Saturday, May 30, 2009

Productive Training Session

I laid two tracks at roughly 7:45 pm, one for Dustin and one for a puppy. I didn’t decide which puppy until later – the one I could get to lie down on hand signal/command with less fuss, which was Bruce, pictured.

At 8:30 (the earliest I could expect to have track scent stronger than trail scent) I took Dustin out to work, and he happily found the starting article by the flag to lie down on it. Though he seemed to need to think about his reward a little (cheddar cheese) and decide whether he liked it, he went on willingly from there. The turn I’d made went straight into the wind by the time we ran it, and Dustin indicted with a high head that he could find all the articles easily enough from there. I reminded him that “Track” meant “Work the line on the ground” anyway, and he obliged. Released from work at the end, he danced off, gave himself a play reward, and discovered the end of the puppy track. He scooped up the cloth glove, showed it to me just to prove he’d found it, danced in a crazy circle, and dropped it on the track again.

I love that he can tell when we’re not working anymore, and that he has such a tidy mind.

I took Bruce to his track next. This was his second. The first had been a mess; he forgot his down, panicked because he didn’t know what I wanted, and his sister ran over to take the rest of the game away from him. The object of the game at present is only to get him to equate discovering an article and lying down on it with reward. Last night, he discovered the first article and puzzled over it, then dropped readily when I asked him to. He sniffed out the second article on his own, natural line-follower that he is, and flopped. I was impressed.

We did both miss the fourth article of the five. He wasn’t sure what he was looking for, but could tell the scent was concentrated in the right sort of way, and fell all over the place, down here and there and everywhere, and I had a lesson in not using brown leather articles at twilight near brown leaves of similar size while working an inexpert tracker. Still, I expect the next time will be at least as good and possibly better.

Next time I’ll put another out for Wanda, just to see what she makes of it. She had an obedience lesson for steak scraps in place of a track this time. She’s become less grabby at long last, and so I can now get the basic positions (stand, sit, and down) and a passable heel, front, and finish. Of course, getting those without a handful of scraps would be something else again, but the patterns can be obtained and labeled. She’s clever enough.

Both of them are nimble for their age, possibly an effect of their wild first eight weeks. I’d love to start them on the safer bits of agility, ramps and step-over jumps, and they’d enjoy it. First, of course, I have to build the ramps. At present, they make up their own courses around the yard and over sleeping adult dogs.

Friday, May 29, 2009

Organizing Puppy Training

As I may have mentioned, a couple of months ago my spouse and I pulled a couple of half-Golden (or, since we know almost nothing of their mother, perhaps I should say Goldish rather than Golden) puppies out from under a pile of old fence at our new house. The world has not exactly been beating down our door to adopt them, though there have been a couple of nibbles. There are a lot of homeless dogs out there right now, not to mention homeless people.

Now, of course, the pups are almost four months old. Gone are the days of the growling little monster and the catatonically terrified heap of fuzz. That, in fact, lasted about as long as it took me to dish up a little yogurt. The little tykes are now quite sweet. They are crate trained. They have at least a vague grasp of useful words like Sit, Down, and Let's Go. They've proven themselves to have some talents worth developing. Bruce is a natural for putting his nose down and following a trail. His sister is born to do tricks.

Aside from a few cautious ventures onto well-scrubbed floors and around our biggish yard, though, they haven't been out much. I wanted to keep them close to home until I was fairly sure their vaccines had taken. However, 2-4 months is a lovely time for them to learn all those basic useful words mentioned. Lots of praise, lots of cheese, and lots of chicken have given these little tykes the idea that words can relate to actions. It's a good concept and far easier to build in a puppy than in an adult dog. Wanda is learning "Bump!" too, as she is small enough that she'll never hurt anyone by jumping up on purpose, and having that in her repertoire leaves open the possibility of Hearing Ear or other assistance work. She's a nice size for it and exceedingly smart.

However, the vaccines should be in there by now, doing the job, and it's time to start some real work for them. We'll be going more places soon, learning that there are a lot of people in the world and not all of them smell or look the same. We'll be learning that there are dogs other than the big German Shepherds these two are now used to, and that there's more than one cat in the world. And hopefully along with this we're going to learn that "Sit" still means the same thing even in a crowd. Dogs can be very context-dependant. It's why I don't care for collar-yank training, actually -- if the dog learns that "Sit" means "I'm going to yank up on your collar and shove down on your rear" then he doesn't learn what to do off-leash and out of reach, and his handler doesn't learn how to correct the error in translation, either. Too, a physical correction if the dog has genuinely forgotten (and quick, how well do you remember what you learned in third-grade social studies?) doesn't improve memory. It just makes the handler look unreasonable.

If everything settles in as it should, then when we reach that teenagerish stage when the puppies start wondering "What happens if I don't?" we'll still have communication. That's a big conceptual leap, the idea of disobedience, and in some fields (scent work, for instance) a concept the dog really needs to have. Humans can be really, really wrong about what the nose can and cannot determine. A dog who's sure the world comes to an end if he disobeys can fail to communicate all sorts of things, such as that the drugs really aren't in that locker or that the person really is up that tree, just because the handler is so certain of being right. The teenagerish age is when my dogs learn a whole new set of useful words: Today, All the Way, and What Was That. In order, that means to do it faster, completely, and less creatively.

Did you notice an option not yet mentioned? I do use "No." It's for chewing electrical cables. It's for being about to knock down something very heavy or otherwise dangerous. It's for actually doing something that could be lethal to the dog. If I'm not chanting it constantly, they know this is the Never Ever word. If I say it for getting up from a sit, either they discount the word or I've created unnecessary anxiety. "Phooey" works fine for small mistakes, and who among us is perfect?

Wednesday, May 27, 2009

Research on Scent Dogs

You may want to research a little of the science on canine perception if you like to know how things work before you start trying to do them. How well should you know the nose?

There are several ordinary-language works on how your dog perceives the world. One of the better ones is Chapter 5 of Stephen Budiansky’s The Truth About Dogs. While other sections of the book, such as the pro-alpha stance and the aggression discussion, may be debatable, he’s done a very nice job of summarizing the science on canine perception, particularly on sight and scent. Too, the illustrations of color perception are worth a thousand words. You will see very soon why your silly dog can fetch his blue toy out of the grass so much faster than his red one! Knowing this, you can make your scent articles easy to see for you and invisible to him or plant your decoy in the woods wearing bright red in summer so you can watch your dog's alert as you get close.

For scent work, Pearsall and Verbruggen have an excellent first four chapters in Scent. Their discussion of skin rafts, the microscopic few-cell bits of our surface which we shed constantly, is coherent and understandable, though it may give clean freaks the heebie-jeebies for a little while. Likewise, the length of time a fingerprint can persist is enlightening when your dog takes off on a trail you were pretty sure couldn’t possibly be there. The training portion contradicts some of the science end of the book. I bought the book used, refer often to the first parts for understanding how human scent is deposited and perceived, and largely ignore the rest.

If you want to get technical, try the “Sources” section of the Budiansky and start Googling. There are some excellent articles out there, though many of them are hidden behind the academic-access filter. Another resource is to accept common-source articles. For instance, through searching one academic article, I found accessible ones which cited it, though I could only reach the abstract of the original. “The Use of Scent Detection Dogs” in the Irish Veterinary Journal is beautifully thorough and relatively recent, published in 2006. I have to agree with the final sentence: the limit on how we can utilize a trained dog’s nose is primarily our own imaginations. Find fire ants? Exterminated mice for safe removal before decomposition? The use of human growth hormone by athletes? First, someone has to think to try it.

Even if you never train your dog to search for a specific thing, knowing how sensitive his nose is may make you more likely to listen when he wants you to know you smell different to him. Untrained dogs can still alert their owners to everything from cancer to pregnancy, whether the children are smoking something they shouldn’t, and household risks. However, the humans of the household still have to be willing to take the “I smell something different” body language seriously.

Working The Young Dog

Dustin (above) has now been with me for about a year and a half. I don't work him nearly as hard as I worked Sunny; he doesn't demand it of me, for one thing. At this age, she was deploying on searches. He is still working on formal tracking at a pretty low level and air-scents more or less as he sees fit.

One reason for this aside from his drive (and mine) being lower is that I have no current training partner. For air scent, there's only so far you can go by thanking the dog for pointing out your neighbors -- though it's a great start! Tracking, at least, I can do by myself for quite some time yet.

The method I picked up from Mary Adelman (yes, her again!) with asides from the book by Lue Button is to put out straight lines with articles dropped on them and to teach the dog to find the articles by finding the line. At the articles, the dog should down and take his reward. The chief difference between Adelman and Button is the age of the track suggested. The first recommends working in the one-to-four hour window of track scent, which is primarily crushed vegetation with only traces of human scent. The second suggests working at a full day's age. At this point the crushed-vegetation scent is gone, but a good bit of the human scent is also. Working back from this point apparently worked well for her. With Sunny I did the hour-old tracks first, with very short stretches between articles, and she jumped to day-old tracks easily. However, since human "trail" scent is very strong in the first 45 minutes or so, and in the fourth to tenth hour of the trail or so, when we started running trails with the SAR team which liked to work quickly, she became a little drunk on the scent and followed air currents too easily. She worked well in the tracking window, and once she'd worked a while on the hotter trails, she figured out that if the scent appeared to go one way on the breeze and another on the ground, she should look to the grass.

For now, Dustin is getting articles around every twenty paces -- not too exactly, as while a dog cannot count to twenty as far as I know, they do get a feel for "There ought to be an article around here somewhere." He has to work out turns and some changes of vegetation. I've learned that he doesn't care to work across pine needles and greatly prefers tall grass to short. He's learned that if he works at it he finds the articles and gets cheese. We're getting somewhere.

What's an article for us? I don't lock him in on leather for schutzhund or on gloves and wallets for AKC tracking; someday we may need to find an actual person or run the VST. Real people and VST tracklayers are prone to dropping things like water bottles, soda cans, business cards, and other oddities. He does get some gloves and leather bits. He also has to identify voided credit cards, metal washers, keyrings, and the like. To my surprise, just as Sunny did, he likes metal articles. I like the cloth ones I made from a pair of hot pink sweatpants, as the color is highly visible to me and utterly lost in the grass for the dog.

Friday, May 15, 2009

Commands and Communication

"Training your dog's nose" may be an inaccurate title. The average dog knows how to use his nose just fine, just as readily and easily as you use your eyes. The trick is not in teaching his nose; it's teaching him a vocabulary.

My last scenting dog, Sunny, started off with "Search" to find airborne scent and "Find another one." The latter was useful for tracking: find another thing with the tracklayer's scent on it, and another, and another -- surprise! Those footprints in between link them up. From there we developed "Track."

What I didn't specifically train for some time was article search without a connecting track. However, one fine day, I needed to have a key in my car to run the air conditioner and another to lock it while I was not in it. In finally shutting everything down and leaving the car, I lost one of the two keys. Several hours later, returning to the car, I discovered this. Note: gray plastic-and-steel key, gray parking lot, nighttime with depth-perception-destroying sodium lights. I went home and collected Sunny, then went back.

I showed her my keyring and told her "Take scent," a command she already knew. Then I pointed to the ground as though we were tracking and said "Search" as though we were air-scenting. She looked puzzled for a moment, thoughtful for a second moment, then put her nose down and began casting around.

A few sweeps of the nose later, she took a few steps straight ahead, looked back at me to say quite plainly, "Is this what you wanted? You need new glasses," and lay down on my key. Twelve feet from where we'd started, it had been completely invisible to me. To her, it stood out just fine. Since we could communicate pretty well, she had figured out the key was probably what I'd had in mind, and since I use positive training, she was willing to take a chance on guessing.

When evaluating a training system, whether for scentwork or otherwise, it's worth asking, "What else can I get from this?" In this case, by working from a "Find another one" tracking system from Mary Adelman's method and also making sure my dog knew air scent was a useable and viable option, I "bought two, got article search free." Later, when we took up trailing, which relies on air scent deposited along any surface to which skin rafts can stick, it was far less work than it would have been starting from scratch because my dog knew about scent flows and footprints. Extending the metaphor, we took trailing at a discount.

Wednesday, May 6, 2009

Track, Trail or Airscent on in!

Welcome to the new home of Sniffydog! I am moving away from for a variety of reasons.

At present, I am working a young German Shepherd Dog, Dustin, in AKC tracking. He is working himself in air scent while I research the military's Silent Scout training -- a post later on this subject, or possibly several. My past experience includes training another GSD in tracking, trailing, air scent, and cadaver while helping friends train their German Shorthair Pointers, Bloodhounds, Labrador Retrievers, Belgian Malinois, and mixes thereof for one or another thing off that list. I am the sort of person to capitalize proper breed names most of the time.

For practice, I also snatch up the nearest rescue to test a new training method on. If it'll work on the baby chow mix or the elderly and traumatized GSD, it will probably also work for your dog, and possibly even on your cat, rat or horse. These methods usually involve setting up a situation to get the desired response, labeling the response, and rewarding with food, praise, petting, and more food.

My main research fallbacks include (but this is the short list) Mary Adelman, Lou Button, Susan Bulanda, and Stephen Budiansky. Further "highly recommended readings" will follow. So will the occasional wacky post on canine misadventures around the house.