cute behaviors in your dog. What new conditioned reinforcer will you start creating?
Over the weekends of August 21 and 22; and August 28 and 29, canine volunteers and their humans came to visit When Hounds Fly after classes ended to participate in the first set of canine cognition research experiments, conducted by Krista Macpherson from the University of Western Ontario.
It’ll probably take quite a while before any sort of paper or findings are produced, so here’s my lay-person’s attempt at describing what went on, and what I anecdotally observed. (Krista will have to analyze hours and hours of video footage to actually see if anything meaningful happened)
Krista was interested in determining whether or not dogs understand the concept of timing – so this is the experiment she designed:
Two Manners Minder remote training devices were setup to dispense food on different intervals – one was set to dispense every 15 seconds, and the other was set to dispense every 60 seconds. After multiple repetitions, were dogs able to figure out how frequently each one paid off, and proactively move towards the machine that was ready to pay off next?
At first, the dogs were introduced to the Manners Minder machines. The machines are set to produce an audible tone prior to dispensing (the tone is normally used as an event marker, i.e. clicker, for training). Relatively quickly, each dog learned that the tone meant that the machine was going to pay.
Next, the machines were moved over and placed on mats created by taping different colored bristol boards together. The bristol boards serve as a visual indicator for analyzing the position of the dog relative to either Manners Minder during the task.
The owner and the dog start at a chair placed in between the two machines. The owner is wearing sunglasses, as dogs often can pick up on where humans are looking at for cues and information (i.e. the owner might start staring at the next machine to pay, and the dog will very quickly learn to go to the machine the owner is looking at).
What I found particularly interesting was something that Krista said about the differences between dogs and other animals. When doing these sorts of tests on rats or pigeons, the variance in behavior is relatively small. What was particularly interesting to see is the wide range of behaviors and strategies that each dog used.
Some dogs were very thoughtful and deliberate in when they chose to move back and forth between machines. Others used brute force and just simply went back and forth between machines almost non-stop. Many dogs (especially ones trained using shaping) initially believed that behaviors they performed could influence whether a machine would pay out (common behaviors included intense staring, pawing, mouthing, nose-targeting different parts of the machine – my Beagle, Petey, offered a play bow to the 60 second machine; Arlo, Emily’s hound, offered a favorite trick – the armpit sniff – to the machine; Farley, Jenn’s Beagle, howled at the machine as if to say “Feed me now!!!”).
A handful of dogs appeared to really be “thinking about it” and there were times where it seemed like they went over to the 60 second machine to get their payout right on time. My Beagle, Petey, just decided to pick up one machine and drag it closer to the other so he didn’t have to travel as far!
Some of the dogs are highly trained (agility dogs, advanced obedience, or just a really well trained family dog) and others were pretty raw (brand new rescue, zero training) – since the activity has nothing to do with performing behavior, it seemed to me that trained dogs did not necessarily figure things out faster or better.
Here’s a few more photos I took of some of the dogs that volunteered in August. If your dog tells you they’d like to participate, just email me at email@example.com and I will include you in future calls for volunteers. Krista is going to be running another set of experiments on the weekend of the 18th and 19th.
As many of you may know, my rescue Beagle, Duke, is dog reactive. Initially, he was fearful and aggressive towards dogs in all settings (off leash, on leash, at the dog park, anywhere!) and before we began working with him, the mere smell of a dog (a dog’s scent lingering in the hallway of our condo even thirty minutes earlier) would trigger a howling fit.
Along the way, we got Duke’s reactivity down to a point where walks became sane again and Duke was able to return to attending group obedience classes. But, after nearly four years of working with him, his reactivity is reduced, but not cured, and he is what I’d call “manageable”, but still requires a lot of management. Many dogs (including ones I’ve been hired to help with) DO become “cured” from this work but Duke isn’t there yet.
While completing my studies with the Karen Pryor Academy, Casey Lomonaco of Rewarding Behaviors Dog Training introduced me to a new technique for rehabilitating aggressive dogs called Behavior Adjustment Training (BAT). Created by Grisha Stewart of Ahimsa Dog Training in Seattle, BAT is a protocol that uses operant conditioning to help dogs learn socially acceptable ways to deal with things that scare them (dogs, people, objects, etc.)
Recently, I ordered and viewed Grisha’s first DVD on the subject – Behavior Adjustment Training (BAT) for Fear and Aggression from Dogwise.com. This DVD is a live recording of a lecture she did on BAT in Novato, California, in January 2010. It is NOT intended for average dog owners (its way too dog-geeky) but is a perfect watch for professional dog trainers and behaviorists that have a solid understanding of operant conditioning and classical conditioning.
In a nutshell, Grisha explains that reactive dogs have learned that lunging and barking sends scary things away. Either the handler, due to embarrassment and wishing to stop the dog from rehearsing unwanted behaviors, backs away and creates distance when their dog reacts – or the scary thing (the other dog owner, for example), leaves because they are afraid of the lunging dog.
In dog-geek terms, using the Antecedent-Behavior-Consequence model we have:
Antecedent = Scary Dog is approaching
Behavior = Bark, Lunge, Growl
Consequence = Scary Dog goes away
BAT is all about changing this sequence to:
Antecedent = Scary Dog is approaching
Behavior = Any calming signal or distance increasing body language – look away, sniff the ground, turn away, scratch, yawn
Consequence = Scary Dog goes away.
There are three stages of BAT, which I won’t go into detail with here – but in its third, purest stage, no food or toys are used. The simple functional reward of being allowed to leave by offering an appropriate calming signal is the sole reward.
When I first heard about BAT, I found the concept interesting, but I am delighted to say that with some preliminary exercises I did with Duke, the results have been quite positive. Today, by coincidence, just before completing this blog entry, I had a chance to work with a reactive dog client with some ad-hoc BAT exercises at the nearby dog park. Both of us were extremely pleased with the progress we were able to make in decreasing distance and keeping the dog calm and under threshold. The relaxed body posture and demeanor of the reactive dog was significantly different than when we previously were using traditional food-based classical counter-conditioning techniques. The dog’s owner commented on our swift progress and how much her dog seemed to “enjoy the game” of BAT.
I look forward to working more with BAT and sharing this new technique with the reactive dog owners in our community.
For average dog owners wishing to learn more, Grisha has put out a second DVD called Organic Socialization that appears to be better filmed and presented in a way that average dog owners can absorb and comprehend the information. That being said, for this or any other aggression issue, consult a positive reinforcement trainer or behaviorist that has a solid understanding of operant conditioning and classical counter-conditioning when you get started.
Here is a short clip from Organic Socialization including a brief clip from a real life BAT session: