Existing draft · inherited source review / markers
The trainers use cue poisoning to describe unwanted associations or confusion around a trained word; the examples emphasize clear meanings, timing and the individual dog’s response.
On this page · 22 sections
Answer
A trained word can become confusing or predict an experience the dog avoids. In Deuce’s owner lesson (l3IAjG4Zk18, 00:45–02:09), the trainer explains this through equipment timing and through using down to mean stop jumping when the dog has learned it as a position. These are related clarity problems, but they are not the same event. The equipment sequence belongs to that dog’s taught program; the page is not a stand-alone instruction for applying a correction.
Cue consistency also appears without equipment. King’s lesson (0Ryn-sQhWK0, 15:05) distinguishes recall from moving along together. Zeus’s go-home lesson (FIORySoF90o, 11:20–11:44) explains how allowing a walk-by can change what come asks for. In Lola’s session (so5_rbhUZ7U, 10:59–11:34), the handler catches repeated cues and explains the response pattern he wants to avoid teaching. These examples make the intended behavior, rather than simply the word used, the useful thing to watch.
The records also show trainers adjusting an exercise when they think its associations are becoming unhelpful. Bo’s trainer lowers criteria after a difficult attempt (Jc0G4EtRIS8, 07:28–08:07); Dolce’s session proposes pairing a disliked exercise with tug (cp1qRyv1b2U, 00:26). In another outing, the trainer reads Luna’s lowered tail and decides to wind down (FN9P4ETmTiQ, 10:48). Those are particular observations and decisions, not proof that every hesitation means a cue has been poisoned.
Several owner lessons describe repair as potentially taking weeks. Those statements are trainer estimates, not a measured recovery timetable for all dogs. The records describe more than one possible response, including retraining, changing a word or changing the exercise. They also contain differences in how trainers discuss corrections around other dogs. The source comparisons below retain those questions rather than presenting one simplified sequence as the answer in every setting.
Owner takeaway
Keep a short record of what each cue asks your dog to do and use the vocabulary taught in your own lessons. If the response becomes hesitant or different, note the setting, your words and what happened next so your trainer can review the sequence. Ask for a demonstration when timing or equipment use is unclear. These examples support checking understanding and adjusting the exercise; they do not establish a universal correction sequence or a fixed number of weeks to repair a cue.
Across the cases
Deuce
Deuce's first owner lesson is where the concept is taught most completely, and it is taught to somebody who has just acquired an e-collar. The rule comes first - the word, then the collar, never the collar on the word - and it is flagged as one of the few things the trainer intends to repeat at every remaining contact, which tells you how he ranks it. Then comes the mechanism in the plainest form in the corpus, followed immediately by the scene the owner will actually recognise if she gets it wrong: the dog trotting along, hearing sit, backing away, asking what he did wrong. Then the price - two and a half to four weeks, or a new word, which he calls awkward. And then, in the same breath, the version that has nothing to do with equipment: the most common owner mistake in this corpus, shouting 'down' at a jumping dog when down already means lie down, which is why the prohibition marker exists at all. A separate session adds the breed-specific route to the same failure: a Rottweiler's confident look reads as stubbornness when it is often confusion, and pressing the wrong reading is how you poison a command by hand.
Cuz if you say sit and you hit the e-collar, I know, bro, you're good. If I I say sit and then you hit the e-collar, that dog is going to go, "Oh, sit's a bad word."
Watch at 00:45 ↗ · Current transcript match · Audio confirmation pending
it can take like three to four weeks. I'm going to say maybe two and a half to three weeks sometimes just to like unpoison the word.
Watch at 01:23 ↗ · Current transcript match · Audio confirmation pending
it's it's important that you use the right word. So, like some people their dogs will jump and they're like, "Down, down." And I'm like, "Down is like when you lay down."
Watch at 01:54 ↗ · Current transcript match · Audio confirmation pending
hand poison a command thinking that they're being tough because they have this like kind of look and they're very confident so when you see them like lock up sometimes you're like oh he's just being stubborn but genuinely they can be kind of confused
Watch at 02:07 ↗ · Current transcript match · Audio confirmation pending
Dutchess
Dutchess's record holds the rule and the running audit. The rule is delivered to her owners with the mechanism and the repair cost attached, and the worked example is a down: correct before the word and she learns that down means the collar, which takes three weeks to undo. Two further refinements are hers. The first is that the tool itself can be poisoned as well as the word - she has a habit of running from collar pressure, which the handler is careful to say is not evidence that it hurts, and he therefore holds two constraints at once: gentler, so she does not become frightened of the collar, and clearer, so the way out is findable. The second is that he counts. Mid-sequence he names the boundary out loud - not into bad territory yet, but he is tracking how often he is hitting the button - and elsewhere in the same session he declines to correct a slow response because he does not think she knew.
Quotation held: its full wording does not match the current transcript. Preserved-track provenance is recorded in the private audit.
with it. So, nothing wrong with that. We just have to make sure that number one, I don't make her scared of the e-collar, so I got to be a little more gentle. And number two, make it really clear.
Watch at 00:46 ↗ · Current transcript match · Audio confirmation pending
minute here. We're not getting into bad territory yet. If I get if I keep hitting that e-collar. Hey, Duchess.
Watch at 12:40 ↗ · Current transcript match · Audio confirmation pending
Yeah. Okay. So, why don't I eat car there? Well, she gave me a face. She was looking around. I don't think she knew, honestly.
Watch at 14:09 ↗ · Current transcript match · Audio confirmation pending
Luna
Luna's file contains the corpus's best statement of the rule, its price, its most important extension and its sharpest internal disagreement. The rule and the price are handed to her owner in the standard form, with the cost stated as lost program time rather than as principle - three weeks of repair inside a four-week program. The extension is that the unit at risk is not always the word: when a down cost her, the concern the handler states is that the whole outing will acquire the association, that she will conclude being out here means a bad experience every time - and for a dog whose entire treatment plan is being taken places, that is the failure that would undo everything. The repair is a completed loop in a single file: stop, walk, sniff, praise, and check the tail is back to normal before doing anything else. And the disagreement is the argument about triggers: hitting the collar every time she sees a dog teaches her that dogs mean pain, and a dog with nervous tendencies generalises that into dogs being dangerous. The alternative builds prediction instead - about ten repetitions and she performs the redirect herself, having decided it is not worth it - which from outside looks exactly like a dog who does not care.
um and then you know that's classified as like poisoning a command and to unpoison a command can take 3 weeks and when you're program is four weeks. That kind of sucks, right?
Watch at 09:33 ↗ · Current transcript match · Audio confirmation pending
their tail's a little low right now so we need to wind down because she thinks out here means sucky experience every time it's not good
Watch at 10:48 ↗ · Current transcript match · Audio confirmation pending
Source withheld by the current visibility or collection review.
one more thing I want to go over is so I say, "Let's go." Each time she sees a dog and we condition eventually to the point where she sees a dog, she goes, "Oh, he's going to say, "Let's go." And if I don't listen, I get the e-collar. So, it looks like she just doesn't care about the dog. But in reality, she's actually predicting what's going to happen and just says it's not worth it.
Watch at 21:24 ↗ · Current transcript match · Audio confirmation pending
Finn
The existing page is built on Finn, and the reason is that with him the risk is named as the governing constraint rather than as a caution. The handler says outright that he is scared to poison any command with this dog, because he builds aversive associations very quickly - and then qualifies it, which is what makes it usable: not necessarily a bad thing, but it means no sloppiness is affordable. Everything else in his file follows from that. The markers are delivered deliberately quietly, so that the words themselves do not add to nerves he already has too much of. Physically placing him into a sit is rejected as a teaching method because it produces inevitability rather than meaning, and he does not like the position at all. And the equipment gets the same treatment: a short leash held back is described sarcastically as a way to manufacture reactivity, which is why the store trip is done on a retractable line. On the other side, his ordinary sessions protect the recall by releasing him straight back out after every one, so coming in never costs him the freedom.
right there, I'm a little careful because I'm really scared to poison any command with him, especially. Because he does seem to build, like, aversive behavior very quickly. which is like not great for training I mean it's not necessarily a bad thing but it doesn't mean you really can't allow for any sort of mistake
Watch at 11:12 ↗ · Current transcript match · Audio confirmation pending
Quotation held: its full wording does not match the current transcript. Preserved-track provenance is recorded in the private audit.
What I'm doing is I'm not doing the butt push to teach the command. That won't teach you anything. What it does is that he does not like to be in that position at all.
Watch at 12:32 ↗ · Current transcript match · Audio confirmation pending
Quotation held: text matches, but timestamp or full-span transcript confidence still needs review.
Lola
Lola's material gives the concept its failure state and its most uncomfortable admission. At the go-home lesson the lead trainer describes asking people to show him how they correct and watching them say sit and hit the button in the same instant - "you just taught the dog the sit means you're getting punished" - and then describes what such a dog looks like, standing there not knowing what to do with its legs shaking. Then he says he did it himself: the first dog he ever trained, a dachshund, bolted at the word sit, and he felt terrible about it. A second session applies the same idea to repetition rather than to punishment: saying 'down' over and over would make the word mean 'down, eventually', and the implied stay becomes void. And the second in-program lesson supplies the most consequential technical distinction in the corpus - a collar fired with no command attaches to the person or dog she was looking at, while a 'let's go' followed by the collar is a punishment for not listening, and those are two very different worlds.
some people still were like I just like I was like show me how you do and they're like sit boom and I'm like yeah you just taught the dog the sit means you're getting punished which that's a really bad thing
Watch at 18:56 ↗ · Current transcript match · Audio confirmation pending
first dog I ever trained actually when I was learning how to train I had a little dachshund and I I said sit and he like bolted away cuz I had like trained it the wrong way and I felt really really bad
Watch at 19:11 ↗ · Current transcript match · Audio confirmation pending
Quotation held: its full wording does not match the current transcript. Preserved-track provenance is recorded in the private audit.
Quotation held: text matches, but timestamp or full-span transcript confidence still needs review.
bo
Bo supplies the only instance in the corpus of a trainer catching the damage as it happens and running the whole repair on camera in ninety seconds. Too many failed heel repetitions in a hard store produce a self-diagnosis - he might have poisoned that a little - immediately followed by the mechanism, which is his own criteria rather than the dog's willingness. Then the sequence: a break, a deliberate repair step he names 'reconciliation', the criteria lowered, and several clean heels. Bo's file also holds the ethic in its gentlest form. A sit arrives badly and unhappily in public, and the handler declines to run it again for the camera's benefit - the record stands in other videos - and the only explanation offered for the failure is that dogs sometimes just feel strange, with no mechanism invented to cover it.
Come on. Alright, I might have poisoned that a little bit.
Watch at 07:28 ↗ · Current transcript match · Audio confirmation pending
I might have been a little bit too high on the criteria end there.
Watch at 07:47 ↗ · Current transcript match · Audio confirmation pending
Okay, good boy. Good heel. Good job. Lower the criteria a little bit. Let's go. Come on, Bo.
Watch at 08:07 ↗ · Current transcript match · Audio confirmation pending
He'll get over it. I won't press on that command again to prove to you that it's perfect, but he is doing well to sit.
Watch at 02:53 ↗ · Current transcript match · Audio confirmation pending
Dolce
The protection dog's file adds three protections that are about experiences rather than words. The first is the hardest to argue with: she slips and falls on an obstacle, the exercise is abandoned, she is comforted, and there is no second attempt - a handler choosing the dog over the demonstration with a camera running. The second is a refusal to spend a command he was not willing to enforce: he uses an informal recall on purpose, knowing she did not want to go that way, and says afterwards that she did not disobey a command because none was given. The third is a repair session in its own right, aimed at a heel that had gone slightly sour - pair it with a tug, let her win something, use a more encouraging tone - and the reason given is the small-win principle stated before it is applied. One more moment belongs here: a correction withheld because the behaviour did arrive, slowly, with his finger already over the button.
Quotation held: its full wording does not match the current transcript. Preserved-track provenance is recorded in the private audit.
Quotation held: text matches, but timestamp or full-span transcript confidence still needs review.
Honestly, even if it's just a little bit, I think it'll help just to make her feel like she's winning something.
Watch at 01:25 ↗ · Current transcript match · Audio confirmation pending
so X I almost punished there you look kind of long to do it I didn't punish because you did it like the last I got like my finger was over the button I was like all right cool
Watch at 06:41 ↗ · Current transcript match · Audio confirmation pending
King
King's file contributes the variant nobody calls poisoning and everybody does: degradation by misuse. Using 'come' when you mean 'let's go' messes up the recall, and the handler ties the rule to the moment it will matter rather than leaving it abstract - the reason you are saying come is usually that you want the dog not to be over there. The worked failure is specific: another dog appears, you say come, he runs towards you and loops straight back out, because you have spent weeks teaching him that come can mean walk past me. The remedy is vocabulary discipline: come means arrive and stay in my area until released; let's go means move along with me.
people will use come like it's let's go and that messes up the recall right
Watch at 15:05 ↗ · Current transcript match · Audio confirmation pending
Quotation held: text matches, but timestamp or full-span transcript confidence still needs review.
Quotation held: text matches, but timestamp or full-span transcript confidence still needs review.
Zeus
Zeus's handover runs the same degradation argument from the other side, with the everyday scene spelled out: walking towards the door, saying come while you keep walking, and the dog passing you on his way inside - at which point come has started to mean 'walk by me', and the command you will need when another dog appears is gone. Two other protections are his. The equipment must never become the predictor: showing the dog the remote is banned outright, in unusually strong terms, because a threat display transfers the power from the handler to the object, and the collar is worn six to eight hours a day so that it predicts nothing. And the recall is deliberately kept from always meaning work: he is called back, given affection, and released from that spot without being asked for anything, so that coming in does not acquire a cost.
it's very important that you keep those two commands separate because this is what happens all the time people do this we're walking up close the door Zeus is over there right Zeus is over there walking up to the door says Zeus come and you're still walking and then he walks by you into the door and he didn't follow come at all
Watch at 11:20 ↗ · Current transcript match · Audio confirmation pending
Quotation held: text matches, but timestamp or full-span transcript confidence still needs review.
go like hey and they show them never I ban you from doing it please if I see you doing that or I hear of you doing that or you tell me you did that I'll be mad right I'll get on you for that because that will cause you problems
Watch at 63:07 ↗ · Current transcript match · Audio confirmation pending
so that way he doesn't think that every single time i call him back he's going to get a command like heel or sit or down or something else
Watch at 09:49 ↗ · Current transcript match · Audio confirmation pending
Kodak
Kodak's file supplies the rule that keeps a command from being spent and the one instance of a skill being stopped for the day on these grounds. The rule is stated twice, once in a session and once in a published rules video, in almost identical words: if you give the actual command you must reinforce it, it is not optional, and whatever the dog did to avoid it will be repeated for ever because it worked. The informal recall - "come here" rather than "come" - exists precisely so a command is never issued and then left hanging. And on a walk home the handler stops working the heel rather than pushing it, on the stated ground that pushing it here would poison it, and recaps it in a play session instead. In the same breath he spells the cue aloud, letter by letter, so the dog does not hear it.
And the reason that you might not give the actual command sometimes is because if you give the actual command, you must reinforce it. You must negatively reinforce it. It is not optional. You can never let commands become optional, right? Because whatever thing they did to avoid doing the command, they will repeat in the future right they will do it again and again and again and again because it works
Watch at 16:31 ↗ · Current transcript match · Audio confirmation pending
and they will think that is the solution to the pressure and you cannot allow that
Watch at 17:03 ↗ · Current transcript match · Audio confirmation pending
I'm not going to go for any more A-T-E-L because I'm going to start recapping that and pushing that again in another play session, not here on the walk. That'll probably just poison it for him.
Watch at 28:48 ↗ · Current transcript match · Audio confirmation pending
Roxy
The two-dog household gives the concept its most humane instance and its most important owner-facing rule. The instance: a radius correction lands far harder on the household's second dog than the handler expected, and within thirty seconds he names it as a negative experience, apologises to the dog by name, turns the level down and puts the collar away for a stretch of the session so the dog can cool down and have fun - then, later, considers that the reaction may have been confusion rather than pain and retests carefully rather than abandoning the tool. The rule is about the recall and it is absolute: you cannot be angry with her when she finally comes back, ever, because she will remember it and conclude that coming back is what went badly - punishing the return is punishing the right answer. A third session states a limit on the method itself: negative reinforcement applied to an unclear criterion, such as a vague heel, can cause real problems.
i will say that was kind of like a negative experience for him a little bit there i didn't know he was going to be so sensitive i'm sorry ramsay um so i have it turned down we're not going to use the e-collar for you know just a little bit during the session just just let him cool down and have some fun
Watch at 13:44 ↗ · Current transcript match · Audio confirmation pending
you, you cannot be mad at her when she gets to you ever. Um because if you are mad at her when she finally does come to you Yeah. she will remember that. She'll think I won't come back next time.
Watch at 15:26 ↗ · Current transcript match · Audio confirmation pending
Right. Um so if you punish her when she gets back, think of it as you are punishing her for doing the right thing. She finally came back.
Watch at 15:43 ↗ · Current transcript match · Audio confirmation pending
Quotation held: its full wording does not match the current transcript. Preserved-track provenance is recorded in the private audit.
Stormie
Stormie's sessions add the marker argument, the frequency argument and the clarity test. The marker argument is the clearest statement in the corpus of why the word matters more than the correction: the soft marker with the collar and the hard marker with the collar look identical from outside, one word apart, and to the dog they are two entirely different instructions - go back into what you were doing, or never do that again. The frequency argument is a trainer's own early failure, produced as evidence: a lab he kept drilling started running away from him and changed his whole body language, and what was wrong was not the command but how often it came. The clarity test is the practical guard against poisoning by misdiagnosis: when a command has gone backwards he stops saying the word entirely and supplies the position with the leash, which fixes a clarity problem on the spot and rules it out otherwise. And her down work carries the guardrail explicitly - she has to be pushed into the shape, so the drills are kept deliberately fun, and a repetition in the wrong shape is rejected so that down does not become the word for a belly rub.
and again if you say ant verse e-collar and then nope verse e-collar it looks exactly the same except minus like one word to the outside but to her brain she's taking the info as two very different things
Watch at 10:23 ↗ · Current transcript match · Audio confirmation pending
there was a lab I was working with like when I was first learning how to train dogs and he did really good and then I kept drilling commands all of a sudden he was like running away from me and like not listening and his whole body language changed and I found out that yeah it was like wasn't necessarily the command that was bad it was the frequency of it that was encouraging him to like repress himself
Watch at 47:15 ↗ · Current transcript match · Audio confirmation pending
So, this is something that I do sometimes is sometimes when the dog uh does a command and I teach a couple other commands. We come back to it like to heal in this case. Um I try not to assume that the dog is just like actively avoiding doing what I want him to do. So, what I'll do is I stop saying the word and I just start doing the position with the leash. And what that does for me is that if there was some kind of confusion, some kind of problem with the clarity, then that'll fix it right then and there.
Watch at 03:27 ↗ · Current transcript match · Audio confirmation pending
Quotation held: its full wording does not match the current transcript. Preserved-track provenance is recorded in the private audit.
Louie
Louie's file gives the concept a mechanism from the e-collar side and a rare admission from the marker side. The mechanism is about phases: the collar starts at a level that means only 'there is a way out of this and the leash will show you', and moving to a level that makes non-compliance costly before the dog knows the required response puts heavy pressure on a dog who does not know how to escape it - which is named as producing panic, confusion and suspicious associations. The admission is about the punisher word: the handler says plainly that Louie is not yet nervous of it, which means the word has not reached its full power, and that they will keep conditioning him - while adding, in the same passage, that he is not a huge fan of the word. And the ordinary calibration is here too: a correction that read as a little harsh gets answered by dropping to marker-only on the next repetition rather than by repeating it.
Quotation held: text matches, but timestamp or full-span transcript confidence still needs review.
Quotation held: text matches, but timestamp or full-span transcript confidence still needs review.
all right that's the first time he's responded to that so I'm only gonna do the actual like marker next time and if he backs off then we'll do it because I think that was a little harsh
Watch at 00:35 ↗ · Current transcript match · Audio confirmation pending
Sage
Sage's contribution is small and revealing: a junior handler, mid-session, hesitating to ask for a sit he might not be able to enforce, and saying why out loud - he does not want to poison the command. It is the only place in the corpus where the principle appears in the mouth of somebody clearly not senior, which is the evidence that it has been taught rather than improvised. The cue is low-confidence in the transcript and is recorded as such, but it is consistent with the surrounding exchange about collar levels on a different collar model.
Quotation held: text matches, but timestamp or full-span transcript confidence still needs review.
Source excerpts
Cuz if you say sit and you hit the e-collar, I know, bro, you're good. If I I say sit and then you hit the e-collar, that dog is going to go, "Oh, sit's a bad word."
Watch at 00:45 ↗ · Current transcript match · Audio confirmation pending
You're going to be walking and Deuce is going to be like trotting and you're going to say sit and he's going to be like, "Huh?" And he's just going to back up and be like, "Huh? What did I do wrong?" And that's one of the worst things
Watch at 00:57 ↗ · Current transcript match · Audio confirmation pending
then you either have to use a new word, which is kind of like awkward, or you have to like untrain that sit is a bad thing and then like retrain it to be what it's supposed to be
Watch at 01:06 ↗ · Current transcript match · Audio confirmation pending
it can take like three to four weeks. I'm going to say maybe two and a half to three weeks sometimes just to like unpoison the word.
Watch at 01:23 ↗ · Current transcript match · Audio confirmation pending
it's it's important that you use the right word. So, like some people their dogs will jump and they're like, "Down, down." And I'm like, "Down is like when you lay down."
Watch at 01:54 ↗ · Current transcript match · Audio confirmation pending
it's just not the idea is not even getting to them, right? So, if he jumps, like the nope is really important to use.
Watch at 02:09 ↗ · Current transcript match · Audio confirmation pending
so that's that's the main thing. It's just like knowing what you're doing, right? So, it's just important that it goes in that order. And I'll probably mention that. I probably won't mention it again today, but I'm going to mention it to you probably next session and then the go home day, too, because I That's one of the few things I'm like, "This is really important."
Watch at 00:28 ↗ · Current transcript match · Audio confirmation pending
I get rottweilers are interesting because they're tough They have a lot of like push put it but you can I find with like rottweilers just due to their temperament If you're not careful you can hand poison a command thinking that they're being tough because they have this like kind of look and they're very confident so when you see them like lock up sometimes you're like oh he's just being stubborn but genuinely they can be kind of confused
Watch at 01:56 ↗ · Current transcript match · Audio confirmation pending
Quotation held: its full wording does not match the current transcript. Preserved-track provenance is recorded in the private audit.
Quotation held: its full wording does not match the current transcript. Preserved-track provenance is recorded in the private audit.
Quotation held: its full wording does not match the current transcript. Preserved-track provenance is recorded in the private audit.
with it. So, nothing wrong with that. We just have to make sure that number one, I don't make her scared of the e-collar, so I got to be a little more gentle. And number two, make it really clear.
Watch at 00:46 ↗ · Current transcript match · Audio confirmation pending
do it. So, again, not driving her to the point where she's like, "I can't do it, right?" But, we're just applying some pressure. Cuz that is a healthy amount of pressure.
Watch at 02:18 ↗ · Current transcript match · Audio confirmation pending
minute here. We're not getting into bad territory yet. If I get if I keep hitting that e-collar. Hey, Duchess.
Watch at 12:40 ↗ · Current transcript match · Audio confirmation pending
Yeah. Okay. So, why don't I eat car there? Well, she gave me a face. She was looking around. I don't think she knew, honestly.
Watch at 14:09 ↗ · Current transcript match · Audio confirmation pending
But that's the two. and then you say the word and then the e-collar always comes after because if you say sit and you hit the e-collar after she doesn't listen what Luna's brain is going to go is oh man every time he says sit I get shocked which means sit is a punisher
Watch at 09:10 ↗ · Current transcript match · Audio confirmation pending
um and then you know that's classified as like poisoning a command and to unpoison a command can take 3 weeks and when you're program is four weeks. That kind of sucks, right?
Watch at 09:33 ↗ · Current transcript match · Audio confirmation pending
Source withheld by the current visibility or collection review.
Source withheld by the current visibility or collection review.
one more thing I want to go over is so I say, "Let's go." Each time she sees a dog and we condition eventually to the point where she sees a dog, she goes, "Oh, he's going to say, "Let's go." And if I don't listen, I get the e-collar. So, it looks like she just doesn't care about the dog. But in reality, she's actually predicting what's going to happen and just says it's not worth it.
Watch at 21:24 ↗ · Current transcript match · Audio confirmation pending
their tail's a little low right now so we need to wind down because she thinks out here means sucky experience every time it's not good
Watch at 10:48 ↗ · Current transcript match · Audio confirmation pending
we'll come back to the down a little bit later maybe she needs to open up a little more
Watch at 09:27 ↗ · Current transcript match · Audio confirmation pending
you see the tail is a lot more like she her tail kind of flops down but for a second it was like lower than usual so i wanted to make sure we open her back up
Watch at 11:45 ↗ · Current transcript match · Audio confirmation pending
right there, I'm a little careful because I'm really scared to poison any command with him, especially. Because he does seem to build, like, aversive behavior very quickly. which is like not great for training I mean it's not necessarily a bad thing but it doesn't mean you really can't allow for any sort of mistake
Watch at 11:12 ↗ · Current transcript match · Audio confirmation pending
What I'm doing is I'm not doing the butt push to teach the command. That won't teach you anything. What it does is that he does not like to be in that position at all.
Watch at 12:32 ↗ · Current transcript match · Audio confirmation pending
Quotation held: its full wording does not match the current transcript. Preserved-track provenance is recorded in the private audit.
Quotation held: text matches, but timestamp or full-span transcript confidence still needs review.
some people still were like I just like I was like show me how you do and they're like sit boom and I'm like yeah you just taught the dog the sit means you're getting punished which that's a really bad thing
Watch at 18:56 ↗ · Current transcript match · Audio confirmation pending
anytime you see a dog and you're like sit and they're like they go into like this thing of like I don't know what to do and like their their legs are shaking yeah it's not good
Watch at 19:04 ↗ · Current transcript match · Audio confirmation pending
first dog I ever trained actually when I was learning how to train I had a little dachshund and I I said sit and he like bolted away cuz I had like trained it the wrong way and I felt really really bad
Watch at 19:11 ↗ · Current transcript match · Audio confirmation pending
so right there I made a little mistake there okay what was my mistake well I kept saying the word over and over
Watch at 10:59 ↗ · Current transcript match · Audio confirmation pending
Quotation held: its full wording does not match the current transcript. Preserved-track provenance is recorded in the private audit.
Quotation held: its full wording does not match the current transcript. Preserved-track provenance is recorded in the private audit.
Quotation held: text matches, but timestamp or full-span transcript confidence still needs review.
don't punish because she didn't do anything to get punished for instead you just go like oh come on let's go if she then continues to fixate the last thing and go harder then you and punish
Watch at 37:56 ↗ · Current transcript match · Audio confirmation pending
He'll get over it. I won't press on that command again to prove to you that it's perfect, but he is doing well to sit.
Watch at 02:53 ↗ · Current transcript match · Audio confirmation pending
That was random. He did not seem super happy there to listen to it.
Watch at 02:30 ↗ · Current transcript match · Audio confirmation pending
These things happen sometimes even with with dogs that are doing a lot of the training.
Watch at 02:42 ↗ · Current transcript match · Audio confirmation pending
Come on. Alright, I might have poisoned that a little bit.
Watch at 07:28 ↗ · Current transcript match · Audio confirmation pending
I might have been a little bit too high on the criteria end there.
Watch at 07:47 ↗ · Current transcript match · Audio confirmation pending
Come here, dude. Yes. Reconciliation. Reconciliation.
Watch at 07:39 ↗ · Current transcript match · Audio confirmation pending
Okay, good boy. Good heel. Good job. Lower the criteria a little bit. Let's go. Come on, Bo.
Watch at 08:07 ↗ · Current transcript match · Audio confirmation pending
Quotation held: its full wording does not match the current transcript. Preserved-track provenance is recorded in the private audit.
Quotation held: text matches, but timestamp or full-span transcript confidence still needs review.
So we're going to be trying to implement some things to help her. to maybe not you know maybe trying to combine it with something enjoyable like a tug toy see if she can come around and uh start to associate it with something more positive
Watch at 00:26 ↗ · Current transcript match · Audio confirmation pending
so X I almost punished there you look kind of long to do it I didn't punish because you did it like the last I got like my finger was over the button I was like all right cool
Watch at 06:41 ↗ · Current transcript match · Audio confirmation pending
people will use come like it's let's go and that messes up the recall right
Watch at 15:05 ↗ · Current transcript match · Audio confirmation pending
Quotation held: text matches, but timestamp or full-span transcript confidence still needs review.
it's very important that you keep those two commands separate because this is what happens all the time people do this we're walking up close the door Zeus is over there right Zeus is over there walking up to the door says Zeus come and you're still walking and then he walks by you into the door and he didn't follow come at all
Watch at 11:20 ↗ · Current transcript match · Audio confirmation pending
Quotation held: text matches, but timestamp or full-span transcript confidence still needs review.
um so that's what can happen is a dog can become collar wise here's how it happens you're in here he's running around he doesn't have the collar on you guys are having a good time and then he doesn't listen to you and you go all right fine then i'm gonna go get your collar
Watch at 64:06 ↗ · Current transcript match · Audio confirmation pending
go like hey and they show them never I ban you from doing it please if I see you doing that or I hear of you doing that or you tell me you did that I'll be mad right I'll get on you for that because that will cause you problems
Watch at 63:07 ↗ · Current transcript match · Audio confirmation pending
so that way he doesn't think that every single time i call him back he's going to get a command like heel or sit or down or something else
Watch at 09:49 ↗ · Current transcript match · Audio confirmation pending
And the reason that you might not give the actual command sometimes is because if you give the actual command, you must reinforce it. You must negatively reinforce it. It is not optional. You can never let commands become optional, right? Because whatever thing they did to avoid doing the command, they will repeat in the future right they will do it again and again and again and again because it works
Watch at 16:31 ↗ · Current transcript match · Audio confirmation pending
And the reason that you might not give the actual command sometimes is because if you give the actual command, you must reinforce it. You must negatively reinforce it. It is not optional. you can never let commands become optional right
Watch at 4:24 ↗ · Current transcript match · Audio confirmation pending
I'm not going to go for any more A-T-E-L because I'm going to start recapping that and pushing that again in another play session, not here on the walk. That'll probably just poison it for him.
Watch at 28:48 ↗ · Current transcript match · Audio confirmation pending
i will say that was kind of like a negative experience for him a little bit there i didn't know he was going to be so sensitive i'm sorry ramsay um so i have it turned down we're not going to use the e-collar for you know just a little bit during the session just just let him cool down and have some fun
Watch at 13:44 ↗ · Current transcript match · Audio confirmation pending
maybe he understand because i will say dogs will freak out to the ecops they don't understand so he might have just been confused there we'll see if turning it back up yields good results i just want to be really careful here
Watch at 14:58 ↗ · Current transcript match · Audio confirmation pending
you, you cannot be mad at her when she gets to you ever. Um because if you are mad at her when she finally does come to you Yeah. she will remember that. She'll think I won't come back next time.
Watch at 15:26 ↗ · Current transcript match · Audio confirmation pending
Right. Um so if you punish her when she gets back, think of it as you are punishing her for doing the right thing. She finally came back.
Watch at 15:43 ↗ · Current transcript match · Audio confirmation pending
Quotation held: its full wording does not match the current transcript. Preserved-track provenance is recorded in the private audit.
Quotation held: text matches, but timestamp or full-span transcript confidence still needs review.
and again if you say ant verse e-collar and then nope verse e-collar it looks exactly the same except minus like one word to the outside but to her brain she's taking the info as two very different things
Watch at 10:23 ↗ · Current transcript match · Audio confirmation pending
It's either a, Oh, I need to go back into this. Like I messed up. Let me go back into this behavior or it means, Oh, that's a bad behavior. I can't do it ever again.
Watch at 10:37 ↗ · Current transcript match · Audio confirmation pending
the way it is you say sit the window is active and the window is closed once you give the release if you keep saying sit over and over you can train the dog that it's like they learn that oh i need to hear the word over and over and then that defeats everything that i taught them and makes your life harder
Watch at 40:23 ↗ · Current transcript match · Audio confirmation pending
there was a lab I was working with like when I was first learning how to train dogs and he did really good and then I kept drilling commands all of a sudden he was like running away from me and like not listening and his whole body language changed and I found out that yeah it was like wasn't necessarily the command that was bad it was the frequency of it that was encouraging him to like repress himself
Watch at 47:15 ↗ · Current transcript match · Audio confirmation pending
So, this is something that I do sometimes is sometimes when the dog uh does a command and I teach a couple other commands. We come back to it like to heal in this case. Um I try not to assume that the dog is just like actively avoiding doing what I want him to do. So, what I'll do is I stop saying the word and I just start doing the position with the leash. And what that does for me is that if there was some kind of confusion, some kind of problem with the clarity, then that'll fix it right then and there.
Watch at 03:27 ↗ · Current transcript match · Audio confirmation pending
Quotation held: its full wording does not match the current transcript. Preserved-track provenance is recorded in the private audit.
Quotation held: its full wording does not match the current transcript. Preserved-track provenance is recorded in the private audit.
Quotation held: text matches, but timestamp or full-span transcript confidence still needs review.
all right that's the first time he's responded to that so I'm only gonna do the actual like marker next time and if he backs off then we'll do it because I think that was a little harsh
Watch at 00:35 ↗ · Current transcript match · Audio confirmation pending
Quotation held: text matches, but timestamp or full-span transcript confidence still needs review.
How the trainers explain it
The stated mechanism is association and order. A word acquires the meaning of whatever reliably follows it, so the correction's position in the sequence decides what the dog learns: put it after the command and the dog learns that failing to comply produces the consequence; put it on the command and the dog learns that the command produces the consequence. Two handlers state that in almost the same words about two different dogs - "she can learn that down means the e-collar" (vRPhOZq5WJ4, 12:06) and "every time he says sit I get shocked which means sit is a punisher" (ldFe0hOs5NQ, 09:10). The observable failure state is described rather than theorised: a dog who hears the word and backs away asking what he did wrong (l3IAjG4Zk18, 00:57), or stands there with his legs shaking (7bEyPjj32bY, 19:04), or bolts (7bEyPjj32bY, 19:11). The repair cost is asserted rather than derived - three weeks, or two and a half to four - and is given the same figure independently in two dogs' files, which suggests a house number rather than an estimate. A second mechanism covers the version with no correction in it at all. A command word used as a general prohibition cannot deliver the prohibition, because the dog either performs the behaviour the word names or takes nothing from it: "it's just not the idea is not even getting to them" (l3IAjG4Zk18, 02:09). This is why the prohibition marker exists as a separate word, and why the marker choice is described as carrying the whole instruction while the correction looks identical from outside (60YpxVWYv20, 10:23). A third mechanism is about the unit. The association attaches to whatever is present and salient, not only to the cue - which is why a collar fired with no word attaches to the dog or person in view (biFPNDgXsQY, 37:14), why a trigger can be poisoned instead of a behaviour (_TVllc3GPIQ, 34:33), and why a whole outing can acquire the association if the session ends badly (FN9P4ETmTiQ, 10:48). A fourth is about dose rather than content: a command is not damaged by being corrected but by being asked too often, and the evidence offered is a trainer's own early case (A5tEoaHw99Y, 47:15). And a fifth, stated as a limit on the method itself, is that pressure applied to an unclear criterion is actively harmful rather than merely ineffective (wfSef-M7-vU, 22:21) - which is the same idea as poisoning by misdiagnosis, where a confused dog is read as stubborn and pressed (EL-La0eKp0Q, 01:56). Inferred, not stated: nobody explains why three weeks, why the repair is so much more expensive than the damage, or how a poisoned word is actually rehabilitated - the only method named for it is abandoning the word.
Differences to review
Whether a reactive dog's trigger may be corrected directly, or only her failure to redirect.
First perspective
Never correct the sighting. Hitting the collar every time she sees a dog teaches her that dogs mean pain, and a dog with nervous tendencies generalises that into dogs being dangerous. Give a generic redirect and correct the failure to move along; what that builds is prediction, and after roughly ten repetitions she performs it herself.
Source withheld by the current visibility or collection review.
Source withheld by the current visibility or collection review.
one more thing I want to go over is so I say, "Let's go." Each time she sees a dog and we condition eventually to the point where she sees a dog, she goes, "Oh, he's going to say, "Let's go." And if I don't listen, I get the e-collar. So, it looks like she just doesn't care about the dog. But in reality, she's actually predicting what's going to happen and just says it's not worth it.
Watch at 21:24 ↗ · Current transcript match · Audio confirmation pending
Second perspective
Correct the behaviour directly when it is large enough. A full bark-and-lunge is penalised on its own, with the collar and no command - and on another dog the protocol is a marker plus a relatively high tap the instant he moves toward another dog, optionally extended by thirty to forty seconds of social pressure, so that he concludes it was a terrible idea.
So like there's what he's describing is a difference between negative reinforcement and a penalty. Right? So one is pressure to do a command. The other one is a penalty for a bad thing they did.
Watch at 37:30 ↗ · Current transcript match · Audio confirmation pending
not going to pay but for something like that i want him to think oh my god that was the worst idea ever i never that was a terrible idea
Watch at 41:05 ↗ · Current transcript match · Audio confirmation pending
How to read this comparison
The two positions may be reconciled by the size of the behaviour - a look gets a redirect, a lunge gets a penalty - and Lola's lesson comes closest to saying so, drawing the line at whether the dog has actually done anything yet. But the Luna argument is stated as a general rule about nervous dogs and not as a rule about looks, and it names the exact protocol the other files run. A family who has watched both will not know which one applies to their dog, and the corpus contains no session in which the two are compared.
Whether a command word may ever be repeated.
First perspective
No. The window opens on the word and closes on the release; saying it again teaches the dog to wait for the repetitions and voids the implied stay.
Quotation held: its full wording does not match the current transcript. Preserved-track provenance is recorded in the private audit.
the way it is you say sit the window is active and the window is closed once you give the release if you keep saying sit over and over you can train the dog that it's like they learn that oh i need to hear the word over and over and then that defeats everything that i taught them and makes your life harder
Watch at 40:23 ↗ · Current transcript match · Audio confirmation pending
Second perspective
The rule against repeating a cue is dismissed as context-dependent - and then immediately qualified. The occasion is a handler with his arm inside a crate with a growling dog, being told to say sit as many times as it takes.
Quotation held: its full wording does not match the current transcript. Preserved-track provenance is recorded in the private audit.
Quotation held: its full wording does not match the current transcript. Preserved-track provenance is recorded in the private audit.
How to read this comparison
The self-qualification is what makes this worth recording rather than an outright contradiction: the same speaker says there are some rules and that contingency matters most of the time. But the two rules are given to two different audiences - the no-repeat rule to families at handovers, the context rule to a colleague in a crate-aggression session - and no transcript marks the difference. In practice the corpus is full of repeated cues inside a single attempt (eight sit cues in one public session at iUKpFfxMeGM 02:13), which suggests the rule is about repetitions rather than about syllables, but nobody says that either.
Whether the existing page's description of the concept still covers it.
First perspective
The published page (concept/dont-poison-the-command.html) is built on a single sighting: some dogs build aversive associations very quickly, so with them a command delivered carelessly can attach fear to the word, and a poisoned cue is far harder to repair than to avoid. Its source is named as Finn's Home Depot session, with Connor as the speaker and 'single sighting' as its stated confidence.
right there, I'm a little careful because I'm really scared to poison any command with him, especially. Because he does seem to build, like, aversive behavior very quickly. which is like not great for training I mean it's not necessarily a bad thing but it doesn't mean you really can't allow for any sort of mistake
Watch at 11:12 ↗ · Current transcript match · Audio confirmation pending
Second perspective
It is no longer a single sighting and it is no longer about sensitive dogs. Fifty-six sightings across fourteen dogs now stand behind it, including the mechanism taught to owners with a named repair cost (vRPhOZq5WJ4 12:06; ldFe0hOs5NQ 09:33), the commonest owner version, which involves no equipment at all (l3IAjG4Zk18 01:54), degradation by misuse of a word (0Ryn-sQhWK0 15:05; FIORySoF90o 11:44), a repetition not repeated after it frightened the dog (ya4jUrUhjtw 13:52), a skill stopped for the day on these grounds (BLUUidxbsnw 28:48), and the venue rather than the word as the thing at risk (FN9P4ETmTiQ 10:48). The page's own quote is also mildly misquoted: the page renders the last clause as "any sort of sloppiness" where the dump has "any sort of mistake".
it's it's important that you use the right word. So, like some people their dogs will jump and they're like, "Down, down." And I'm like, "Down is like when you lay down."
Watch at 01:54 ↗ · Current transcript match · Audio confirmation pending
their tail's a little low right now so we need to wind down because she thinks out here means sucky experience every time it's not good
Watch at 10:48 ↗ · Current transcript match · Audio confirmation pending
I'm not going to go for any more A-T-E-L because I'm going to start recapping that and pushing that again in another play session, not here on the walk. That'll probably just poison it for him.
Watch at 28:48 ↗ · Current transcript match · Audio confirmation pending
How to read this comparison
The page is superseded in scope rather than contradicted in substance, and its framing has a practical cost: an owner reading it concludes this is a risk for sensitive dogs specifically, when the record shows the commonest instance is an ordinary owner using 'down' to mean 'stop it'. The page's short description ('A fearful dog attaches aversion to a cue fast; sloppiness is not affordable') and its search terms are both keyed to the narrow case. Its quote should also be re-cut from the dump, since the version on the page does not match the transcript.
Whether a soured command is repaired by pairing it with something good, or only by abandoning the word.
First perspective
Repair it. A heel that has gone sour gets a session of its own: pair it with a tug, let her win some of the time, use a more encouraging tone, break the work up, and plan to test without the toy later. The stated principle is that letting her win something inside a disliked exercise changes how it feels to her.
So we're going to be trying to implement some things to help her. to maybe not you know maybe trying to combine it with something enjoyable like a tug toy see if she can come around and uh start to associate it with something more positive
Watch at 00:26 ↗ · Current transcript match · Audio confirmation pending
Honestly, even if it's just a little bit, I think it'll help just to make her feel like she's winning something.
Watch at 01:25 ↗ · Current transcript match · Audio confirmation pending
Second perspective
The only remedies named at the owner lessons are a new word or a long untrain-and-retrain, priced at two and a half to four weeks - and no session shows the untraining being done.
then you either have to use a new word, which is kind of like awkward, or you have to like untrain that sit is a bad thing and then like retrain it to be what it's supposed to be
Watch at 01:06 ↗ · Current transcript match · Audio confirmation pending
it can take like three to four weeks. I'm going to say maybe two and a half to three weeks sometimes just to like unpoison the word.
Watch at 01:23 ↗ · Current transcript match · Audio confirmation pending
How to read this comparison
Not a disagreement between people, but a gap: the corpus prices the repair twice and never demonstrates it. The nearest thing to a demonstration is Dolce's tug session, which is aimed at a command she dislikes rather than one that has been paired with a correction, and Bo's ninety-second reconciliation loop, which is aimed at a command the handler has just mildly damaged in the same session. Whether a genuinely poisoned word has ever been rehabilitated on film is not recoverable.
Questions and answers
Can a word really go bad, or is that just something trainers say?
The trainers use this term for a cue acquiring an unwanted association. Luna’s explanation discusses the possible work involved in repairing it (ldFe0hOs5NQ, 09:33), Deuce’s contrasts retraining with using a new word (l3IAjG4Zk18, 01:06), and Lola’s trainer recalls an early mistake (7bEyPjj32bY, 19:11). These are explanations and personal accounts. The mentioned repair durations are estimates from those speakers, not a measured timetable established by the recordings.
um and then you know that's classified as like poisoning a command and to unpoison a command can take 3 weeks and when you're program is four weeks. That kind of sucks, right?
Watch at 09:33 ↗ · Current transcript match · Audio confirmation pending
then you either have to use a new word, which is kind of like awkward, or you have to like untrain that sit is a bad thing and then like retrain it to be what it's supposed to be
Watch at 01:06 ↗ · Current transcript match · Audio confirmation pending
first dog I ever trained actually when I was learning how to train I had a little dachshund and I I said sit and he like bolted away cuz I had like trained it the wrong way and I felt really really bad
Watch at 19:11 ↗ · Current transcript match · Audio confirmation pending
What is the actual rule about the order of the words?
Command or marker first, correction second, always. The reasoning given is that a correction with nothing in front of it has no context, so the dog attaches it to whatever was most recently said - which is the command. Put the marker in front and the dog learns that failing to comply is what produces the consequence. In the first weeks home the marker is backed up every single time so that it means something; later the collar becomes a last resort. The analogy used with one family is the parent counting down - the reason a dog respects the warning is that the warning has always been true.
Quotation held: its full wording does not match the current transcript. Preserved-track provenance is recorded in the private audit.
Quotation held: its full wording does not match the current transcript. Preserved-track provenance is recorded in the private audit.
Quotation held: its full wording does not match the current transcript. Preserved-track provenance is recorded in the private audit.
My dog is jumping up. Why can't I just say 'down'?
Because down already means something to him, and this is named as the commonest owner mistake in the whole corpus. If he obeys, he lies down, which is not what you were asking for. If he does not, the message simply fails to arrive - "the idea is not even getting to them". Either way the thing you meant never reaches him, and the word gets used in a context where it never works, which is how a command loses its meaning without anybody correcting anything. This is the reason the method keeps a separate prohibition marker: the marker carries 'stop that', so the command words are left to mean positions. The same logic covers the other everyday collision, which is using the recall when you mean 'come along with me'.
But, uh just talking about like on the topic of communicating with the dogs, it's it's important that you use the right word. So, like some people their dogs will jump and they're like, "Down, down." And I'm like, "Down is like when you lay down."
Watch at 01:51 ↗ · Current transcript match · Audio confirmation pending
it's just not the idea is not even getting to them, right? So, if he jumps, like the nope is really important to use.
Watch at 02:09 ↗ · Current transcript match · Audio confirmation pending
Quotation held: text matches, but timestamp or full-span transcript confidence still needs review.
My dog is reactive to other dogs. Should I correct him when he sees one?
This is the one place where the corpus disagrees with itself, and you should know that before you choose. One trainer argues explicitly against it: hit the collar every time she sees a dog and what she learns is that dogs mean pain - and a dog with nervous tendencies goes from there to dogs being dangerous, which is manufacturing reactivity out of nervousness. His alternative is to correct the failure to move along, never the sighting, and he says what that builds instead: after roughly ten repetitions she performs the redirect herself, having worked out the consequence and decided it is not worth it, which from outside looks like a dog who does not care about other dogs. Other files in the corpus do correct the behaviour directly when it is large - a full lunge gets the collar on its own - and one lesson draws the line at whether the dog has done anything yet: a look that has not become anything is not punished at all. Nothing on film compares the two approaches.
Source withheld by the current visibility or collection review.
one more thing I want to go over is so I say, "Let's go." Each time she sees a dog and we condition eventually to the point where she sees a dog, she goes, "Oh, he's going to say, "Let's go." And if I don't listen, I get the e-collar. So, it looks like she just doesn't care about the dog. But in reality, she's actually predicting what's going to happen and just says it's not worth it.
Watch at 21:24 ↗ · Current transcript match · Audio confirmation pending
don't punish because she didn't do anything to get punished for instead you just go like oh come on let's go if she then continues to fixate the last thing and go harder then you and punish
Watch at 37:56 ↗ · Current transcript match · Audio confirmation pending
A repetition went badly and he seemed frightened. Should I try again so he doesn't learn he can get out of it?
Not that day, on the evidence here. When one dog slipped and fell on an obstacle the exercise was dropped on the spot, she was comforted, and there was no second attempt - with a camera running, which is the part that makes it worth quoting. In another session a sit arrived badly and unhappily in public, and the handler said he would not press the command again to prove it was perfect, because the proof exists in other videos and the repetition would have been for the audience rather than for the dog. Note that this sits alongside the opposite instruction elsewhere in the corpus - do not quit when a behaviour gets ugly as help is withdrawn - and the two are not reconciled. The distinction the sessions seem to draw, without stating it, is between a dog who is resisting and a dog who has been hurt or frightened.
Quotation held: its full wording does not match the current transcript. Preserved-track provenance is recorded in the private audit.
He'll get over it. I won't press on that command again to prove to you that it's perfect, but he is doing well to sit.
Watch at 02:53 ↗ · Current transcript match · Audio confirmation pending
Sometimes they can just feel a little strange.
Watch at 02:50 ↗ · Current transcript match · Audio confirmation pending
How do I tell whether he is refusing or confused, before I correct the wrong one?
There is a specific test on film, and it takes about thirty seconds. Stop saying the word altogether and simply put him into the position with the leash. If the problem was clarity, that fixes it on the spot; if he braces against it and plainly knows, you have your answer. The handler describing it says the point is not to assume the dog is actively avoiding what he wants - and he notes a cause most people would not consider, which is that teaching other commands in between can be what made this one go backwards. The reason to bother is that the cost is asymmetric. Correcting a dog who was refusing loses you a repetition; pressing a dog who was confused can cost you the word, and it is likeliest with a confident-looking dog whose confusion reads as toughness.
So, this is something that I do sometimes is sometimes when the dog uh does a command and I teach a couple other commands. We come back to it like to heal in this case. Um I try not to assume that the dog is just like actively avoiding doing what I want him to do. So, what I'll do is I stop saying the word and I just start doing the position with the leash. And what that does for me is that if there was some kind of confusion, some kind of problem with the clarity, then that'll fix it right then and there.
Watch at 03:27 ↗ · Current transcript match · Audio confirmation pending
I get rottweilers are interesting because they're tough They have a lot of like push put it but you can I find with like rottweilers just due to their temperament If you're not careful you can hand poison a command thinking that they're being tough because they have this like kind of look and they're very confident so when you see them like lock up sometimes you're like oh he's just being stubborn but genuinely they can be kind of confused
Watch at 01:56 ↗ · Current transcript match · Audio confirmation pending
Yeah. Okay. So, why don't I eat car there? Well, she gave me a face. She was looking around. I don't think she knew, honestly.
Watch at 14:09 ↗ · Current transcript match · Audio confirmation pending
Is there anything else that can go bad, apart from the word?
Three things, and the biggest of them is the outing. When a down cost one nervous dog badly, the risk the handler names is not that 'down' will go bad but that being outside will - she will conclude that out here means a bad experience every time - and for a dog whose entire treatment is being taken places, that is the failure that would undo the program. His protocol off the dropped tail is named and completed in one session: stop, walk, sniff, praise, and confirm the tail is back to normal before ending. The second is the trigger: a correction given with no command attaches to whatever the dog was looking at, which is a person or another dog. The third is the equipment - a dog who only sees the remote before a correction becomes collar wise, so showing it as a threat is banned outright and the collar is worn for hours a day so that it predicts nothing.
their tail's a little low right now so we need to wind down because she thinks out here means sucky experience every time it's not good
Watch at 10:48 ↗ · Current transcript match · Audio confirmation pending
Quotation held: text matches, but timestamp or full-span transcript confidence still needs review.
go like hey and they show them never I ban you from doing it please if I see you doing that or I hear of you doing that or you tell me you did that I'll be mad right I'll get on you for that because that will cause you problems
Watch at 63:07 ↗ · Current transcript match · Audio confirmation pending
Teaching outline
The handler can name the five ways a cue gets spoiled in this corpus, run a repetition in the correct order every time, run the clarity test before correcting an unclear failure, and stop a session on the right grounds rather than pushing through.
00:28
The order rule, the mechanism, the recognisable failure scene, and the repair cost, taught to an owner in under two minutes.
07:28
A trainer catching his own damage in a hard environment and running the complete repair loop on camera.
03:27
The clarity test: the word withdrawn, the position supplied by hand, and the diagnosis made before anything is corrected.
10:20
The tail read, the session wound down, and the outing re-sweetened before it ends.
Source reference held under the current eligibility review.
Run ten repetitions of one known command with a stopwatch on your own sequence. Command. Wait. If nothing comes, marker. Wait. Only then the leash or the collar. Say the word once per attempt and count how often you break that. Then run the clarity test twice on purpose: pick a command that has gone backwards, stop saying the word entirely, and supply the position with the leash - and note whether the problem disappears. Then audit your vocabulary out loud: what do you say when he jumps, when he is about to take something, when you want him to come along rather than come in? If a positional word appears in any of those answers, replace it with the prohibition marker. Then audit the end of sessions: pick a walk, read the tail at the halfway point, and if it has dropped, stop working, walk, sniff, praise, and finish on an easy repetition rather than on the hard one. Finally, practise not issuing a command: use the informal version - 'hey buddy, come here' - three times in a walk when you want the dog to move but are not prepared to enforce it, and say out loud afterwards that no command was disobeyed. (Deuce l3IAjG4Zk18; Dutchess vRPhOZq5WJ4; Stormie hc8gbZkTphM; Luna FN9P4ETmTiQ; Kodak acb6qFCsYww; Dolce Au7MBs3rYtY; Bo Jc0G4EtRIS8.)
Correction on the word rather than after the marker (7bEyPjj32bY 18:56).
A positional command borrowed to mean 'stop that' (l3IAjG4Zk18 01:54).
The recall spent when you are not prepared to enforce it (acb6qFCsYww 16:34).
Pressing a confused dog because his face reads as tough (EL-La0eKp0Q 01:56).
Repeating the rep that frightened him, to prove the command still works (iUKpFfxMeGM 02:53).
Correcting the sighting rather than the failure to redirect, on a dog with nervous tendencies (_TVllc3GPIQ 34:29).
Going to fetch the collar when he stops listening, or showing him the remote (KKFbjjFvhqU 64:06).
Ending an outing on the bad repetition rather than repairing it (FN9P4ETmTiQ 10:48).
Asking too often rather than too hard, and reading the resulting flatness as stubbornness (A5tEoaHw99Y 47:15).
Your dog is jumping on a visitor. You say 'down' twice, he keeps jumping, and you press the collar button on the second 'down'. Name the two separate pieces of damage you have just done, and say what you should have said instead. (First, you used a positional command as a prohibition, so the message never arrived; second, the correction landed on the word 'down', so 'down' now predicts the collar - and the repair is priced at about three weeks. The prohibition marker, then the collar behind it.)
Common mistakes
Saying the command and hitting the button in the same instant, so the dog learns that the word predicts the correction.
some people still were like I just like I was like show me how you do and they're like sit boom and I'm like yeah you just taught the dog the sit means you're getting punished which that's a really bad thing
Watch at 18:56 ↗ · Current transcript match · Audio confirmation pending
Cuz if you say sit and you hit the e-collar, I know, bro, you're good. If I I say sit and then you hit the e-collar, that dog is going to go, "Oh, sit's a bad word."
Watch at 00:45 ↗ · Current transcript match · Audio confirmation pending
Using a trained positional word as a general prohibition - shouting 'down' at a jumping dog, when down already means lie down.
it's it's important that you use the right word. So, like some people their dogs will jump and they're like, "Down, down." And I'm like, "Down is like when you lay down."
Watch at 01:54 ↗ · Current transcript match · Audio confirmation pending
it's just not the idea is not even getting to them, right? So, if he jumps, like the nope is really important to use.
Watch at 02:09 ↗ · Current transcript match · Audio confirmation pending
Using 'come' when you mean 'let's go' - saying it while you keep walking - until come means 'walk past me' and is unavailable when another dog appears.
people will use come like it's let's go and that messes up the recall right
Watch at 15:05 ↗ · Current transcript match · Audio confirmation pending
it's very important that you keep those two commands separate because this is what happens all the time people do this we're walking up close the door Zeus is over there right Zeus is over there walking up to the door says Zeus come and you're still walking and then he walks by you into the door and he didn't follow come at all
Watch at 11:20 ↗ · Current transcript match · Audio confirmation pending
Being angry with the dog when she finally does come back - which punishes the right answer, and she remembers it.
you, you cannot be mad at her when she gets to you ever. Um because if you are mad at her when she finally does come to you Yeah. she will remember that. She'll think I won't come back next time.
Watch at 15:26 ↗ · Current transcript match · Audio confirmation pending
Right. Um so if you punish her when she gets back, think of it as you are punishing her for doing the right thing. She finally came back.
Watch at 15:43 ↗ · Current transcript match · Audio confirmation pending
Repeating the cue until the dog answers, which teaches him to wait for the repetitions and voids the implied stay.
the way it is you say sit the window is active and the window is closed once you give the release if you keep saying sit over and over you can train the dog that it's like they learn that oh i need to hear the word over and over and then that defeats everything that i taught them and makes your life harder
Watch at 40:23 ↗ · Current transcript match · Audio confirmation pending
so right there I made a little mistake there okay what was my mistake well I kept saying the word over and over
Watch at 10:59 ↗ · Current transcript match · Audio confirmation pending
Giving a real command you are not prepared to enforce - which teaches the dog that commands are optional and makes whatever he did to get out of it his permanent answer to pressure.
You can never let commands become optional, right? Because whatever thing they did to avoid doing the command, they will repeat in the future right they will do it again and again and again and again because it works
Watch at 16:42 ↗ · Current transcript match · Audio confirmation pending
Reading a confused dog as a stubborn one and pressing harder - specifically likely with a confident-looking breed that locks up.
I get rottweilers are interesting because they're tough They have a lot of like push put it but you can I find with like rottweilers just due to their temperament If you're not careful you can hand poison a command thinking that they're being tough because they have this like kind of look and they're very confident so when you see them like lock up sometimes you're like oh he's just being stubborn but genuinely they can be kind of confused
Watch at 01:56 ↗ · Current transcript match · Audio confirmation pending
Running the repetition again after it hurt or frightened the dog, to prove the command works.
Quotation held: its full wording does not match the current transcript. Preserved-track provenance is recorded in the private audit.
He'll get over it. I won't press on that command again to prove to you that it's perfect, but he is doing well to sit.
Watch at 02:53 ↗ · Current transcript match · Audio confirmation pending
Showing the dog the remote as a threat, or going to fetch the collar when he stops listening - which transfers the power to the object and makes him collar wise.
um so that's what can happen is a dog can become collar wise here's how it happens you're in here he's running around he doesn't have the collar on you guys are having a good time and then he doesn't listen to you and you go all right fine then i'm gonna go get your collar
Watch at 64:06 ↗ · Current transcript match · Audio confirmation pending
go like hey and they show them never I ban you from doing it please if I see you doing that or I hear of you doing that or you tell me you did that I'll be mad right I'll get on you for that because that will cause you problems
Watch at 63:07 ↗ · Current transcript match · Audio confirmation pending
Making every recall the start of work, so that coming back reliably predicts being asked for something.
and see this time i'm making sure that he comes back to me and then i'm just going to release him from here after some loving oh boy right so that way he doesn't think that every single time i call him back he's going to get a command like heel or sit or down or something else
Watch at 09:38 ↗ · Current transcript match · Audio confirmation pending
Drilling - asking too often rather than too hard. The damage comes from frequency, and the tell is the dog's body language changing.
there was a lab I was working with like when I was first learning how to train dogs and he did really good and then I kept drilling commands all of a sudden he was like running away from me and like not listening and his whole body language changed and I found out that yeah it was like wasn't necessarily the command that was bad it was the frequency of it that was encouraging him to like repress himself
Watch at 47:15 ↗ · Current transcript match · Audio confirmation pending
Finishing an outing on a bad experience, so that the place rather than the word acquires the association.
their tail's a little low right now so we need to wind down because she thinks out here means sucky experience every time it's not good
Watch at 10:48 ↗ · Current transcript match · Audio confirmation pending
you see the tail is a lot more like she her tail kind of flops down but for a second it was like lower than usual so i wanted to make sure we open her back up
Watch at 11:45 ↗ · Current transcript match · Audio confirmation pending
What to notice
He hears a word he used to know and backs away, or looks at you as though he has done something wrong.
You're going to be walking and Deuce is going to be like trotting and you're going to say sit and he's going to be like, "Huh?" And he's just going to back up and be like, "Huh? What did I do wrong?" And that's one of the worst things
Watch at 00:57 ↗ · Current transcript match · Audio confirmation pending
You say the word and he freezes, unsure what to do, with his legs shaking.
anytime you see a dog and you're like sit and they're like they go into like this thing of like I don't know what to do and like their their legs are shaking yeah it's not good
Watch at 19:04 ↗ · Current transcript match · Audio confirmation pending
He runs from the collar pressure rather than answering it - which is a strategy, not evidence that it hurts, and it means nothing is being taught while it works.
of has a tendency where she'll just kind of try to blow me off or run away from the e-collar pressure, right? Um It does not mean that the e-collar pressure is hurting her, but her reaction to it is that I can get out of it if I run away. So, we're teaching her that it's not really a viable option.
Watch at 00:23 ↗ · Current transcript match · Audio confirmation pending
He comes when called and then loops straight back out to where he was.
Quotation held: text matches, but timestamp or full-span transcript confidence still needs review.
He has stopped coming back at all, or comes back slowly - and you can remember being cross with him the last time he did.
you, you cannot be mad at her when she gets to you ever. Um because if you are mad at her when she finally does come to you Yeah. she will remember that. She'll think I won't come back next time.
Watch at 15:26 ↗ · Current transcript match · Audio confirmation pending
He only listens once the collar is on, or starts behaving the moment you pick up the remote.
um so that's what can happen is a dog can become collar wise here's how it happens you're in here he's running around he doesn't have the collar on you guys are having a good time and then he doesn't listen to you and you go all right fine then i'm gonna go get your collar
Watch at 64:06 ↗ · Current transcript match · Audio confirmation pending
He has gone flat, or has started avoiding you, and nothing about your corrections has changed - only how often you are asking.
there was a lab I was working with like when I was first learning how to train dogs and he did really good and then I kept drilling commands all of a sudden he was like running away from me and like not listening and his whole body language changed and I found out that yeah it was like wasn't necessarily the command that was bad it was the frequency of it that was encouraging him to like repress himself
Watch at 47:15 ↗ · Current transcript match · Audio confirmation pending
His tail drops on a walk, or the sniffing stops, and you carry on working - the outing is what is going bad, not the command.
nothing less from a breed that is like you know shepherd husky right so if you look at the tail there that experience with the down she did not like that at all so we just need to walk around open her up a little bit
Watch at 10:20 ↗ · Current transcript match · Audio confirmation pending
He has got worse about other dogs since you started correcting him around them.
Source withheld by the current visibility or collection review.
He is confident-looking and locks up rather than complying, and you have been reading that as stubbornness.
hand poison a command thinking that they're being tough because they have this like kind of look and they're very confident so when you see them like lock up sometimes you're like oh he's just being stubborn but genuinely they can be kind of confused
Watch at 02:07 ↗ · Current transcript match · Audio confirmation pending
One particular command has got worse rather than better, while everything else improves.
Everything else is pretty much verbal. She'll do a place, she'll do a sit, she'll do pretty much anything else with verbal. It's just the heels are the only thing.
Watch at 04:07 ↗ · Current transcript match · Audio confirmation pending
do sometimes is sometimes when the dog uh does a command and I teach a couple other commands. We come back to it like to heal in this case.
Watch at 03:32 ↗ · Current transcript match · Audio confirmation pending
Trainer notes
Five distinct failure modes are recorded here and the corpus calls most of them poisoning. (1) Correction on the cue: the collar or the leash arrives on the word rather than after the marker, and the word becomes the predictor. Priced independently in two files at three weeks and at two and a half to four (vRPhOZq5WJ4 12:06; l3IAjG4Zk18 01:23; ldFe0hOs5NQ 09:33). Order is absolute: command or marker first, correction second, no exceptions stated. In the first weeks home the marker is backed every time so it carries; later the collar is a last resort (vRPhOZq5WJ4 12:47). (2) Overloading a word: a positional cue used as a general prohibition. Commonest owner failure in the corpus, and the reason a separate prohibition marker exists (l3IAjG4Zk18 01:54, 02:09). (3) Degradation by misuse: 'come' used for 'let's go' until come means walk past me (0Ryn-sQhWK0 15:05; FIORySoF90o 11:44). Related: never spend a command you are not willing to enforce - the informal recall exists for exactly this, and an unenforced command teaches that commands are optional and installs the dog's evasion permanently (acb6qFCsYww 16:34; Au7MBs3rYtY 31:28). (4) Damage by frequency rather than by intensity: the tell is body language changing and the dog avoiding the handler, and the first-hand case is a lab in one trainer's early career (A5tEoaHw99Y 47:15; and the dosing rule at 46:34). (5) The unit at risk being larger than the cue: a trigger poisoned instead of a behaviour (_TVllc3GPIQ 34:33), a correction with no command attaching to the person or dog in view (biFPNDgXsQY 37:14), a venue acquiring the association (FN9P4ETmTiQ 10:48), or the equipment becoming the predictor, which is collar wisdom - hence the ban on threat displays and the six-to-eight-hour wear (KKFbjjFvhqU 63:07, 64:06). Guards. Run the clarity test before correcting an unclear failure: withdraw the word, supply the position, read whether the problem disappears (hc8gbZkTphM 03:27). Monitor your own correction rate out loud and name the boundary before you cross it (lUcKg-GNhII 12:40). Do not drive a dog past the point where she can answer the pressure, and hold two constraints at once with a dog who runs from it - gentler and clearer (g8X4LPVegS4 00:46, 02:18). Do not move from escape to avoidance before the dog knows the response, or you produce panic, confusion and suspicious associations (G-c3ofWH0HA 17:42). Do not apply negative reinforcement to an unclear criterion; a vague heel is the named example (wfSef-M7-vU 22:21). Where a position has to be forced, pair the drills with deliberate fun and reject repetitions in the wrong shape (HdgdH84a8qs 07:30, 09:02). Withhold a correction when the behaviour did arrive, even slowly (KK6zS1srCOE 06:41). Soften the next repetition when one read too hard rather than repeating it (XMUMAaXKW5w 00:35). Repair, where it is shown: a break, a named reconciliation step, criteria lowered, and re-entry (Jc0G4EtRIS8 07:39-08:07); and for a merely disliked command, pairing with a game the dog can win (cp1qRyv1b2U 00:26). Variations by dog. A fast-aversion dog gets underplayed markers and no sloppiness at all (0JRh17nttVw 11:12; eypco6KEnXM 02:41). A confident-looking heavy breed gets the confusion reading first, because the toughness is a face (EL-La0eKp0Q 01:56). A sensitive dog may react to the collar out of confusion rather than pain, so the level is halved and retested rather than abandoned (4L4vqzKd0sM 14:58). Unresolved and to be settled before it comes up: whether a trigger may be corrected directly.
By trainer
Connor
Named on the existing concept page as the speaker for Finn's Home Depot session, which is where this concept was first recorded. His version treats the risk as the governing constraint on a fast-aversion dog rather than as a general caution, and he qualifies it immediately: not necessarily a bad thing, but it means no sloppiness is affordable. He also rejects physically placing the dog as a teaching method, on the ground that it teaches inevitability rather than meaning. The attribution comes from the page's source field, not from the audio.
right there, I'm a little careful because I'm really scared to poison any command with him, especially. Because he does seem to build, like, aversive behavior very quickly. which is like not great for training I mean it's not necessarily a bad thing but it doesn't mean you really can't allow for any sort of mistake
Watch at 11:12 ↗ · Current transcript match · Audio confirmation pending
What I'm doing is I'm not doing the butt push to teach the command. That won't teach you anything. What it does is that he does not like to be in that position at all.
Watch at 12:32 ↗ · Current transcript match · Audio confirmation pending
Christian
Named in the third person in two of the sessions behind this concept rather than heard teaching it. In Bo's crate-aggression session he is the handler reaching into the crate while the rule against repeating a cue is being dismissed as context-dependent; in Deuce's session he is named as the person who does most of the client assessments. Nothing in this concept's record is a quote from him, so his own version of it is not recoverable.
Christian's just going to take it and just kind of chill there and let him bite. Um, and, and not like take the hand away and, um, then, uh, go to close the crate and just kind of like, see if we can get him to normalize that way
Watch at 1:05 ↗ · Current transcript match · Audio confirmation pending
when you go and you assess the dog for the client which i don't do but i've been on a couple of them and like christian assesses them all or most of them anyways um we got kelton doing some of them too you don't always know like every single little thing they're gonna have because people just don't know
Watch at 12:01 ↗ · Current transcript match · Audio confirmation pending
Equipment context
E-collar
The tool most capable of poisoning a cue, and the one the order rule exists for: it comes after the word, never on it. Levels in the cited sessions run about 15 to 35, with one dog's radius correction halved to about 5 after it read too hard, and the top of the dial described as having been used once in a career. Three further protections: the level is chosen so the dog can still answer it; a dog who runs from it is being strategic rather than hurt, and the answer is gentler plus clearer rather than higher; and the remote is never shown and the collar is worn six to eight hours a day so it predicts nothing.
Quotation held: its full wording does not match the current transcript. Preserved-track provenance is recorded in the private audit.
i will say that was kind of like a negative experience for him a little bit there i didn't know he was going to be so sensitive i'm sorry ramsay um so i have it turned down we're not going to use the e-collar for you know just a little bit during the session just just let him cool down and have some fun
Watch at 13:44 ↗ · Current transcript match · Audio confirmation pending
um so that's what can happen is a dog can become collar wise here's how it happens you're in here he's running around he doesn't have the collar on you guys are having a good time and then he doesn't listen to you and you go all right fine then i'm gonna go get your collar
Watch at 64:06 ↗ · Current transcript match · Audio confirmation pending
The marker words - the soft reinforcement marker and 'nope'
The whole reason a command word never has to carry a prohibition. One marker means go back into what you were doing; the other means that behaviour is not allowed at all. Which one is used is chosen by how bad the behaviour was, and choosing the soft one after something serious reads to the dog as permission to escalate. On a fearful dog the markers are deliberately underplayed so the words themselves do not add to his nerves. One handler states plainly that a marker's weight is manufactured rather than intrinsic, and that he is not a fan of the punisher word even while continuing to condition it.
It's either a, Oh, I need to go back into this. Like I messed up. Let me go back into this behavior or it means, Oh, that's a bad behavior. I can't do it ever again.
Watch at 10:37 ↗ · Current transcript match · Audio confirmation pending
Quotation held: its full wording does not match the current transcript. Preserved-track provenance is recorded in the private audit.
Quotation held: text matches, but timestamp or full-span transcript confidence still needs review.
Leash - and specifically its length
Named as capable of teaching the wrong lesson on its own. A short leash held back is described as a way to manufacture reactivity, which is why store trips are done on a retractable line. Leash pressure is also the tool of the clarity test: the word is withdrawn and the position supplied by hand, which fixes a clarity problem on the spot. And a collar whose adjustment creeps looser under pressure is rejected, with the failure mode named.
Quotation held: text matches, but timestamp or full-span transcript confidence still needs review.
Quotation held: text matches, but timestamp or full-span transcript confidence still needs review.
Quotation held: text matches, but timestamp or full-span transcript confidence still needs review.
Tug toy, used as a repair
The one repair method demonstrated on film for a command that has gone sour. The heel is paired with a tug, she is allowed to win some of the time, the tone is deliberately more encouraging, the work is broken up rather than run back to back, and the plan is to test without the toy later. The stated principle is the small win: even a little of it helps, because it changes how the exercise feels to her.
So we're going to be trying to implement some things to help her. to maybe not you know maybe trying to combine it with something enjoyable like a tug toy see if she can come around and uh start to associate it with something more positive
Watch at 00:26 ↗ · Current transcript match · Audio confirmation pending
Honestly, even if it's just a little bit, I think it'll help just to make her feel like she's winning something.
Watch at 01:25 ↗ · Current transcript match · Audio confirmation pending
Quotation held: text matches, but timestamp or full-span transcript confidence still needs review.
Settings
Finn
Home Depot, where the handler names the poisoning risk as this dog's governing constraint
11:12
bo
Home Depot, where too many failed heels produce the damage and the repair
07:28
bo
a public access trip, where a bad sit is deliberately not repeated for the camera
02:53
Lola
the family's own home at the go-home lesson, where the failure state is described
19:04
Candidate clips
Cuz if you say sit and you hit the e-collar, I know, bro, you're good. If I I say sit and then you hit the e-collar, that dog is going to go, "Oh, sit's a bad word."
Watch at 00:45 ↗ · Current transcript match · Audio confirmation pending
Quotation held: text matches, but timestamp or full-span transcript confidence still needs review.
Come on. Alright, I might have poisoned that a little bit.
Watch at 07:28 ↗ · Current transcript match · Audio confirmation pending
some people still were like I just like I was like show me how you do and they're like sit boom and I'm like yeah you just taught the dog the sit means you're getting punished which that's a really bad thing
Watch at 18:56 ↗ · Current transcript match · Audio confirmation pending
Quotation held: text matches, but timestamp or full-span transcript confidence still needs review.
Source withheld by the current visibility or collection review.
it's very important that you keep those two commands separate because this is what happens all the time people do this we're walking up close the door Zeus is over there right Zeus is over there walking up to the door says Zeus come and you're still walking and then he walks by you into the door and he didn't follow come at all
Watch at 11:20 ↗ · Current transcript match · Audio confirmation pending
Quotation held: text matches, but timestamp or full-span transcript confidence still needs review.
Quotation held: its full wording does not match the current transcript. Preserved-track provenance is recorded in the private audit.
I'm not going to go for any more A-T-E-L because I'm going to start recapping that and pushing that again in another play session, not here on the walk. That'll probably just poison it for him.
Watch at 28:48 ↗ · Current transcript match · Audio confirmation pending
Quotation held: text matches, but timestamp or full-span transcript confidence still needs review.
there was a lab I was working with like when I was first learning how to train dogs and he did really good and then I kept drilling commands all of a sudden he was like running away from me and like not listening and his whole body language changed and I found out that yeah it was like wasn't necessarily the command that was bad it was the frequency of it that was encouraging him to like repress himself
Watch at 47:15 ↗ · Current transcript match · Audio confirmation pending
Quotation held: text matches, but timestamp or full-span transcript confidence still needs review.
Source moments
Source reference held under the current eligibility review.
Source reference held under the current eligibility review.
Case story beats
bo
In the middle of a busy store, after a run of failed heels, the trainer stops and charges the damage to himself out loud - and then runs the entire repair in ninety seconds: a break, a step he names reconciliation, the criteria lowered, and several clean heels. It is the method's error-correction loop happening in public, with nothing edited out.
Come on. Alright, I might have poisoned that a little bit.
Watch at 07:28 ↗ · Current transcript match · Audio confirmation pending
Come here, dude. Yes. Reconciliation. Reconciliation.
Watch at 07:39 ↗ · Current transcript match · Audio confirmation pending
Okay, good boy. Good heel. Good job. Lower the criteria a little bit. Let's go. Come on, Bo.
Watch at 08:07 ↗ · Current transcript match · Audio confirmation pending
Lola
At a go-home lesson the lead trainer describes asking owners to show him how they correct, watching them say sit and hit the button in the same instant, and what that produces - a dog standing there with its legs shaking. Then he says he did it himself, to a dachshund, the first dog he ever trained, and that he felt really bad about it.
some people still were like I just like I was like show me how you do and they're like sit boom and I'm like yeah you just taught the dog the sit means you're getting punished which that's a really bad thing
Watch at 18:56 ↗ · Current transcript match · Audio confirmation pending
first dog I ever trained actually when I was learning how to train I had a little dachshund and I I said sit and he like bolted away cuz I had like trained it the wrong way and I felt really really bad
Watch at 19:11 ↗ · Current transcript match · Audio confirmation pending
Luna
A down costs a nervous dog more than it was worth. The handler reads the tail, stops, and names the thing he is actually protecting - not the word, the whole outing, because a dog who concludes that being outside means a bad experience has lost the only treatment she had. Then he walks her, sniffs, praises, and checks the tail is back before ending.
nothing less from a breed that is like you know shepherd husky right so if you look at the tail there that experience with the down she did not like that at all so we just need to walk around open her up a little bit
Watch at 10:20 ↗ · Current transcript match · Audio confirmation pending
their tail's a little low right now so we need to wind down because she thinks out here means sucky experience every time it's not good
Watch at 10:48 ↗ · Current transcript match · Audio confirmation pending
you see the tail is a lot more like she her tail kind of flops down but for a second it was like lower than usual so i wanted to make sure we open her back up
Watch at 11:45 ↗ · Current transcript match · Audio confirmation pending
Dolce
A protection dog slips and falls on an obstacle with a camera running. There is no second attempt: the exercise is dropped, she is called in and comforted, and the handler asks her out loud whether the surface was too slippery for her. The demonstration lost, the dog kept.
Quotation held: its full wording does not match the current transcript. Preserved-track provenance is recorded in the private audit.
Stormie
A trainer teaching a family how to use an e-collar stops and tells them about a lab from early in his career: he did well, the trainer kept drilling, and then the dog started running away from him and his whole body language changed. The command was never the problem. The frequency was.
there was a lab I was working with like when I was first learning how to train dogs and he did really good and then I kept drilling commands all of a sudden he was like running away from me and like not listening and his whole body language changed and I found out that yeah it was like wasn't necessarily the command that was bad it was the frequency of it that was encouraging him to like repress himself
Watch at 47:15 ↗ · Current transcript match · Audio confirmation pending
Questions for intake
Is there a word your dog used to respond to and no longer does? Which one, and what changed around it?
What do you say when he jumps up, or when he is about to take something? (If the answer is a positional command, this concept applies before anything else does.)
Do you use your recall word when you want him to come all the way in, or also when you just want him to come along?
When he does not do it, what happens next - and in what order do you say the word, the telling-off and the correction?
Has he ever been corrected at the moment he saw another dog or a person, rather than for what he did next?
What happened the last time he finally came back after ignoring you - what did your voice sound like?
Has anything ever frightened or hurt him mid-exercise? What did you do immediately afterwards?
Does he behave differently when the collar is on, or when he sees you pick up the remote?
How many times a day do you ask him for commands, and has he become flatter or more avoidant as that went up?
Are there places he used to enjoy and now does not want to go?
Still to establish
Whether a nervous dog's trigger may ever be corrected directly. Luna's file argues that it manufactures reactivity; Zeus's and Dutchess's protocols do it. Nothing on film compares them, and the two instructions reach owners of comparable dogs.
How a poisoned command is actually repaired. The cost is priced twice, in two dogs' files, at three weeks and at two and a half to four - and no session shows the untraining being done. The only method named is abandoning the word.
Where the three-week figure comes from. It appears independently in two owner lessons, which suggests a house number rather than an estimate, but no source is given.
Whether the no-repeat rule is about syllables or about repetitions. Owners are told never to repeat a cue; the corpus is full of repeated cues inside single attempts, and one session dismisses the rule as bogus while immediately conceding that there are some rules.
How to distinguish a repetition that should be abandoned because it frightened the dog from a rough patch that should be worked through as help comes off. Both instructions exist, in different dogs' files, with no criterion between them.
The existing page's quote does not match the dump: the page has 'any sort of sloppiness' where the transcript has 'any sort of mistake'. The quote needs re-cutting before the page is regenerated.
Whether 'poisoning' as the trainers use it covers all five failure modes recorded here - correction on the word, word borrowed for a prohibition, degradation by misuse, damage by frequency, and a venue or trigger going bad - or whether the corpus is using one word for several different things.
Louie's handler says plainly that he is 'not a huge fan of' the punisher word while continuing to condition it. Nothing else in the corpus follows that up, and it is the only recorded reservation about the marker system itself.
Source limitations
Speaker identity is unrecoverable in nearly every session behind this concept. Only Connor is attributed, and that comes from the existing concept page's own source field for 0JRh17nttVw rather than from the audio.
Sage's 83w6Et4G-WE 09:53 is flagged low-confidence in the Tier 1 note. It is consistent with the surrounding exchange and is recorded as a cue rather than as evidence.
EL-La0eKp0Q 01:56 contains 'They have a lot of like push put it but you can', which is garbled. The sense of the passage rests on the later clauses, which are clean.
60YpxVWYv20 renders the soft marker as 'ant' and 'ash' at different points, and 'verse' for 'versus'. The marker words themselves are not recoverable as sounds from the transcript.
BLUUidxbsnw 28:48 spells the cue as 'A-T-E-L' because the handler is spelling it aloud so the dog does not hear it; the word is the heel cue.
KK6zS1srCOE 06:41 opens 'so X I almost punished there', where X is probably the dog's name mistranscribed.
Luna's trspGmYuPAw and several Kodak sessions run on the en-US caption track from an earlier Whisper pipeline; capitalisation is erratic and lines break mid-sentence.
4L4vqzKd0sM and wfSef-M7-vU are filed under roxy but the dog being handled in the cited moments is the household's second dog.
Of the 56 Tier 1 sightings behind this concept, 14 were set aside. The fourteen dogs here are the ones with narrated instances; the rule is stated at almost every owner lesson in the corpus, so its frequency is understated by the quote count.
Connected library collections
These links follow eligible video assignments in the reviewed library. A source can feature more than one animal; the older research-folder name is not used to merge identities.
Related concepts
- What do I do when my dog ignores a command?
- Is my dog confused or is he refusing?
- Should I repeat the command if my dog doesn't move?
- Does my dog actually know the command?
- How does my dog know when he can stop?
- What is the difference between come and let's go?
- What if correcting my dog doesn't work?
- What makes a correction fair?
- Whose fault is it when a command fails?
- My dog shuts down during training - what do I do?
- What are the words I actually have to remember?
- How do I train my dog without treats?