I’ve been trying to give all of my students some tips to improve their mechanical skills as trainers. Over the last few sessions I have come up with my “5 Tips to Successful Training”. These are just very simple changes in the mechanics that can make the world of difference in the fluidity of your training and the rate of learning from your dog. I’m frequently bringing up these topics to my students/clients and those who improve these skills begin to see a difference in their dog during training.
I noticed that this session I’ve been much more focused on pointing out these tips to my students and I’ve definitely seen the handlers become some of the most skilled that I’ve worked with (which is fantastic).
Timing
Really focus on your timing and make sure you are using your marker word/sound. “Yes!” “Good!” *Click* are very important in this type of training. When you clearly mark a behavior you are becoming much more clear in your communication with your dog. They know, without a doubt, what it is that earned them the reward so they can replicate it. If the treat just happens, dogs often do not know exactly what caused the treat to happen so they do not know what to repeat so it can take a few more repetitions to connect the dots. If you have precise timing with your word/click you can cut down the time it takes to make a connection to the desired behavior.
To practice timing I suggest people sit on a bench at a mall and mark/click as people pass a certain point, or as their left foot hits the ground, or other similar moments in time. You can have a friend bounce a ball and mark the upper and/or lower vertex (highest/lowest point in the bounce).
Quiet Hands
When I was growing up I was very into horses and riding (shocking right?). I took dressage lessons for 5 or 6 years I think starting when I was just 5. I remember learning a lot about keeping my hands quiet and instead communicating by shifting my body weight and leg pressure. I think I took that to heart because it’s something I’m always reminding my students. Keep the leash quiet. Don’t rely on the leash to guide your dog or communicate with your dog. We can use our body language, hand targets, vocal cues, and even lures to maneuver our dog around… we don’t need to rely on tugging on the leash. I noticed that lots of people (with prior training experience or not) would raise up the leash waiting for a response from the dog… it wasn’t a leash pop but simply raising their arm until the leash was very tight. This tightening of the leash often hinders the response of the dog so I was regularly remind them to “lower your leash hand” or to relax the leash.
One of my favorite things to practice quiet hands is to work dogs off-leash or on a hands-free leash. If there is no leash or the leash is not in your hands, it’s a lot more difficult to have a leash being tight or used to maneuver a dog around.
Rate of Reinforcement
The quick and easy response is… professional trainers reward more frequently. There was a study done (I cannot find the source) but it looked at the rate of reinforcement of trainers v.s. typical dog handlers and trainers were rewarding at double the rate of typical students. In a 90 second session with a dog (for fast behaviors like sits, downs, or “follow me” game) I aim for 30+ reinforcements. My goal is to treat my dogs at least every 3 seconds. Some human handlers don’t even get 30 rewards in 3 minute long session. If you are working your dog on “fast” behaviors, increasing your rate of reinforcement can do wonders for building focus and improving the response.
I try to practice by getting 10 treats in my hand and try to reward 10 behaviors within 45 seconds (to start, eventually I decrease that to 30 seconds). Often times it’s maintaining that rate of reinforcement that is difficult so I start small and help them work their way up.
Do Not Repeat Cues
Be patient with your dog in the beginning. Do not be afraid to wait for 30 seconds for them to respond to a cue. When you repeat the cue you are simply undermining yourself. You are teaching your dog to ignore you the first time you ask for a behavior. When you become aware of your rate of repeating cues (are you a “sit, sit, sit sit” or a “sit… SIIIIIITTT” type person) you can begin to start focusing on keeping quiet when you would normally want to repeat the cue. It’s terribly difficult to just wait it out but it can be the difference between having a dog who responds to “sit” and one who responds to “sit, sit, sit”.
For students who are quick to repeat a cue (sit, sit, sit sit), I tell them to say “sit” and then take in a long slow breath and then breathe it out long and slow. This generally gets them through the first moments they would normally repeat the cue. Similarly for people who say “sit…siiiiiit” I have them cue the behavior and then when they feel the urge to repeat it to, instead, breathe out a slow breath. It’s building an incompatible behavior 🙂
Precision Placement of Rewards
There is a big difference between bringing the reward to the dogs nose and having the dog come to the reward. Depending on what behavior you are working on you could do either. One of the most common loose leash walking mistakes I encounter is that the people are putting the reward to the dog’s nose. Eventually the dog pushes further and further away from the person OR they start getting their hip/butt out so they are more perpendicular to the person and side-stepping along. Instead of giving the treat TO the dog, rewarding right next to the seam of the pants brings the dog to exactly where you’d like them so they tend to stick closer since they have to get back to that position anyhow to get the treat. When you are asking/rewarding for a down, instead of rewarding if the dog breaks the down (in between the click and the treat delivery) use the treat to get them back into the down position and reward in the desired position.
Just be mindful of what you want… if you want to reward a down, make sure the pup is still in a down. If you want to reward LLWing why give the treat 18″ away from you? Dogs will return to the place where rewards have happened, use that to your advantage. So yes, even if your dog breaks the down between the click and food delivery they will know what earned them the reward BUT you can send home two messages that maintaining the position is also rewarding since the food happened in that position.
Tena,
I’m trying with Delilah but she gets so excited when she sees the food. I am becoming very frustrated with her. She WILL NOT stay in a down. Jodi fails.
Jodi, if you are using a clicker you simply withhold the click. Instead of clicking the instant the dog’s elbows hit the ground you can say a nice rewarding word “goood” “yesss” “thanks” but withhold the click until she’s stayed in the down for a second or two. then you slowly increase the amount of time before she gets the click.
Or you can rapid feed with her in the position. Her elbows hit and you are giving her probably 2-3 rewards/second… then RELEASE HER from the down. Eventually you just reduce the number of treats… instead of 2-3/sec you do 1/sec, then 1 every 2 seconds, 1 every 5sec, and remember to release her. eventually she’ll understand that if she says in the down there is always opportunity for reinforcement and that you’ll tell her when she can move again.
These are the two “try this first” that i use with dogs who pop up from downs (COUGH* Rio)
I feel so woefully inadequate. I think I found someone who can SHOW me, that is the learning method that works best for me.
But I will implement this and see how we do.
Don’t feel inadequate! Goodness! I wasn’t born with my training knowledge… i had to learn it… we all start somewhere! no need to feel inadequate! You are learning, as we all are!
I’m sure you’ll do great! Glad you found a trainer who meshes well with you!!
I love the idea of teaching timing by clicking every time a ball bounces on the ground. It’s a brilliant suggestion and I think it would have helped me a lot back when I was learning. Not that I am not still learning.
Reward placement is so key! That is something I forget a lot. Over the weekend one of my trainers reminded me of that during the fun match. Shiva kept jumping in front of me instead of running beside me like she is supposed to, and it’s probably because I haven’t been good at rewarding by my side with the hand closest to her. Back to the beginning I go!
These are little things why I try to stay in at least one class with the dogs.. because I don’t always notice when I’m not being a precise as I could be! I love having someone point out when i have sloppy clicks, less than precise reward positions etc.
great suggestions – i’ll have to work on that. I avoid repeating commands too much (i htink), but Gwynn has the SLOWEST sit ever. he just takes ages to actually respond to the command, as if he didn’t hear it at all… hard to resist saying something over 😛
Lexy for slow sits, you can raise the criteria … so it’s not just butt on ground but it’s butt on ground within 5 seconds… then within 4 seconds then within 3 seconds. If a sit happens after your time limit… too bad for the dog. So you have to start with a time limit where your dog can be successful but you simply up the criteria to make the sits faster 🙂
Great tips. From a totally theorhetical not-pet-owner perspective, I want to know why “Sit sit sit” doesn’t work the same way as changing a cue? The first stimulus eventually predicts the second, so you get a SOONER response?
I think because it’s not consistent (though on paper it does have some validity). If the dog S-I-Ts on the first SIT they are rewarded… if they s-i-t on the third sit they are rewarded there isn’t necessarily a link between each “sit” to build that same predicting the next cue. Plus I think it sets a precedence that if the dog ignores the first cue that they will still get rewarded if they eventually comply after the 5th cue. Also, given the way i’ve heard people use multiple cues, I wouldn’t be surprised if dogs start hearing “sitsitsitsit” as one word/cue so “sit” doesn’t actually mean butt to floor.
REALLY good question though!!! I love it!
If I’m teaching a new cue, though, I’ll give NewCue (NC), Old Cue (OC).
NC–OC–behaviour–treat
NC–OC–beh–treat
NC–beh!!–treat! and maybe a party cause I’m a bad trainer like that
NC—- — — OC–beh–treat
NC–OC–beh–treat
NC–beh–treat (no party cause last time the dog forgot!)
NC– — beh–treat
NC–beh–treat
NC–beh–treat
NC– ———OC–beh–treat
I do help the dog if they fail or can’t figure out what I want. I almost never repeat my new cue, but just add the old cue back in.. Seems this might be similar to the inconsistency of someone in a training class?
I think the precedence is set the same way.. that if the dog just waits for the old cue, the behaviour will be rewarded. I think this is (should be) only a problem if the behaviour is NOT taught with R+. An R+ trained cue should be an OPPORTUNITY, right? I’m telling you what you CAN do if you want what I have! So, assuming the dog is in the game, they should WANT to do it as soon as they know what you want.
The last bit, it sounding like one big cue, yes. I could see THAT being an issue.
I’m curious though.. how often do you actually see dogs who wait for the 3rd or 4th cue, once sit (or whatever) has been trained to fluency with R+? I use this in agility a lot.. Tunnel TUNNEL TUNNEL!!!! Mostly, because I can (no rules on cuing) and maybe as a keep going signal? not sure how it works functionally.
Cool discussion.
Tena, I know you’re in the process of starting up your own business. I’m in the early planning stages for my own walking business and was wondering if I might pick your brain about your experiences sometime. If you’re feeling chatty maybe drop me a message at percipient [at] gmail [dot com]!
By the way, this is a great post – very helpful for people just starting out, or for more seasoned dog folks who could use a refresher.
Early on in agility training I realized the dogs went where the reinforcement was delivered — away for the folks with object-rewarded dogs and close in for folks with food-rewarded dogs. That clicker training pioneer Bob Baily has a saying, I;m told, “Click for action, Reward for position”.
Personally, I need to get fluent in use of my new Flip video. Nothing like watching a video fo your training & competition sessions….
Excellent post. It’s great to get reminded of the basics and have them explained so well, thank you:)