Forum Replies Created
-
Posts
-
Good morning! Lots of great stuff here 🙂 Glad you got to have a fun dinner with a friend, that is always better than video editing LOL!
Looking at the videos:
Wing wraps – I think you can now go to treats that require little or no chewing 🙂 Chewing takes time, puppies lose their flow when they have to crunch and chew… – what do you have that she likes that is soft and easily swallowed, for speedy reinforcement? It will makes things more efficient for her 🙂
She is doing well on the wrapping here! So now you can switch your stanchion out for something bigger – laundry basket, barrel, cone – eventually a wing – so she has to go around something bigger and can’t see you as well.
When that goes well on a bigger object, you can work up to standing 🙂Hand touch – the mechanics are looking sharp here! So now, to prep for what is coming: just do it on one hand so she is not going back and forth. Leave the target hand locked out in position for a couple of reinforcement in a row – then play tug, then you can switch hands.
The touches look good to your palm, really strong! But for the purposes of what is coming, I do recommend a target of some sort (for saliency as a big target is bigger than your hand :)) I know, I haven’t told anyone what we are using this for LOL! but it is coming next week, I SWEAR hahaFocus Forward – I think this session was more about the collar grab/reset than the focus forward. The focus forward is fine, she is happy to look at the treat and move to it 🙂 As with the wrapping, I recommend a treat that requires less chewing – string cheese is great for this. And also, you can (and should) also do this with a toy.
So about the harness grab and her not wanting to come back… that is great info! So for this game, use a reset cookie to bring her back and turn her around so she is facing the right direction. Then you can either slip a finger on her harness or in front of her chest, then throw the treat or toy and release quickly. Definitely don’t bring her back by the harness as she really didn’t like that, and it will bleed over into making this game sad-making. Separately, work on a reset procedure of having her follow your hand for a treat, turn around by following the hand, then a gentle touch then an immediate release to the toy or treat. And definitely gets toys involved in all stages of this – she is going to be more like to reset when you have a couple of toys involved 🙂
Toy play – You can now incorporate having a couple of toys here, a bit of back and forth with different toys. The back and forth should both keep her engaged and also less interested in chewing on the one toy 🙂 and we will get a look at her reinforcement preferences (helps us to plan training and also to balance the value of various reinforcements). And that can also start to develop the retrieve as a nice by-product, because she will be coming back to you for the next one and not having any one-on-one time with the toy LOL! And you can add in talking her her, sometimes a little touch of the harness and let go so she gets a little more harness love built in too, eventually working up to moving her by the harness while tugging. We might never need to move her by the harness during training, but we sure do want her to not move away from it if we reach for her.
Decel – this is going well too! Remember to work both sides and not just left side 🙂 When she gets to you, feed the treat closer to your leg (your hand can be touching your pants) and with her head turned in a bit, especially on the pivots. Don’t keep her head straight or turned away – we do want her head position to be turning towards us here. She did really nicely so you can add a little action by walking away after the initial cookie toss, so she sees a little motion after she eats the tossed treat – then you can add the pivots in after the decel. If you have a longer hallway, you can add even more room and motion, she will like that 🙂 And for this one: toy play! You can totally get toys involved in this too!
Pre-Game with the prop – she might have been a little tired here but I don’t think she was was done – she had just started to get some good hits going when you moved the prop further away by a significant amount (to her!). She was not ready for that, had some errors, then got frustrated and jumpe don your head. So it was not so much of a “I am done” moment, it was more of a “I don’t know what you want now” moment. With puppies in particular, I live by the 2 failure rule in training – if she asks 1 question or has one failure, big red light goes off in my head. If she has 2 questions/failures, I make it easier immediately to recover the high rate of reinforcement. Then you can more gradually increase the difficulty.
She had one failure before you moved it then another right after you moved it – then 2 more at about 1:00 then she was frustrated. So even though she had reinforcement between the 2 initial failures, there were still 2 failures so I suggest making it easier (by moving it back to where it was) when that happens. Then in the next session, you can move it out again by but an inch, not 6 inches.>>I might need to find a new Thing since this bag is kind of slippery when she steps on it.
Or attach it to something that elevates it a little while also anchoring it, like duct tape it to a bath mat?
Nice work here! Let me know what you think!
TracyHi!
>>Holy crap! I let my OCD take over! Poor Spot, he’s such a good boy to put up with me.
Yes, set a timer or get a couple more dogs that need training, then you will never go too long with any one dog haha!!!
>> I figured out that there is a sh*t ton of ash in the dirt there though, so it won’t really be ready to use till I get some rye grass planted. Dogs, toys and I were soot covered when we got home. LOL!
OMG! Never a dull moment, right?
On the video: This went really well!! He was a really good boy with the threadles and balance reps here!
Things started off really nicely, lots of success. He is collecting nicely on the turn jump when you ask for it (even if it is the wrong cue, he still understands it as collection, good boy!)
On little error at :25:
Bear in mind that there is a positional cue associated with threadles (being in the gap between the jumps, not further past the first wing of the next jump) and only later on with more experienced dogs do we proof it by having you way far ahead. So at :25, you ended up getting way far ahead. You had a nice cue and he gave a good collection on the turn jump, but then you accelerated out of it, which ended up putting you in serp position between the uprights of the next jump as he was deciding to threadle or not – so he chose to read position and he serped. I think eventually the thredle verbal will override position (an you can open your upper body back to him a bit more too) but not yet – for now, aim to be also in the positional cue like you were at :34 which went a lot better (but had a little too much decel) and also :42 which was really strong and didn’t have any decel 🙂
It is a fine line of getting into that positional cue and not too far ahead and not too far behind, so moving a little closer to his line for now will help and not blasting ahead too much.>> When I did the 180 the first time I used a right command and got the threadle. After that I just used jump and he got it.
on the balance there – I think he was just in threadle mode after a bunch of threadle reps, which is pretty common. On the next rep, you were clearer with the closed shoulder and position so he got it. I am not sure if the right cue got more threadle focus for him and the jump cue was a better option, or not – so next time, use the jump cue on your first balance rep and see how he does. But also, balance more frequently so he doesn’t get locked into the threadle vortex and start threadling all the things, all the time 🙂
The traditional threadles at the end looked really good! He is really doing a great job of giving collections on that turn jump, which is setting up a super nice threadle!!!!
Well done on these! Let me know what you think!
TracyGood morning!
This was kind of a baby dog obstacle course! This didn’t seem like too long of a session (short blasts of each thing), but it depends on how much time in between each game you gave her – she is young so maybe she needed a couple of minutes in the middle.Good tugging to start the first video! One suggestion is to have the treats in your hand as you take the toy out of her mouth, so you can click and be immediate with the reinforcement – that way she can keep offering and you can have very fast rewards. You can also click then tug which will get her off the board, so she can then offer getting back on or another click 🙂
Good adjustment to move your hand slowly to get the turns! When you were too quick, she couldn’t quite follow what you wanted there.
Fit bone: on this session (and the others), try to stand still and let her offer, rather than moving around to try to get her to do more. You can release her to tug to get her on and off the fitbone so she doesn’t rely on your motion, then just let her offer – your movement and your hand in and out of pockets will cause her to wait for that as a cue rather than offer. We like the offering because it builds independence and confidence 🙂 You can also click and toss a cookie so she gets off then offers getting back on. Do you have 2 fitbones? That can give her a larger playing field to get back feet on too 🙂
I think it was on this video you mentioned needing 3 arms LOL! I feel that pain! You can get rid of the clicker in favor of fast cookie delivery (use a verbal ‘yes’ then deliver the cookie), or tuck the toy into your armpit so your hands are free for the clicker and food.
3 – cooler:
This also went well! You can see if she wil keep tugging while you take he treats out of your pocket, that will also help with quick delivery when she makes that first offer. As with the others sessions here, don’t move for the shaping parts – move during the tugging, but then be stationary so she can offer independently. She was a good girl to get up on the cooler!!!! That is not easy!!Last video – she got back into the tugging when you moved it away from her (prey drive) so remember to move it away and not towards her or upwards. You were moving it towards her at first so she was not as interested as when you let her chase it – good adjustment!!!
For whatever reason, the gravel was interesting to sniff – it is possible that she simply has not had experience training on the gravel so she was investigating the environment. It is also possible that she was a little concerned about the noise on the gravel and that is how she was showing it? Or, 3 sessions were good but she was a little mentally tired with 4 of these short sessions. Hard to know! I think she did best when you moved the tugging back onto the grass – so when you train again, do this game (wobble on gravel) first and see how it goes! She seemed happy to make it move so it will be interesting to see the next session 🙂One more idea:
The cooler, wobble board and fitbone are all compression exercises for her because of her size (they would be extensions for my Papillon lol!) – she has to gather herself into a smaller shape to be able to be on them. That is great, and now we can also work on extensions so she can stand comfortably or even a little stretched out. You can build extension setups by putting a couple of things together, like 2 of the fitbones, or the wobble board plus fitbone so her front end can be on one and her hind end can be on the other.Nice work here! Let me know what you think!
TracyGood morning! Good to see you here and in puppy class too LOL!
The remote reinforcement concept is a great one for puppies, I wish I had started it earlier with my older dogs. My youngsters transitioned to trials really easily because they had a background in remote reinforcement.
>>Here I’m going with “Let’s go get your…” cookies or toy in the future. I know it’s kind of long but I feel I’m most likely to actually use it in a training or show context vs just saying “cookies.”>>
I am with you on this one – I think the ‘let’s go…..’ sentence is great because the dogs will cue off the “let’s go” and the rest of the sentence (I am pretty sure there is a happy conditioned response to the word “cookies” or “toys”), as well as the context and tone of delivery. And it will feel very natural to us at the end of a run, rather than having to remember an exact precision marker in that moment. I mean, I am sure we can work on getting a precision marker instead or we can teach the context with this more casual marker and have it work out beautifully 🙂 I personally think we humans have enough verbal cues on our plates that MUST be precise that we can be a bit casual here and let the contextual cues and brilliance of the dog help us.
On the video:
This went really well! A couple of ideas for you:Try not to pair “yay” or “good boy” with the immediate RR marker because yay or good boy will instantly become the marker and the dog will stop what he is doing, because he predicts heading to reward (ask me how I know this can happen very easily LOL!! Oops!) And then we get errors and frustration when we don’t want it to be the marker but yet… it is. So – you can praise, that is all good – but if you praise and then want to run to the RR, give yourself a second or two between the praise and the ‘let’s go’ cue. And, sometimes just praise and don’t head to the RR. That way we keep praise as a separate entity and not in any way paired with any delivery of reinforcement.
>> got his nose coming up near my face instead of an elevated hand touch. That’s usually a sign of frustration,
Yes, it was good feedback from him! The high nose touch is stimulating for him in general, I think he had some vocalization in moments when he was not necessarily frustrated – and that is also interesting to note because that was a little later in the session, when he might have been getting depleted despite the high ROR.
For these types of sessions, definitely set a timer. Because things are happening really quickly and the reinforcement is nearby, there are a lot of reps in 4.5 minutes. And I think he was a bit toasted by the end, especially when asking for more and more behavior and when the behavior doesn’t have a lot of action to it. The first minute had a lot of behavior – and then by 3 minutes, you could see him changing his behavior and spending more time near the reinforcement. This is really interesting because the ROR was really pretty high – but if sessions are long and repetitive, we will still get mental fatigue/check outs. We are starting to see more and more studies supporting depletion/mental fatigue in dogs leading to shifts in arousal & errors, even when there has been a high rate of reinforcement. Fascinating stuff!! So, we build mental endurance with shorter sessions that get longer gradually, over time, as the brains mature.
So for this type of session where you can get a lot of behavior really quickly, depletion can happen juts as quickly: you can break it down to 90 seconds or less, then have a break, then come back later for another 90 seconds, etc.
When things are ‘bigger’ (like sequences) and there is not as much behavior happening in that minute and chains/sequences are longer, I think 4.5 minutes is no problem at all because the endurance is in place for the mentally challenging stuff, and if it is not mentally challenging, then it is easier to stay focused.Great job here! Let me know what you think!
TracyGood morning!
I hate when that happens with the internet! I have terrible internet here at home, I feel that pain 🙁And coffee is a good thing – 2.5 cups in to my day haha!
>>. If the value of the food is too high, he’d rather food. If the value is too low, he’d rather the frizz. I have not tried active food movement with it, so that may be a good place to start.
>>A lot of his not taking food comes with over arousal. I know I need to practice modulating but what’s the best way to do that. If he’s above threshold, he usually won’t tug and may or may not take food.>>Think of it as behavior and not as reinforcement, then. Behavior changes in certain contexts so the first step is to identify those contexts, plan, then train behaviors. And rather than considering “food” as one reinforcement, break it into several different ones: cheese, kibble, MM, etc. That is no different than frisbee, ball, tug toy, hollee roller as having different values.
What puts him above threshold? Make a list. (There is distraction list homework this week anyway :))
It might be a fluid list that changes as he grows up and gets more experienced, this is totally normal.And, looking at the reinforcement hierarchy list: what are guaranteed reinforcements when he is NOT over threshold?
And for the back and forth between food and toys, which is one of the most useful procedures in animal training, two approaches:
– high value food and toy
– low value food and toySo for high value food – high value indicates it is a reinforcement and the toy is therefore lower in value. That means the value building procedure is about the toy, not the food. So you can work the building value for toys procedure in this case. I really like having a lot of action when building value for toys – run a jump-tunnel sequence while holding some cheese, throw a toy that stands even a remote chance against the cheese… any interaction with the toy gets marked with ‘snacks’ or something that indicates the availability of the high value food. And because value building is really just shaping, remember to do this with successive approximations: start with tiny slices of the behavior (interaction with the low value toy) and, over time, build it up to the more complete interaction.
And then the opposite happens for the low value food – it is a different item, and a different behavior. Eat the low value food, get the high value toy or activity. Eat a piece of kibble (criteria: swallow), get the ‘go for a swim’ cue. Of a frisbee toss, or open the door for him to have a run in the yard. Value for that low value food will come up REAL FAST lol! When Voodoo was about Fever’s age, he didn’t have a real interest in food. But he liked running in the yard and he liked agility trials. So in order to go for a run or to enter the ring, he had to execute the cued behavior of eating a treat. (criteria: swallow!) Food value came up really fast and now food is really high value in all situations.
Let me know what you think!
TracyGood morning!
Nice session here!!! This was good information about the value of placed reinforcement – dead tenny or dead frisbee are not necessarily perceived as reinforcement but the lotus ball is. It was a great reminder that ‘reinforcement is in the eye of the beholder’ – in other words, the dog decides if the thing is actual reinforcement or not LOL!
So that gives us 2 thing to remember –
– when trying to build a behavior, use what she perceives as reinforcement – lotus ball if it is a placed reinforcement, for example. This wing wrap is mainly to help us sort out the procedures but if you were legitimately trying to shape a new behavior or bring the behavior into a new environment: always choose what you know is guaranteed to be reinforcement. And these little experiments allow us to plan for that because we get great feedback from Keiko!– And separately, build the value of the ‘dead’ reinforcement to get her driving to it and picking it up, so it is available in your toolbox of reinforcement at some point in the future. I do that with toy races from the puppy games 🙂 And also shaping it a bit: a tiny bit of interaction followed by the toy then moving (like the friz on a leash – if she interacts with it or touches it with her mouth, you can reward that by moving it).
And when she will interact with the ‘dead’ toy at all, you can install that into the ‘building value for toys’ concept by putting it into a series of behaviors that have a long reinforcement history (that she also really enjoys :)) and following it with a higher value reinforcement. So for example: Do a tunnel, send to a ‘dead’ tennis ball. when she interacts with the dead ball, even just a little: mark it with the marker to chase the frisbee or tug on it.
You can see how well that type of value building works, with the frisbee tugging! You basically put frisbee tugging into an existing reinforcement process – wing wrap then tug on the moving toy – so it didn’t take her long at all to begin to tug on the frisbee rather that wait for it to be thrown!
Great job here! Let me know what you think!
TracyGood morning! There is actually a lot of good stuff happening here!!!
Value looks good, he seems to understand how to hit is and he is moving to it. The only question he has is how to leave your hand and leave the reinforcement. So, we tweak it a little to shift the value off of you. 2 ideas:
I know the instructions says to reward back at you, and you’re being quite perfect. So perfect, in fact, that value has shifted to being so tight that he doesn’t want to leave you 🙂
So, let’s shift value back to leaving you: send then click then toss the treat out past the pillow. It will get him driving to it a lot better and less interested in your hand. Now, if he loses that value of coming back, we can change placement – that is pretty normal to do with puppies.
If you feel like you need a 3rd arm, you can drop the clicker and send with one hand, toss with the other.
One other idea: he is not always ready for tue next send after the cookie. So call him back to you, give him a little ready dance… then send. That helps prep him for the send and gets him jazzed up too! And it folds in some handler focus versus send focus 🙂Let me know what you think! Nice work!
TracySimple! Remote reinforcement 🙂 in training, when the reinforcement is present, it also serves as a cue in many ways.
Without the presence of the Reward, her arousal changes and the behavior changes.
Putting all of the behaviors into the remote reinforcement procedures will totally help! I dod it with the weaves in the demos posted yesterday.
Separately, get her as stimulated as possible in training (do 5 tunnel in a row til she is frothing :)) then do a stopped behavior. At the beginning, the Reward can be present but then work up to being able to do this with remote reinforcement.
Let me know what you think!
TracyHi! You can randomize the toy placement when it is “pre-placed” and don’t let her see you place it there. 🙂 Ping pinging where the reward is (close, far, closer, further, and so on) will help build the duration of this behavior which in this case translates to the number of jumps she takes.
I like the helper idea! But the #1 thing as that the helper does not become a target 🙂 so when someone else is throwing, have the person in random areas anywhere on course except that last jump 🙂 and the person should also randomize where to throw the Reward, so it is happening early and later, also ping ponging it.
Last idea – because we see this at the end of courses, you can build it into Remote reinforcement by back chaining:
Last jump then let’s go to reward
2 last jumps then let’s go
3 last jumps then let’s go
So the remote reinforcement and the leash are the targets that can always be there 🙂
Let me know what you think!
TracyHi!
The dead toy reinforcement is definitely a part of the get it procedure that you can build up, which makes it useful for weave training as you described. Away from weaves, what does she does on a get it to a dead toy on the ground?You can also move the toy more than 6 feet away, the “pressure” of the toy being so close might be eliciting the stalking behavior. Try it 10 or 15 feet away and see what she does!
T
Hi! In trying to figure out how to get more tug, the first question is: what will he put his mouth on? Can be traditional things like toys or frisbees or balls, or untraditional things like shoes, old socks, paper towel rolls, feathers…. we can see what he likes to put in his mouth and build it from there!
TracyHi!
The next one is on December 9 at 7pm.
>> If only they would allow cookies in the agility ring!
Do this with toys! And his leash 🙂
T
Hello! Welcome back!
She was really catching on! What a good girl! This all looked really strong.
She did best when you led out watching her so keep eye contact as you move away, for now. And letting her settle into position before you step away might be helpful (but some dogs just want us to leave and get on with it, she will let us know)Glad you did it in front of a tunnel, I was going to suggest that but you already had it and she was perfect. YAY!
The catch reinforcement really solidified the stay with a couple of my youngsters too, I still use it a LOT in training and also in NFC runs at trials. So keep using this a LOT with her, mix it in everywhere 🙂 And, be sure to balance it with releases forward to the jump or tunnel: that will add more stimulation, which in turn will challenge the stay behavior.
Great job here!
Tracy
