Field Notes · An evolving learning library from Angier K9

Cases · Case

Roman

Who Roman is

As the family describe him

Reactive, especially to dogs, and less so than he used to be.
On film

Well, and this is even less reactive than it was a year or two ago, but...

Watch at 0:51 ↗
A reputation for being scary, built almost entirely at the kennel front and on the lead. The owners say so themselves, and they name people staring at him through the kennel as the cause.
On film

So when exactly what when does when are people freaked out by him, what does he do? So it's almost all through the fence yeah and on the leash

Watch at 1:00:41 ↗

He can't go anywhere. He's trapped in there.

Watch at 17:17 ↗

So that's where he's gotten this big reputation of being this scary dog.

Watch at 17:25 ↗
No attack on a person. Bites on the family have been accidents, by their account. The owner also says she does not let anyone else handle him because it could go badly - so the reputation is partly theirs too.
On film

The bites with Roman have been accidents.

Watch at 5:10 ↗

Has he actually bitten people? No, by accident, he's bitten us. That's it, though? Yeah, that's right. He's not attacked anyone.

Watch at 18:30 ↗

and I don't let anybody handle Roman because it could go so badly.

Watch at 18:24 ↗
Growling at passing dogs from the start, from under a chair when they first got him. A dog raised him as a puppy (Elwyn, the household's neutral 'babysitter' dog), and he knows the house dogs.
On film

was when we first got him he i mean he's sitting underneath the chair not with that big and he's already growling at other dogs that are passing by him

Watch at 44:51 ↗

he we have one that pretty much roman was his puppy growing up and that's elwyn's our neutral he's like our babysitter dog

Watch at 57:07 ↗
A ball dog above everything. Daily fetch at the river where they used to live, a find-it game the owner invented there, and a habit of hiding the last ball in the woods when he is tired.
On film

pretty much he'll give it up he's always wanting to fetch when he we're living at the river pretty

Watch at 12:35 ↗

When we had him at the river, we'd play this at the field, find it, and I would make him sit

Watch at 27:22 ↗

he plays harder than most of ours do i mean he'll go till he's dropping especially if you've got him in the water

Watch at 1:01:26 ↗
A quick learner who has been easy to teach; the difficulty, in the owner's words, is two dogs and two hands.
On film

roman's a quick learner he's a he's a smart dog

Watch at 3:16 ↗

he's very i mean i haven't had any trouble with teaching him other stuff yeah it's just when they're two dogs and I've only got two hands I can only hold them so far apart.

Watch at 3:35 ↗

As the trainer describes him

Friendly, playful and neutral with a stranger who approaches without fear. The first thing the trainer establishes, with the leash dropped and then with the owner holding him back, is that the approach itself produces nothing but excitement.

On film

Yeah, hey dog, yeah, that's a pretty neutral interaction, I mean he's amped up because he wants to play but he's not He's not trying to do anything

Watch at 13:44 ↗

so that's interesting yeah so he's um so all he did just now it's not like a evil aggressive dog

Watch at 19:25 ↗

we were laying on the grass together petting i was petting him he's awesome

Watch at 48:54 ↗

The problem, narrowed at the second visit to three things: some leash reactivity, barrier aggression at crates and kennels, and reacting to dogs he cannot get to. Whether he would actually attack a dog, the trainer does not know and says so.

On film

like I said before Roman's a great dog I don't see anything actually wrong with him. He's really friendly for the most part He's got a little like bit of leash reactivity and some barrier aggression with crates on kennels and then he's got like that thing where he reacts to other dogs when he can't get to them but I don't know if he would actually attack a dog I'm not sure it seems unlikely to me but it's possible

Watch at 7:56 ↗
Exceptionally play-driven, and on that basis very trainable. The trainer's enthusiasm here is a claim about potential, not a result.
On film

His eyes dilate so crazy when he sees the ball come out.

Watch at 20:11 ↗

That dog, you could teach him to do basically anything you want very quickly. He could do obedience competitions and stuff like that pretty easily. Because any dog that's really, really driven for play, you can make their obedience not just functional. You could make it flashy if you want.

Watch at 20:27 ↗

get that off you that's good yeah man i can teach that dog almost anything

Watch at 9:33 ↗

The program, stage by stage

Evaluation at the owners' propertyTurning point

The trainer drops the leash to see what Roman does, walks around him, has an owner restrain him while he approaches (to separate restraint or leash frustration from aggression), and finds a neutral, playful dog. He tests toys: Roman gathers them and sits beside them but lets them be picked up and drops on cue, wants only balls that bounce or squeak, and does not always bring the ball back. The owner fits a prong collar; a first pressure test shows Roman braces against the leash, which the trainer reads as an oppositional reflex and names leash yielding as the first job. Most of the hour is the trainer explaining his method to the owners: evaluate, build cooperation through play, put rules on the game, teach with negative reinforcement, then create obligation - and, above all, a conditioned 'no'. He contrasts it with a previous trainer's sit-stay calm time, which he calls habituation that helps only where things are already under control. Verdict: a great dog, fixable in 'a few weeks'. No bite occurs.

On film

Because the main kind of thing with this is that I just want to see what he actually does.

Watch at 1:13 ↗

No biting has occurred sorry to disappoint you

Watch at 36:30 ↗

this uh you have a great dog seriously roman's a great dog there's nothing wrong with him that i can see um he's he's just had that bad experience when he was little

Watch at 1:01:36 ↗

Second visit: the fence, a 'nope', and obedience through the ball

Roman is walked along a fence toward another dog, Davey, and gets a 'nope' and a correction when he fixates. The trainer is candid that it was not done the way he would normally do it: the timing was off, the correction was not strong or novel enough (Roman went back to it), and 'nope' should have been taught first inside a game, not introduced at the trigger. He diagnoses the recall as not obligated. Then he plays ball with Roman and runs come, sit, down, out and place with the game as the reward, using the prong lightly. He explains a flexi-lead recall plan, the three Ds, proofing in places like Lowe's, and the release marker. His conclusion is the line the case is remembered by: Roman needs exactly one thing, a conditioned punisher, and he could build it in one session. The owners' account in the same visit is that the remaining work is other dogs. The rest of the recording is Egg's evaluation.

On film

Nope. Good. Good.

Watch at 2:02 ↗

that one it was like the event wasn't like the full punishment that I want because he did go back to it a little bit at first and so probably need to be novel

Watch at 5:16 ↗

I'll tell you he needs exactly one thing and like nothing else he just needs a condition Punisher that's literally it I could show I could I could do it for you I could do it for you in like like one session

Watch at 49:22 ↗

Skills

Toy possession and the out

At the startOwners report he gives tug items up and hides balls in the woods; the trainer tests for guarding.
By the endNot guarding: he sits by his toys but lets them be picked up and drops on request. By the second visit 'out' sits in a sequence of cues with the ball. Whether it is a clean out rather than a trade is not stated.

How it was taught

Tested rather than taught in the first visit: a ball thrown while the trainer approaches a second toy, a rope toy offered, drop-it exchanges. The trainer states the aim of an out without trading.

On film

walked back over dropped both of them dropped his ball dropped the ball i threw for him and sat next to the other toy and just kind of looked at me

Watch at 19:58 ↗

starts paying attention that drops the other one and it's just will he let you pick a toy up when he's sitting there yeah no problem yeah he'll drop it when i ask you

Watch at 20:27 ↗

let me see it let me see it hey drop it drop it good drop it good job drop it

Watch at 28:25 ↗

I'm going to see if we can get him to out the ball instead of trading.

Watch at 38:00 ↗

Roman, Roman, come, come, out, sit, down, down, go, go,

Watch at 25:53 ↗

Leash yielding

At the startWalked on slip leads, never a prong. Braces against the first pressure.
By the endDiagnosed and prescribed, not taught on camera. In the second visit the prong is used lightly for the down, which is the only filmed use of leash pressure on him.

How it was taught

Not taught. The method is described: feel the dog's first response to pressure, teach him to give to it, so that later pressure can stay small.

On film

Have you guys used prong collars at all? We have not. We've mostly used the slit leaves.

Watch at 39:29 ↗

Lee shielding is you just do this he feels it see what's his first response bracing

Watch at 50:24 ↗

same thing they pull against pressure and so that's what he's doing he's just doing an oppositional reflex he's pulling against the pressure you teach them to yield to the leash and then everything gets easier

Watch at 50:47 ↗

You're just like, you saw what I was doing with him before. It's like, it's like literally like, I'm just babying the crap out of it.

Watch at 35:50 ↗

Come, sit, down and place through the ball game

At the startHe already knows some commands (the owners have taught him; the trainer says he has 'a little bit of training on him'). The recall comes when he wants it to.
By the endRepeated come, sit, down and place with the game resuming as the reward, in the second visit.

How it was taught

Cue, then the game comes back as the reward; a light prong prompt for the down.

On film

roman down down all the way down good job dude

Watch at 14:46 ↗

Place.

Watch at 17:26 ↗

So you get the game back. You get the game back.

Watch at 19:43 ↗

he's actually a really like good dog and he's already got a little bit of training on him

Watch at 48:59 ↗

Recall

At the startNot obligated: he comes if he wants to, and is suspicious of a random 'come' (is he being put in the car?). He does not always bring the ball back.
By the endA plan, explained and not practised on camera: a 26-foot flexi on a prong, cue come when he is distracted, the lightest pop that works, and release him back to the ball or to the spot he was sniffing.

How it was taught

Not taught on camera.

On film

Roman come here here Roman and so his recalls not obligated it's like if he wants it then he'll come

Watch at 6:21 ↗

sometimes a random Roman come here he's like you're worried what you got for me are you putting me in a car

Watch at 6:34 ↗

yeah that's real good especially since you can toss the ball yeah yeah see he doesn't like to bring it back i see not all the time not all the time

Watch at 1:02:24 ↗

But, like, I'd normally use, like, the 26 on a dog like him.

Watch at 31:54 ↗

can either with him you can throw a ball and then he goes and get the ball that's his reward or you could just be like okay and then like encourage him to go back out to sniff the spot he was sniffing

Watch at 32:52 ↗

The conditioned 'no' (a conditioned punisher)

At the startNone. The owners have no word that stops him; the trainer says that is why they cannot call him off when a loose dog runs at him.
By the endOne 'nope' with a correction at the fence in the second visit, which the trainer himself says was mistimed, not strong or novel enough, and introduced in the wrong context. The one-session build he offers is not filmed.

How it was taught

Described, not built: teach the word inside a game and in obedience first, pair it with a meaningful aversive (prong or e-collar) so it transfers to the voice, then use it in the situations it must generalise to.

On film

and you get that by using a conditioned punisher because it transfers whatever punisher you use prong collar e-collar to your voice it's like pavlov's dogs right

Watch at 56:30 ↗

like in a typical situation I would not introduce the like word nope with the situation like triggering situation I introduce it in in the game in obedience within a game and then he learns okay nope means cut it out right

Watch at 4:08 ↗

that one it was like the event wasn't like the full punishment that I want because he did go back to it a little bit at first and so probably need to be novel

Watch at 5:16 ↗

Release marker

At the startNo formal release; the owner uses 'let's go'. Sit-stays were better when they lived at the river.
By the endExplained ('okay', charged until the dog gets zoomies); not built on camera with Roman.

How it was taught

Not taught.

On film

So does he have a release marker?

Watch at 51:55 ↗

Yeah, I don't really have a word.

Watch at 51:59 ↗

hand i'll step off say let's go and then he's gotcha

Watch at 52:14 ↗

because ideally if you if you condition the release marker correctly when you say okay my favorite thing to see is that the dog gets zoomies

Watch at 37:17 ↗

How the method adapted

With two visits, 'adaptation' mostly means what the trainer chose not to do with this dog and why, plus one self-correction. The choices are consistent: lowest conflict, play first, and pressure kept small.

The evaluation was run in the open with the leash dropped, and then the restraint was added deliberately, to find out whether the trouble was the dog or the lead.

Because the dog's reputation came from the kennel front and the lead, and the trainer thinks kennel-front evaluations measure the kennel. Holding him back as the trainer walks up separates restraint frustration from aggression.

On film

hold him back as I walk towards you and see if that is the thing that is an issue because that might just be restraint and like leash like slash

Watch at 13:20 ↗

he wants to hurt people and it's like that doesn't that's not that's like if you evaluated me when i had a migraine

Watch at 16:43 ↗

The owner, not the trainer, put the prong on.

The owner says she is always the one who puts collars on him. The trainer thinks Roman would have let him, but would not risk a conflict he did not need.

On film

How does he do with collars going on and off? Pretty good, but I'm always the one putting them on to him.

Watch at 38:40 ↗

I'm gonna let you put it on him for now. I think that he'd let me put it on him, but I don't want to create any kind of conflict unnecessarily, if that makes sense.

Watch at 39:19 ↗

The reward followed Roman's preference, not the kit. He ignored the rope toy, so the plan became to put his kind of ball on a rope so it could be used for tug games with rules.

He only wanted a ball that bounces or squeaks, and the trainer wants a toy he can hold on to so the game can have rules.

On film

I threw that and he only wants the ball.

Watch at 22:03 ↗

He's like, uh-uh, I know what that is, it's not a ball. It doesn't bounce.

Watch at 31:42 ↗

It would be also nice to play like a competitive game. Then you could start putting like rules on it. So the way we do it is you build a game where you start putting rules on it, and that's how you train them.

Watch at 27:10 ↗

The trainer corrected himself on camera about the fence work: he introduced 'nope' at the trigger, which he says he would not normally do.

His stated sequence is word first inside a game, then the trigger. He had gone straight to the fence, and says he would time it better, use the 'nope' first, and make the consequence novel next time.

On film

and then I would time it a little bit better than I did and make sure that I use the nope first

Watch at 4:01 ↗

like in a typical situation I would not introduce the like word nope with the situation like triggering situation I introduce it in in the game in obedience within a game and then he learns okay nope means cut it out right

Watch at 4:08 ↗

Pressure on Roman was kept deliberately light, with a promise to go only as high as each place requires.

The trainer's rule is that the level of negative reinforcement rises with the distraction, place by place; at home, with a dog who wants the ball this much, very little is needed.

On film

You're just like, you saw what I was doing with him before. It's like, it's like literally like, I'm just babying the crap out of it.

Watch at 35:50 ↗

Well, maybe over there, it's like, hey, down. And he doesn't do it. And then I got to go a little harder. And then he goes into the down. And then he learns, oh, wait, it's obligated here too.

Watch at 36:01 ↗

What the family was told

Fear makes the dog it expects: how people approach him matters

or maybe they've been told that he's aggressive and so then they're afraid of him and so they walk up with that fear and he feels it and then he goes what's wrong with you

Watch at 14:30 ↗

and then it creates the scenario it's like a self-fulfilling prophecy

Watch at 14:42 ↗

A lot of, like, her dad will have the issue of, like, looking straight at a dog and going, oh, this dog doesn't like me. And it's like, no, because it looks like you're staring him down.

Watch at 1:00:55 ↗

Neutral is not the same as under pressure

between that and putting pressure on a dog right so like i might be able to walk up neutral and not do anything and he's like okay but if i put a prong collar on and was like sit he might be like oh i'm not doing that i'm gonna bite you

Watch at 14:56 ↗

the behavior you want to see is that they go okay let's work let's cooperate

Watch at 15:41 ↗

The rubric: evaluate, play, rules on the game, then teach with pressure and release

So this is what we're doing now. We're evaluating the dog. We're saying, okay, what is this dog actually about? You let them do what they're actually going to do and not like an artificial thing, right?

Watch at 16:02 ↗

but then you show them how to get rid of the pressure so you just go like this you don't have to like or like step hard on the leash and like try to stomp them into the ground that's just going to create extra stress and then shut them down and they don't want to play with you

Watch at 47:08 ↗

and then you show them how to get rid of it and as soon as they hit that position close to the one he's in now where they're in a down then you go okay and that's their release word and then they go and then you're like boom whip out the toy

Watch at 47:32 ↗

In dog training, play is God. Like, that is, like, the thing that makes everything else possible. If you can get a dog to play, you won.

Watch at 1:01:19 ↗

Obligation, and why the e-collar

so if you want to create obligation the first 80 percent you do all positive reinforcement is fun right

Watch at 40:31 ↗

because you can't always beat the environment squirrel versus hi dad's here squirrel every time right

Watch at 40:43 ↗

used to use obligation the problem is all of those are blunt instruments they're not nuanced

Watch at 41:35 ↗

don't care if they listen to the beep I want them to listen to me so then my vocal is the the warning

Watch at 42:41 ↗

none of it can be optional because if it's optional then you might you then you don't get it when you need it the most

Watch at 6:53 ↗

because if you make it non-optional in a way that's not not clear it's unclear to the dog or you make it non-optional in a way that's too hard on the dog or you make it non-optional in a way that's way too soft on the dog you either get really bad fallout and they develop weird behaviors

Watch at 7:07 ↗

Slip lead versus prong

with the slip lead because you're going to hurt him with the slip lead

Watch at 49:13 ↗

either put way too much pressure on them because you're trying to go too fast and not showing them how to respond to the pressure and so then there's just lots of pressure and it just induces panic

Watch at 51:06 ↗

What the previous trainer's calm time can and cannot do

so positive counter conditioning can like get you to a certain place and that's what she was having you do that's uh like it's it's almost like habituation right and it does have a lot of value it can work really well um but it gets you to a point and then not further

Watch at 53:50 ↗

habituation they get used to like oh there's other dogs around the people are nice to me things are cool like it's not that bad right but then that doesn't help really at all in the the situation i described over here

Watch at 55:20 ↗

He does not have to like other dogs; he does have to be callable off them

romans never like oh i want to be best buds with all the dogs right turns into like your stereotypical like golden retriever that might not ever be roman right yeah that's okay yeah that's fine right

Watch at 56:40 ↗

A dog having a conditioned punisher, like a conditioned word that means, hey, don't do that, and I mean it, is one of the most effective possible things for safety reasons.

Watch at 58:19 ↗

I would never want to send a dog home to anybody without it.

Watch at 58:34 ↗

but all the condition Punisher is is it's um it's a word that like most people you just use no or nope or something like that and that just means don't do that that's it

Watch at 49:32 ↗

Forbid the sliver, allow everything else

i try to make sure that it's clear in their minds like everything else is fine you can do lots of stuff right you just can't do this one little sliver

Watch at 11:48 ↗

with a dog then you have real clarity and dogs thrive with like real clarity real consequences that make sense to them that are predictable and that are under their control because then it's contingent

Watch at 13:38 ↗

Recall on a flexi, then proofing by the three Ds, then real places

the problem is that like with that long line you see all the like line management I have to do

Watch at 31:15 ↗

And then the best way to do it is you take him on a walk where there's a bunch of really interesting stuff, and you wait until they're really distracted, and then you tell them come.

Watch at 32:01 ↗

the flexi on a prong collar and you just kind of like pop it and just like slight and if that's all that it needs then that's cool

Watch at 32:34 ↗

have you ever heard of the three d's distraction duration and distance

Watch at 34:02 ↗

for me the last hurrah for most dogs like the last thing that is like okay we know we're good is going fully off leash in public

Watch at 38:17 ↗
In the family's own words

When he saw the girls on the outside of the fence, he started to be fussing to them, and then when they didn't react, he got over pretty quickly.

Watch at 17:38 ↗

we would do like the other trainer with his trying to get him less reactive we just have him in a sit stay yeah and just chill for 20 minutes slack leash

Watch at 52:35 ↗

things from happening yeah well that's what only one morning i could jump over his fence and i had roman on leash didn't realize the other dog was out

Watch at 58:43 ↗

yeah so normally he knows that we're pretty much his escape route if he freaks out

Watch at 5:29 ↗

so it's like mostly just with other dogs that he kind of needs to like learn

Watch at 29:55 ↗

Outcomes

What the trainers reported

A great dog with nothing wrong beyond a bad early experience; fixable 'in a few weeks'. Note that the words after 'nothing wrong with him that i can see' run into what may be an owner speaking ('we've done a lot of the training'), so the 'bad experience when he was little' clause may be the owner's, not the trainer's.

The trainer, end of the first visit

On film

this uh you have a great dog seriously roman's a great dog there's nothing wrong with him that i can see um he's he's just had that bad experience when he was little

Watch at 1:01:36 ↗

a lot of the training yeah not sure about it how strong his heel is you can you can fix this stuff in a few minutes like in a few weeks like you can fix this to be really great

Watch at 1:01:47 ↗

The fence problem could probably be eliminated in a session or two, through play and obligated, positively reinforced obedience.

The trainer, second visit

On film

yeah I mean we could probably eliminate it entirely very quickly maybe even within just like a session or two um just by um doing play with him uh sharpening up his obedience and making it where it's obligated and positively reinforced

Watch at 5:48 ↗

He needs exactly one thing, a conditioned punisher, buildable in one session; after one session of new teaching with some negative reinforcement, the trainer would be confident taking him to Lowe's or a park.

The trainer, second visit

On film

I'll tell you he needs exactly one thing and like nothing else he just needs a condition Punisher that's literally it I could show I could I could do it for you I could do it for you in like like one session

Watch at 49:22 ↗

I mean yeah I mean I'd be confident to take him out to Lowe's and to like a park or something like that if I just had a like a session with him where I just got to teach him something brand new that makes sense where I got to use some negative reinforcement and and establish just a little bit more authority, then he could go to Lowe's no problem

Watch at 30:29 ↗

The conditioned punisher is 'not complicated' to get.

The trainer, first visit

On film

and that's something that's not complicated to get with a dog you can get it

Watch at 59:24 ↗

Less reactive than a year or two ago; never attacked anyone; a quick learner.

The owners

On film

Well, and this is even less reactive than it was a year or two ago, but...

Watch at 0:51 ↗

Has he actually bitten people? No, by accident, he's bitten us. That's it, though? Yeah, that's right. He's not attacked anyone.

Watch at 18:30 ↗

roman's a quick learner he's a he's a smart dog

Watch at 3:16 ↗

What the footage shows

A neutral, excited response to a stranger approaching, including while restrained on the lead by his owner; no bite in either visit.

On film

Yeah, hey dog, yeah, that's a pretty neutral interaction, I mean he's amped up because he wants to play but he's not He's not trying to do anything

Watch at 13:44 ↗

No biting has occurred sorry to disappoint you

Watch at 36:30 ↗

Toys given up on request; no guarding seen by the trainer.

On film

I'm not seeing huge issues here with this kind of thing

Watch at 29:05 ↗

Bracing against leash pressure on a prong at first fitting.

On film

same thing they pull against pressure and so that's what he's doing he's just doing an oppositional reflex he's pulling against the pressure

Watch at 50:47 ↗

Fixation at the fence toward another dog, a 'nope' with a correction, and a return to it - by the trainer's own account.

On film

Nope. Good. Good.

Watch at 2:02 ↗

that one it was like the event wasn't like the full punishment that I want because he did go back to it a little bit at first

Watch at 5:16 ↗

Many repetitions of come, sit, down, out and place in a ball game, and a dog relaxed enough afterwards to lie in the grass being petted. The audio cannot show how clean the repetitions were.

On film

So you get the game back. You get the game back.

Watch at 19:43 ↗

yeah my dogs well Romans on sleepy time now he was just he's just been laying down

Watch at 24:59 ↗

we were laying on the grass together petting i was petting him he's awesome

Watch at 48:54 ↗

The ideas Roman's case runs on