Aidan3

Good point.

When your dog "franticly" (or not) offers behaviours, it would be because the dog has learnt that ONE behaviour just might get the reward. Correct? So, with the dog expecting and knowing the reward is on offer, and we don't give it (because the dog hasn't offered the behaviour we, in our minds, want), would you regard that as negative punishment? No, by definition and by function it would be an extinction procedure and not negative punishment.

You gotta watch those locals! Your guys did well though, that's something to be happy about. I went for a walk in the bush at 6:00am on Sunday morning, everything white with frost, must be the time the locals let their out of control dogs off the leash. One dog kept having a go at us for about 20 minutes, I found the owners back at the car park. Talk about lazy! Why walk your dog when you can just let it off at the car park to chase wallabies and harass other dogs?

Would you always make that conclusion? Frantic offering of already learned behaviours is not necessarily a sign of stress. true but it can be stress as well So long as we're clear that not all frantic offering of behaviour is a sign of stress. I wouldn't want anyone thinking they were stressing their dog out just because they weren't the best free-shaper in the world, or that a little of this isn't completely normal.

I was talking specifically about teaching targeting so I'm a little confused about your reply? Are you concerned that using a target might cause frustration or result in unreliable responses? A target is usually only used early on so it won't bring reliability (only reinforcement of the end result will do that), and in my experience a target is possibly the least frustrating way to teach a new behaviour possible - no guesswork, no force, clear and direct communication, and no focus on the food.

Would you always make that conclusion? Frantic offering of already learned behaviours is not necessarily a sign of stress.

I've seen worse room trashings from dogs who didn't have SA if it makes you feel any better :-) Go to www.clickertraining.com and search for my article on Separation Anxiety. In it is an exercise which you can begin right away, "The Calming Yo Yo". Lots of good feedback on this exercise but not many people will ever actually do it, probably because it doesn't make a lot of sense to anyone but applied behaviour analysts...

My recommendation for using a head halter is to invest in a "balance leash" which is a leash with a clip at either end. One end is clipped to the head halter, the other the flat collar. That way you only "use" the head halter when you really need to, it is also safer, and much easier to wean onto a flat collar when you are ready. When you are ready, you unclip the head halter but still leave it on the dog for a couple of weeks when you walk.

Fascinating, entertaining, waffle.

I'm not often openly critical of people, but psychologytoday.com lost all credibility with me when they gave LCK a platform and unfortunately they lend credibility to LCK that is not deserved. He is a writer of fiction, and presenting fiction as non-fiction in what might be regarded by many as a credible publication only serves to confuse the reader.

It's stupid, but I just couldn't figure out what I was doing wrong with my password on my laptop. I made both logins substantially similar, no deception. Although if I ever say anything really stupid I'll blame it on the other Aidan.

The dog in the clip spends a good amount of time watching my face, but if you notice I also click deliberately for looking in other places as well. If I were teaching heeling and eye contact was an early criterion I'd be clicking that eye contact over and over and over. You get what you click and eye contact would be stronger than following the food in about 30 seconds. Even despite the very high rate of reinforcement used in this early session the dog does not spend a whole lot of time following my hand because I click for other things. What anyone watching this clip needs to keep in mind is that we have 1:30 of footage, and in that time, despite not lowering the rate of reinforcement and having the treat bag on my left side this dog has already learned stuff that isn't "follow my hand around". She's learned to come back from a tight leash, and a reasonable proximity for casual loose leash walking. She is free to sniff, free to look at anything she likes, and is doing this with another dog who is, at most, 10' away who she is clearly very interested in. So with all this freedom, she is still spending far less time trying to check out the other dog than she had been prior to this session. There is a lot of information for the dog there. A ton of things get clicked. Dogs are smart, seriously smart.

I don't see any examples of food being deliberately used as a lure (although she does become interested in the food in my hand at times), but I do see change in the dog's behaviour. At the beginning she doesn't have much clue about where she should be, at the end she has a very good idea. The response I was looking for was being increased through +R, i.e we got more of the response we were reinforcing. Every time she got to the end of the leash she brought herself back. In the early stages I don't really care where they are so much so long as they learn very strongly to come back from a tight leash without thinking about it. Believe me, Tess had had no shortage of more traditional training before this. This is Tess' third 10 minute session. The camera person has a small dog, Tess is dog aggressive and has picked up and shaken a small dog before. She actually looks pretty happy to see the dog in this clip, not threatened at all - and why would she? It's a positive, non-threatening environment with only good outcomes. Pavlov is always on your shoulder. What you need to take note of is what exactly is the dog doing at the time of the click? At first it looks like I'm teaching her to sniff the ground! She figures out pretty quickly I'm clicking for something else. She starts to prance there for a bit, so I click when her feet hit the ground. If you're not paying attention it looks like I'm rewarding her for prancing, but the clicker makes it very clear to Tess, it happened precisely when her feet hit the ground. I click for her head in all sorts of positions, she gets clicked for eye contact a lot, but I also deliberately click when she's in position but not looking at me. This is not heeling, after all. I also click before she hits the end of the leash at one point, probably looks like I'm reinforcing lunging but again, you have to take a snapshot of exactly what happens at the time of the click. It can be hard to get your head around at first, and I'm sure even a lot of clicker trainers will have questions about what I'm doing here (and maybe they would be right, too, I think I make a lot of mistakes in the clip). In the last few classes we had Tess was walking on a loose leash past other (reactive) dogs in close proximity. We also had her heeling around other dogs dragging a long line. It's probably not something you are used to seeing and early on, as you point out, it looks like she is all over the place. How long is the clip, a couple of minutes? And does she look the same at the end as in the beginning? If you'd seen her on a check chain you'd think even the first 10 seconds was an improvement Admittedly, despite all the classes, her owner was not competent with a check chain but the point is she became competent with my method very quickly and her dog started to enjoy being good.

I'd have to see it to make an informed comment. If it weren't for the urination (and a couple of other things) I would guess that they were playing, my GSD and one of my cats do exactly this - but it's important to note that in their relationship there is no biting or forceful pinning (or urination, but interestingly they do seem to have 'bases' where they will always leave each other alone). As others have mentioned the swishing tail can be a sign of anxiety or anger. I do not believe that cats and dogs 'fight' for dominance and there is no real evidence to support what you have been told. They might try to establish a territory, or one might see the other as prey, or they might dislike each other but I doubt you will ever get a stable or contextual hierarchy and it's not particularly useful to use a dominance model to diagnose or treat this sort of problem and can do more harm than good, I'll be blunt about that. Saxonpup has asked about something I have mentioned in another thread which might be a little too complicated to explain here but I'll try. You can positively reinforce calm responses and teach each animal that the presence of the other is a 'cue' for these calm responses. So basically, you teach them some calm behaviours, then reinforce those behaviours while they are together. Even better if you can have them relax together in close proximity but I think actually achieving that can be difficult without someone who is confident that they can do it without creating stress or avoidance. By getting them to relax around each other they soon start to accept each other as part of the family. It can also help to teach the cat to escape somewhere high around the dog, so the dog isn't learning to chase. Give the cat plenty of escape options, even if it means setting up some props (a step ladder, for e.g) If you are familiar with clicker training you can attempt to 'capture 'these calm behaviours as they occur. They don't need to be on cue. If you see something you like, mark it with the clicker, then reward it with food. You can do this for the dog and the cat also, although it's not as easy with cats because you normally need to control their feeding to accommodate training (but worth a try, some cats are big gutses). Some things you might click for: - not chasing - sitting - laying quietly - not swishing the tail (cat) - pausing or being still around each other (cat or dog) It sounds like the dog is playing (although I can't tell for sure without seeing it), but the cat is not seeing it the same way. I would definitely step in and remove pup. It might be worth keeping pup on a leash around the cat for a while, just until the cat settles down and until you can reinforce some nice, calm behaviours in the dog around the cat.

Sorry if you think your posts are being twisted and turned, I don't believe that is the case (bar the "all dogs you have ever owned" comment). When you referenced Miss RB you said that her dog had "no recall", those are your words, and clearly her dog has "some recall". I agree that a dog who fails a recall hasn't been conditioned in that situation (i.e "doesn't understand") but at what point can you say they have "no recall"? And how else are we supposed to interpret "no recall"? It's a very black or white comment, and countenanced with some of your other comments it leaves some questions - how are we supposed to interpret your reported "100% recall" if "no recall" actually means "some recall"? Are you being objective or subjective? I don't want to make a big deal out of it, I'm just being honest with you. It could account for some of the responses you have received. I havent seen a reliable positively trained recall though??? I am not being cocky. I don't think you're being cocky, I'm just uncertain about how objective you are. And please don't take this the wrong way, I'm not saying you're not objective, but the evidence does not make it clear that you are. If I had no doubt at all I wouldn't bother to bring it up.

I didnt say that. I said this. I am sensing from your embellishment of exaggerration In fairness to Miss RB, you did say "no recall" which is an exaggeration. You also said you've never seen a reliable positively trained recall and that all your dogs have had 100% recall, regardless of breed or background. It all sounds a bit too "black or white", which could explain some of the responses you have seen.

Haha, that's getting very specific! Is anyone familiar with "matching law"? Basically, the bigger the reward for an unwanted behaviour, the more frequently you will need to reward the behaviour you want instead (or the better the rewards you will need to use). So yup, fairly appropriate for nearly any sort of hound and a lot of terriers, indigenous breeds, or lines selected for very high drive. If they've been able to chase the bunny a lot of times, or got a long way into the prey sequence, it's going to be harder than if it's something that's only happened once, or you've always got them back before they've got too far into the chase (a couple of examples in this thread of people who appear to do this and claim very strong recalls). Lots of reinforced repetition builds stronger responses (note that I have not made a distinction between + or - reinforcement here). For practical purposes, reinforcer frequency can trump reinforcer magnitude. The sort of experiments they run in labs include things like 8 hour training sessions with thousands of repetitions, and there is no way (IMHO) a pigeon is "counting" how many pellets they got for pecking one key vs another but matching law seems to hold true whether it's just a few trials or hundreds of trials. So (until we can read pigeon minds, and we're working on it) we can probably assume they aren't putting a whole lot of cognition into deciding which key to peck.

Hehe, I was once effusively praised for my "horse whispering skills" - I can't even ride! I just understand reinforcement really well, the observers did not.

Well said. 100% of what? If your dog never encounters anything that would make a recall a real challenge, and this might be unique for this dog, then you will never know. If we actually want to collect data we have to impose limits. I'm more interested in what a dog does in a particular situation as a measure of results; e.g dogs in my reactive and aggressive dogs classes are measured for recalls away from other dogs, and then I only know the probability of what they will do in similar situations. Anything could happen tomorrow.

Broadly speaking there are two common responses to stress, flight (hide away) or fight (stand up to it). Either are an attempt to reduce stress, what you have described would be called a "distance increasing behaviour", dog barks or charges and other dog backs off (hopefully).

This method for using food is also very effective, probably a bit simpler with clearer instructions that what I have offered above: http://www.shirleychong.com/keepers/Lesson6.html Just remember, just because your dog ignores food while she is running off to play with other dogs and kids doesn't mean that food isn't a reinforcer (it's just less exciting than other dogs and kids, this does not mean we can't build a reliable recall away from those things with food) so you can include food in Step 2 of Recall Redux.

Can you describe how you tried to use treats? When she runs off, I call her name and pull out a treat (making a rustling noise) but she just keeps on running. She's not food orientated (my Lab is!), which is unusual for Goldens OK, I think a lot of people might be making this mistake. That's what we might call a "bribe", or if we're being really kind a "lure" The idea is not to try to entice the dog back with a food treat, but to condition a behavioural response using positive reinforcement. So we start off in a low-distraction environment, maybe your hallway or living room, and we just call the dog back and forth between two or more people, giving a treat each time to reinforce this fairly predictable response. If no helper is available, call your dog then toss the treat to get him away from you so you can call him again. Then we take it outside, just in the back yard. No other dogs around. Build it up nice and strong in these low-distraction environments. Don't reinforce the slowest responses, the ones where he takes a while to respond or dawdles. We want fast responses, and fast returns. We might even take a couple of steps back when he's close to encourage this, or toss the food treat behind us. When we're consistently getting fast responses we might start to artificially add distractions in controlled set-ups. So we might have someone hold some food in their hand, not letting the dog get it, while we recall them past the helper. Or have someone holding a ball. The idea is that we set the dog up for success, if he doesn't succeed we have nothing to reinforce and reinforcement is what builds the behaviour, it is vital to conditioning. When he is really good at recalling past set-up, controlled distractions where we can ensure that he is responding as we want, then we can introduce a long-line (tracking leash) and take him places where we can work on more difficult recalls. Again, keep setting him up for success. If you aren't at least 80% sure that he will come back, don't call him. That's what the long-line is for. If you take him to the dog beach or park, don't have him on the long-leash, and don't expect to call him either, or at least not while he is distracted. Wait until he is calm or looking for you and your probability of success is above 80%. I hope that helps. Hopefully you can see there is an important distinction between bribing and reinforcing a dog with food. The aim is always to condition a response and it doesn't happen overnight. All [healthy] dogs are food motivated, or they die. It's not quite that black and white, but if you really have any food motivation problems, write in with details.

Can you describe how you tried to use treats?

Good onya, Alfie02, that's a really nice thing to do and it will make a big difference for this dog.

I've seen a lot of people learn how to clicker train, in person and on-line. The sample size that informs my opinions is in the tens of thousands. If the OP is a member of this forum they have a more than adequate support base. That is my opinion, if you have a different opinion then I respect your right to it and will not invalidate it. You attempted to invalidate Corvus' opinion because you believed it to be harmful, but you did not provide anything other than an opinion to support that claim.