![]() |
|||||
STAMP!
AATM
|
|||||
|
Saying you're an animal trainer is a little like saying you're a doctor. There
are countless specialties, subdivisions, techniques, etc. yet all animal trainers have some common ground. Likewise, while there are vast differences in different species and how they take to being trained, there
is definitely a common ground in the training of almost any species of animal. I
am neither a specialist nor an all-arounder. I've spent most of my life working
with dogs, cats, and birds but have also worked with a variety of barnyard beasts, the occasional turtle, and a handful of
rodents. After years of pondering the differences and similarities I decided
to experiment teaching the same behavior to a dog, a cat, a camel and a parrot. The
choice of behavior was simple, to go to a target. The follow up was to use that
simple behavior to create a more complex behavior. I hope that following my adventures
will shed some light on how to train different species using similar techniques while respecting their species-specific behaviors. Teaching an animal to target in the first key to opening up a channel of communication
that allows one to 'explain' all sorts of behaviors to an animal quickly and with relatively little stress. It is not unreasonable to think of target training as the most important step in training any animal. An animal who will go to and remain at a target can be shown how to go into a cage,
walk on a lead, come to its handler, remain still for husbandry procedures, almost any basic behavior can start with simple
targeting. Training the camel, my task was to use targeting to get the animal
to willingly enter and walk through the barn. She had developed a fear of going
into the barn and her owners could not convince her to enter it. I chose to train
the cat to go into a cage, something many of my clients have trouble getting their cats to do.
The dog was to learn to bring a message to someone and jump on them until he was noticed. The parrot simply had to fly to my hand or to his cage on cue, he's my spoiled baby so he got off easy. Those who think that animal training is a long process that requires large amounts
of patience will be interested to know that the time spent training each animal for this experiment varied from ½ an hour
to 1 hour maximum. Granted, some of the animals already had training experience,
but the fact remains that training was not a long drawn out procedure. PART 1 TWIGGY THE CAMEL
Twiggy is an almost two-year-old female bottle raised camel who is the first camel her owners have ever had. They are experienced animal people, and did a wonderful job raising and socializing her. She is friendly, easy to halter, and easy to lead, but has otherwise had no training. My previous experience with camels was limited to handling camels trained by other trainers, and even that
experience was fairly limited. Because of my inexperience with camels and Twiggy's
inexperience with training, I thought she'd be perfect for the experiment. I arrived at the farm in the late morning. I
wasn't exactly sure what I would work on with Twiggy other than teaching her to touch the target, but I figured I'd think
of something when I got to know her. Before heading out to the barn I chatted
with her owners and discovered that she would not walk through the barn, a rather typical horse barn with stalls, feed, storage
and tack rooms, and wide corridor down the middle. She had not had any bad experiences
in the barn, but no one had ever been able to make her go through it. My first
thought was, 'Great! I'll teach
her to walk through the barn!" My second thought was what had I gotten into now
as I pictured myself struggling with this very large youngster inside the dark recesses of the barn. No matter, I had already explained to everyone that I had never trained a camel before and didn't really
know what I was doing. Her owners had been instructed to remove her favorite
food treats from her menu for the day so that I would be able to use them as reinforcers.
She is partial to carrots so I neatly sliced carrots into 75 bite sized morsels and headed out followed by an audience
of four family members and friends as well as a photographer.
My training plan was to start by teaching Twiggy to tough the target with her muzzle while she was on one side of the
fence and I on the other. If that went well I would go ahead into the field with
her and work on target touches in different directions (up, down, left, right) as well as following the target. My next step would be to have her follow the target into the barn a few steps. I figured that I could accomplish that much in one session if she was compliant, and was ready to back
off my goals if she was more difficult. Twiggy headed toward us as soon as she
saw us walking down the path, and was standing with her head stretched over the fence by the time we reached her. The distraction of so many people all milling around talking was a bit much for both of us so I settled
everyone a short distance away where the would see us but could not be directly distracting.
I had decided to use Kayce Cover's favorite conditioned reinforcer, the letter 'X', along with the intermediate reinforcer
'xxxx' to train Twiggy. I stuck
te farget in front of Twiggy's muzzle, she lipped it curiously, "X!" I handed
her a piece of carrot. She and I repeated that scenario about six or seven times. I could see the light coming on in her eyes.
I could also see that it was taking too long for her to pick the carrot piece out of my hand and that distracted both
of us from the training. I switched to slipping the carrot into the side of her
mouth, which worked much better. I moved the target about a foot away from her
muzzle, she started to reach toward it, and 'xxxxxxxx' she touched it, 'X' and another piece of carrot. Again I moved the target about a foot away from her muzzle, again she reached toward it as I kept up a
soft stream of x'z and rewarded her with a crisp 'X!' and a piece of carrot for success.
The third time I did this she started moving toward the target, 'xxxxx' then stopped.
I stopped saying x. She looked at me for a long moment pondering the situation. She reached forward again. I started
x-ing again. She stopped. I stopped. She started. I started. She touched the target, "X!". She did the same thing one more
time, experimenting with getting me to start and stop x-ing. I knew she understood
the concept when on the next try she went purposefully toward the target with one ear tilted toward me to make sure I was
continuing with the x's. We were ready for the next step.
Once I was inside the field with Twiggy she had to spend a few seconds exploring me from head to toe before getting
her mind back to work. Rather than let her learn to ignore a target when she
wasn't in the mood to work, I put the target away until she'd gotten her curiosity under control. She made the transition to targeting with me in the field smoothly and effortlessly. I started to up the ante, asking for a step forward rather than just a each. She progressed rapidly, almost always moving to the target immediately and confidently as soon as I had
positioned it. The photographer could not see us clearly and needed to get into
the enclosure with us. I put the target away again for a moment while she explored
this new person and brought it out once she'd had her fill and was again turned toward me.
She headed for the target, 'xxxxxxxxx' she turned back toward the photographer, I stopped x-ing. She cocked her head toward me questioningly. I started x-ing
again. She took another step toward me, 'xxxxx' she paused, silence. Another step, a pause, and finally the last step to the target, 'X!'.
I slipped several carrot pieces into her mouth, praised her, and gave her a good scratch on the neck. I wasn't sure she'd like these things, but it seemed like she would, and she did.
The next step was to teach her to follow me at a polite distance using the target as an indicator of where her head
should be positioned. I had plenty of experience with horses that crowded and
mugged while they were being led and did not want her to get into that kind of habit.
She tried to walk with her chin on my shoulder a couple of times, but moved herself back to the target when she realized
the intermediate bridging (xxxxx) had stopped. I x-ed continuously when she was
right, and reinforced with a big 'X!' and a carrot piece at variable intervals. All
was progressing smoothly and quickly. She was much easier to train than I had
anticipated. I started asking for longer stretches of attention, for turns and
circles and the like. Suddenly Twiggy stopped working and wheeled around away
from me. One of the spectators had decided to come around the side of the barn
for a better view. Behaving like and being with the tremendous range of peripheral
vision of a typical prey animal, Twiggy was immediately aware of this change in her environment. Being bottle raised and imprinted on humans, she was more interested in exploring this new appearance than
in bolting away from it, but either way, her concentration on training had been broken.
We had been working steadily for five minutes, so I decided to give her a break for a few minutes to explore the visitor
and reflect on her experiences with the target. After a brief break Twiggy was clearly excited to get back to work. As soon as I re-appeared with the target she walked over and touched it.
This time I did not 'X!' right away, but continued the soft stream of 'xxxxx' for a brief moment then 'X-ed' and reinforced. I repeated the target touch several times, sometimes reinforcing immediately, sometimes
asking for more duration before reinforcing. Only occasionally did Twiggy take
her muzzle off the target before I released her. She was well on her way to understanding
the nuances of our communication. The soft 'xxxxxxx' of the intermediate bridge
was reassuring her that she was on the right track, just as silence indicated to her that she had slipped off the path to
the right behavior. She performed with energy and confidence. We were ready to take our first step inside the barn. Twiggy
followed me (and the target) up to the barn entrance. We stopped with me just
inside the barn and Twiggy just outside of it. I reinforced her with several
pieces of carrot and put the target away for a moment while Twiggy stood looking into the barn. I brought the target out and placed it about six inches from her muzzle.
She reached forward without moving her feet and touched the target. I
reinforced and moved the target another six inches forward. She again stretched
forward to touch the target without moving her feet. I continued with her this
way, edging the target a few inches forward at a time, 'xxxx-ing continuously as she reached cautiously toward it, reinforcing
her with 'X!' and a piece of carrot for each touch until her neck was stretched horizontally.
I put the target away for a moment and let her relax. When I brought the
target out again I held my ground but did not go further into the barn. Twiggy
had remained in the same place and again stretched her neck out fully to touch the target without moving her feet. I reinforced her and immediately moved the target another two inched into the barn. I started to 'xxxxxxx' as she tried to extend her neck further. She
realized that her neck had reached its limit and very gingerly moved one front foot forward into the barn. 'X!'
As she munched on her slice of carrot I turned to her owners and asked exactly how far into the barn they were able
to get her. "About as far as she is now" was the answer. This was to be our moment of truth. Twiggy was completely
at liberty to bolt any time the g to be too much, she understood the concept of moving to and touching the target, and she
knew that I'd let her know whether or not she was on the right track with constant feedback.
I moved the target another six inches into the barn. Twiggy contemplated
for a moment then took a slow step into the barn, moving all four feet. We moved
slowly into the barn, Twiggy shuffling her feet forward as little as possible each time I moved the target. When we were about one third of the way through the barn she started to exhibit what I interpret as signs
of stress, ignoring the target and nuzzling bits of straw and fallen feed on the barn floor.
I let her take another break. Interestingly, she did not leave the barn,
even though she was not being reinforced for staying in it. She kept her head
down for a few more seconds then raised it as if to say, "OK, let's get back to work."
I was almost out of carrot so I switched to plain feed. At this point
I felt that the actual reinforcer was not nearly as important to her as the ability to control her environment and 'make'
me explain things to her and give her things. I also had a banana in my pocket
as back up. Apparently bananas are among her favorite foods, but they aren't
the easiest reinforcers to dispense. We made it to the halfway point of the barn
uneventfully, moving a few inches at a time.
Two of the resident horses had become interested in our trip through the barn and hung their heads over stall doors
to watch. Twiggy has grown up with these horses and was not perturbed. We kept on moving. Suddenly one of the horses goosed Twiggy
in the rear. Nothing hard or malicious, but enough to startle her into hightailing
it out of the barn. Damn! There
went all my hard work. Without thinking I waved the target at Twiggy's departing
behind and said 'touch'. Touch is the cue I use for targeting, and even though
I had not yet taught Twiggy the cue I had certainly mumbled the word as I was praising her for her work. As I mentioned before, her peripheral vision is great, and I knew she'd seen the target. I started to mumble 'xxxxx' almost under my breath. She paused. I "x-ed" a tiny bit louder. She turned
toward me. I headed toward her and met her half way. This occasion definitely merited a piece of banana. Despite being goosed by the horse, Twiggy moved through the barn much more quickly
the second time. I never pushed for great distances, but increased the reach
until she was moving ten to twenty inches at a time. The last ten feet of the
barn were particularly scary to her because of a dreaded black matting that she seemed sure would swallow her whole. We went back to inching our way forward. She
repeatedly progressed to having three feet on the mat, but it took several attempts before she got up the courage to put all
four feet on the mat at once. Beyond the mat was the far end of the barn where
all of our audience was waiting to pet and praise her as she gobbled up the rest of the banana. Twiggy had made it across the barn for the first time in her life.
After this first walk through the barn at liberty, I put Twiggy's halter and lead rope on and repeated the trip. As one might have predicted, it was somewhat harder for her to get through the barn
with halter and lead rope as she did not have the same feeling that she could bolt if she needed to. While I was initially careful not to apply any pressure to the lead rope, after we had made it well into
the barn I began to apply a little pressure to the lead rope at times when she started to backtrack. I tried to be careful not to pull to the extent that she became frightened, but at the same time wanted
her to feel that pressure on the halter did not necessarily and immediately mean she was going to be dragged or forced into
anything. Any time I applied pressure I xxxxx-ed so that she knew she was doing
the right thing by accepting this handling. On the one occasion when she looked
a little panicked I loosened the pressure and draped my arm around her neck, rubbing and petting her while talking to her
calmly. She re-gained her composure quickly.
At the end of our training session Twiggy had been able to walk through the barn willingly both at liberty and with
a halter and lead rope. She had clearly enjoyed the experience. Although I knew she must be somewhat tired from the effort(I knew I was), she nuzzled me hopefully as I
packed up to go, obviously hoping I'd take my target out again for another round.
Twiggy will need to repeat all these steps a number of times before she's comfortable and confident enough to walk
through the barn without a target to guide her, but she now had the skills she needs to work though her fears without a great
deal of stress. I lost track of the exact amount of time I spent with her, but
I know that the entire training session, including breaks, was less than an hour long and we used the equivalent of three
carrots, one banana, and a couple of handfuls of feed.
PART 2
I started by tacking the target to the wall at cat height and called The next session took up where we had left off.
Our third and final session entailed affixing the target to the inside of a cage.
I used one of the larger dog crates to make photographing the sequence easier.
Once the target was inside the crate I called
PART 3 BARTOK THE AFRICAN GREY
PARROT
Bartok is a fully flighted 3 and a half-year-old male African grey parrot. He
is my personal companion and house pet. He has been trained to perform a few
tricks, trained with Kayce's Bridge and Target techniques, and also worked with in the more traditional 'pet owner playing
with pet' kind of way. He already had been trained to fly to my hand and to his
cage from my hand when I began the experiment, but I had not worked with him for a while.
In addition, Bartok was going through a moody phase and did not want to be handled.
This seems to be a normal part of Bartok's behavior, every year when he's molting he shuns any physical attention from
me. As he is fully flighted and one room in our apartment has double height ceilings,
he can be quite difficult to manage if he's in one of his 'moods'. I looked forward
to seeing how working on targeting would affect my ability to manage him. In
the past he's always become very compliant when I keep working with him, but he is reaching adulthood and I thought that might
affect his willingness to work.
It is worth noting that both Edison and Stamp were easily able to work with a great deal of ambient distractions, other
animals coming through the room, toys on the floor, people around them, whereas both Twiggy and Bartok were apt to startle
and lose composure at even relatively minor distractions. I think this is a function
of being a prey animal. Predators are also constantly aware of their environment,
but their focus is on catching and consuming prey rather than staying alive, therefore they aren't that bothered by distractions
when they are performing actions that will lead to reinforcement. I don't mean
to imply that predators are not ever bothered by distractions, I'm sure neither Edison nor Stamp would continue working on
my team if a mouse strolled across the room, but the emotional reaction to distraction is not as intense.
I did not use a separate target for Bartok, but instead used my extended hand as a target to land on. I used peanut butter on the end of a chopstick as a reinforcer. I
alternated clicker, verbal bridge (x), and plain old praise as the conditioned reinforcer.
At the start of the training session I placed Bartok on the top of his cage and sat down about five feet away from
him. I extended my hand and cued him to 'fly to me'. He immediately flew to my hand and was reinforced. I then
cued him to 'fly to your cage' which he also did promptly. I reinforced then
sat down again. Again I cued him. He
moved to the very edge of his cage and stretched toward me, but would not take off.
He had noticed the two cats staring at him from the balcony above and refused to move.
He has never had a bad experience with the cats, both of them have received a great deal of training to leave the bird
alone, but he did not feel safe flying below them. I moved closer and closer
until he had only to step from the cage to my hand, which he was willing to do.
Flying back to the cage was less of a problem, probably because it was easy for me to 'launch' him from my hand. I worked my way slowly backward until he was again flying to my hand despite the cats. He then noticed the camera and we had to start from square one. We worked through the camera fear the same way, then called it a day.
This doesn't sound like much progress, but there is a huge difference between Bartok's behavior when he is specifically
asked to target to my hand and his behavior when I just try to manage without cues and bridges. Left to his own devices he is far less compliant and less willing to work through problems than when I
actually follow a carefully structured plan of action to guide him through whatever problems arise. In the five minutes we worked he was able to overcome is fear of flying around the cats and around the
camera, and focus on the behavior I wanted. My biggest shortcoming with Bartok
is that I don't work with him consistently. Every time I find the time to work
with him he progresses by leaps and bounds but I slack off and let him regress when life gets busy. Interestingly, although sometimes there are months between one training session and the next, he always
gets back to speed very quickly and is able to progress from where we had left off.
Clearly the motivation and drive instilled by training has long lasting effects.
PART 4 STAMP THE DOG
Stamp is and eighteen-month-old Norfolk terrier co-owned by my daughter and myself.
He is training for obedience, agility, film work, and earthdog and has a lot of experience with being trained. He understands a wide variety of conditioned reinforcers including several verbal
reinforcers and the clicker. He is also well versed in targeting. He can target with his nose, his paws, his chin, his shoulder and his hip on different cues. He also targets to an indicated target even if it is a novel one.
For example, I will use squares of cardboard, a plastic leaf, a riding crop, a chopstick, basically whatever is on
hand when I am training. For the sake of continuity in the photographs I used
the same target with him that I used with Twiggy and
Session one began with Julian sitting in a chair approximately four feet away from me, holding the target between his
knees so that Stamp could easily see it and reach it without much effort. I sat
Stamp next to me and signaled toward the target with my hand, making sure he had focused on it before sending him with a 'touch'
command. Julian and I both reinforced with a stream of 'xxxxxx's' as he approached
the target, and a terminal bridge, 'X!' when he touched it. Stamp ran back to
me for a treat. We repeated this a few times.
Sometimes I used the cue 'touch', which he already knew; sometimes I used the cue 'find Julian' which he did not know. The first time I changed the cue he stopped half way to Julian and barked at me. I encouraged him with 'touch, find Julian' while indicating in Julian's direction
and he went on to the target. After six or seven repetitions,
Julian moved the target so that it was in his lap, tilted down so that Stamp could see it.
I sent Stamp the same way, and as he approached, in addition to say 'xxxxx', Julian gently waved the target to get
his attention. Stamp jumped onto his lap and we both reinforced him with treats. We repeated this a couple of times then I moved a little further away. Stamp performed well at the longer distance, however, after a couple of repetitions he had gotten too excited
and started barking and jumping about instead of remaining focused on the task. Whenever
Stamp starts his bark fests I take it as an indication that I pushed a lesson too long, so we took a break.
We started session two after a break of a couple minutes. the first send
was again to Julian in the chair with the target in his lap. The second send
was to Julian standing up, holding the target down low in front of his legs. Stamp
was not at all bothered by the change is position. Julian raised the target a
bit with each subsequent send until it was at waist height. When the target had
gotten high enough that Stamp could not reach it without jumping up, he got slightly frustrated and barked at Julian rather
than trying to reach the target. We both immediately stopped 'x-ing' and Julian
gently moved the target about to encourage Stamp to reach for it. After a few
seconds Stamp lunged up and was immediately reinforced. On the next send he jumped
quite purposefully at Julian to reach the target. We then asked for longer duration,
reinforcing only after two or three jumps. Sometimes I pushed for too much and
Stamp reverted to barking and running back and forth between us. We took another
break after he had successfully mastered jumping repeatedly. These two sessions
totaled about ten minutes including the break. We started the third and final
session the next day with a quick review of each step that we had already gone through. Stamp
was very focused on task and performed with very little barking. As soon as we'd
finished the review, Julian held the target so that a tiny bit of it was covered by his shirt.
Stamp went for the target as though nothing had changed. Julian moved
the target further and further up his shirt with each send until it was entirely under his shirt. Stamp continued to run full speed to Julian and jump on him on cue despite the disappearance of the target. At one point in the session someone rang the doorbell
and Stamp, who normally feels it is his duty to charge to the door barking, was clearly torn between the draw of the
door and the draw of the training session. He split the difference and ran to
Julian, jumping on him while barking at the top of his lungs and looking meaningfully at the door.
Once the target was completely out of sight we stopped using it and I started sending Stamp from further and further
away with his message to 'find Julian. At the end of the third session he was
crossing the livingroom and turning the corner to the hallway to jump repeatedly on Julian.
As soon as Julian had touched the message on his collar, instead of ending with a reinforcer, I gave him the command
'here' and he ran to my finger target to be reinforced. To make this task useful
in everyday life it seemed to me that it was important for the dog to understand that the person who sent him was the one
who would ultimately reinforce him, that way the person who had sent the message would always get a confirmation of delivery
by way of the dog returning without the note. CONCLUSIONS
There are a few conclusions I've come to while working on this experiment. The
first is a reaffirmation that target raining any animal is extremely useful for almost any future task you might want the
animal to perform. The target is not only useful to show the animal exactly where
and exactly what you want, it also serves as a focal point which increases drive and concentration regardless of the species. The clearest example of this was the camel.
Twiggy was able to conquer her fears because she had something clear to concentrate on and she knew she was free to bolt if she needed to.
This brings me to the next conclusion, one that I struggle with sometimes but do really believe to be true, and that
is that animals have an easier time learning new tasks if they are at liberty. The
reality of life is that many animals need to learn the skill of working while leashed or haltered, but one should not confuse
the need to leash an animal with the need to use the leash in training. Does
this mean that I believe an animal should never be corrected with a leash? It
depends on the animal. I hade used a leash to control Edison and Bartok when
they were outdoors, but would never dream of giving them any sort of leash correction for disobedience and expecting their
behavior to improve.
These four animals are different in species and in individual character. Their
behavior and response to training reflects these differences. Most important,
however, are the similarities. When clearly guided and reinforced fairly, each
animal was able to learn its assigned task in a very short time. It behooves
us as trainers to question what we're doing rather than what's wrong with the animal when our training progresses slowly and
bumpily. Animals that are untrainable are few and far between, as trainers we
know that but still fall back on wondering why this particular (pick any species) that we are working with is so (fearful,
stubborn, stupid, inattentive ? pick one) when what we really need to focus on is opening the channel of communication. END |
||||