Clicker training, philosophy/methodology question

Description of your first forum.

Clicker training, philosophy/methodology question

Post by Lauren Radne » Tue, 13 Jun 2000 04:00:00



Hi, All,

First of all, I apologize from the bottom of my heart if this topic is a
"groaner" ("oh, NO, I can't believe some stupid newbie is asking THAT
again, HERE we GO with this STOOPID thread ALL OVER AGAIN!!! Sigh...").

That said, here's my question....

I have had four dogs of my own I trained over a span of twenty-plus
years - and other people's dogs, and other species of mine and others,
that I've trained as well -  but let's stick to dogs... Anyway, the dogs
I've trained have, for the most part, been trained with what apparently
is now viewed as "the bad old methods" of using a ***collar. For
example, to teach the dog to heel, you would give the command, and walk
along and, if - after gentler methods were tried, the dog strained ahead
or was inattentive, you would change directions abruptly, firmly
repeating the word "heel", and the dog would get yanked and surprised,
and figure out pretty quickly that he had better pay more attention to
you (you crazy unpredictable human being) than whatever else was so
interesting a minute ago.

At the moment, I have four relatively new dogs at once, of various ages,
and mostly to motivate myself to really get them obedience trained
*now*, and mostly because one of them is already the size of a percheron
horse, and three times as strong, I have signed up for a (very
expensive) six month course of obedience training that (of course,
mostly depending on how much work *I* put into it) is supposed to lead
to a pretty advanced level of trained dog by the end of it all.

This trainer espouses the "clicker" operant conditioning method of
training.

I am trying to be a Good Dog myself, and I am all for training a dog
with positive reinforcement, etc.... I've had a lot of fun watching my
dogs catch on almost immediately when I do things like train them to
heel by praising them *only* when they drift into the position I want
them to walk in, and going silent in mid-syllable if they range out of
that specific position while we are walking. So it's not like I think
you have to club a dog to death if he doesn't obey your command to
"glosnick" the first time, when he obviously doesn't have a clue what
you mean.

However.

I am having some philosophic difficulty with this trainer's methods that
I am obediently trying to use with my dogs....

Before I even met this trainer, I talked with her on the phone, and
expressed my concern that it seemed to me that offering food as a
training reward would result in a dog whose motivation would be "I'll do
it IF you have food," and "I'll do it because YOU can please ME by
giving me food", not "I'll do it because I can please YOU and you're the
pack leader, and I understand the rules, and besides I love to please
you". It also was my observation that food-rewarded dogs tended to obey
more strongly if there was a food reward, and less emphatically - or
perhaps not at all - if there was no food reward.

I was concerned a dog trained this way would, in essence, base his
degree of obeying on whether you had food, whether he was hungry, and
whether the food was desirable enough to compete with whatever the dog
was already interested in at the time.

The trainer said that none of this was an issue, if you were properly
trained in how to "fade" the food rewards from the obedience process. I
was (and am) extremely dubious, but trying to keep an open mind.

This trainer's *own* dog, a three-year old which this trainer has had
since a pup, will not lie down and stay on command, but drifts away
within a few minutes, to be coaxed and cajoled back eventually, when the
trainer notices. Also, this dog jumps on EVERYBODY over and over, which
I associate with a *puppy* behaviour problem, not an *** three year
old dog (belonging to a trainer, no less).

This trainer, on the very first day, had all of us let loose our totally
uncontrolled and untrained puppies and dogs (we were in a very large
room, not outside, thank goodness), and then told us to say the word
"come", once, then scramble all over after our dogs (who were busy
chasing each other with wild delight) trying to shove a food treat under
their nose, and lure them back over to wherever we were when we said
"come"... then click the clicker, and give them the food.

I have a pretty good education in psychology, including operant
conditioning, and more experience in training animals than most people I
meet, and it's been my experience that you never give a command you
can't enforce, or you train the animal it can ignore you. And it's also
been pretty obvious to me that you don't start training an animal in an
area of maximum distraction.

I kept my jaw from dropping at how this trainer handled this first night
of training, but my dog was NOT interested in ANY kind of food treat
that I, or anybody else, could provide, because, of course, NOTHING was
as much fun as chasing some other puppy. For two hours, no less. I
actually lost two pounds when I stood on the scale the next day, so I
guess there was *some* benefit, but I don't think there was any for the
dog - at least, not in the area of obedience training.

Okay, I'll try to stop expressing my extreme dubiousness about the
trainer, and stick to the method...

Here's one of the questions I've had already (it being only the first
week), and I didn't get an answer that really convinced me from the
trainer....

We were supposed to spend the week (and I did) training our dog to,
among other things, "come"... and the method was to say the word once,
then chase the dog down and wave the food under his nose and lure him
back over some distance away, and then click, and give him the food. If
the dog doesn't come, and/or isn't interested in the food, and can't be
lured... then... nothing. You just keep trying to get his attention with
the food he isn't interested in. It's your fault for not having
interesting enough food (or the right pitch of squeaky toy, or appealing
enough belly rubbing techniques, or whatever).

It seems to me that, even by starting in a minimum distraction
environment, with a colossally appealing food treat, and working your
way up through baby-steps of gradually increasing the distraction level,
and gradually decreasing the frequency of the food treat, whenever the
dog hears the word "come", he is still making a decision along the lines
of "am I that hungry? is it that tasty? or would I rather chase this
rabbit? Because if I don't go over there now, it's no big deal... I'll
get that treat later when I feel more like obeying."  I am under the
illusion, or delusion, that the 'bad old methods' result in a thought
process more like, "Yes Ma'am! Right away Ma'am! Even though maybe I'd
rather not, Ma'am!".

I envision a situation where a choke-collar "do it and I praise you
exuberantly, or else I'll enforce it and you'll have to do it anyway"
dog, and a clicker-trained "I really like the food treats, and the
praise, but nothing happens if I don't obey" dog, each see a rabbit and
start to chase it. The rabbit sprints across the highway, and both dogs
follow it. A Mack truck is bearing down on both the dogs. The
choke-collar dog owner says, "Rover, COME!" and Rover says, "Damn! And
that rabbit was so interesting. Okay, here I come!", while the clicker
dog owner says, "Bingo, COME!" and Bingo says "Yeah, maybe, in a minute,
but I am TOTALLY distracted by this WAY cool rabbit and I know for a
fact that there is no food treat on this WORLD that is more interesting
to me right now than chasing this rabbit!"

It seems to me that at least one of the aims of dog training is so that
you can save the life of your dog ("don't eat that <poison>", "come away
from that <rattlesnake> right now", "don't move <further into
traffic>"), and therefore, obedience training should not be a Democracy.
If your dog only sees the rabbit, and you see the rabbit AND the Mack
Truck, and your dog sees pleasant consequences in obeying you, but
nothing compared to the pleasure of chasing the rabbit, and no negative
consequences of *not* obeying you, barring missing a small tidbit of
liver or something... well... it just doesn't seem the best way to
achieve the intended result.

When I asked this question of my trainer during Session Two yesterday,
she said that negatively correcting your dog (using a ***collar for a
correction):
1) caused your dog to fear you
2) caused the dog to label unrelated but coincidental things as "the bad
thing" you were correcting.

I said, politely, that this had not been my experience with my dogs,
that they certainly never feared me, and that I had not seen the
generalization she mentioned.

For those of you still reading this far, I will add that my dogs trained
that way loved me, obeyed enthusiastically, understood the principle
quickly, and it stuck. They were not afraid of me, nor did I seem to
inhibit some wide generalized area of generic behaviours related to
their concepts of walking, or walking on a leash, or exploring the
world, or whatever. That is, if I turned abruptly and said "heel", and
they got surprised and yanked, AND it happened to be a cloudy day, AND
there was a woman with a parasol, AND another dog happened to be
barking, my dogs would walk politely on a leash, but did not seem to
fear or avoid cloudy days, women with parasols, or other barking dogs.
But I digress.

She went on to say over and over that "it's been SCIENTIFICALLY PROVEN"
that this method works.

I am not disputing that. I am questioning the *degree* of
*effectiveness* and the *circumstances* of the effectiveness.

She did say it takes longer (I can see this already).

And she said that "Dogs get into a routine - they are creatures of
routine. If they get into the routine of coming when you call, then they
will do it no matter what."

I did not wish to challenge her further, but I don't know a single one
of us who would follow a "routine" instead of something absolutely
fascinating, tempting, available, and apparently perfectly okay to
do.... "Here Rover, ...

read more »

 
 
 

Clicker training, philosophy/methodology question

Post by DogStar7 » Wed, 14 Jun 2000 04:00:00


Quote:
>What am I missing, PLEASE???

>Sincerely, and somewhat frustrated,
>-Lauren

Hi :)

Read your post and found it very interesting (and funny).

I am of the "old school" of dog training, and  find that it works well for me
and the students that I teach.  I am kind of dubious about this 'positive only'
type of training, but I am sure that there are people it works for.  I know
that in the old school of dog training, there are certain breeds that take
longer to get it than others, I imagine it's the same thing for the clicker
training set.  In my classes, we have about 8 weeks to turn family pets into
well behaved family members, and clicker training would probably take way too
long :)

As for your trainers dog.... my students will see my dog and say " I want my
dog to be like yours"....  and by the end of the 8 weeks, the majority of them
are behaving quite similar to mine.  And none of them are afraid, scared,
cowardly or broken of spirit.

If you are not happy with this class, please do not feel guilty about switching
back to the method you have been successful with in the past.  It is not
abusive; don't let the "postive only" crowd make you feel that way.  You know
your dogs are happy and not afraid, that right there is proof enough:)

I feel that sometimes people try and turn dog training into a complex and mondo
psycological event, when it really doesn't have to be.  JMO :)

Dogstar716
Come see Gunnars Life: http://hometown.aol.com/dogstar716/index.html

"AKC papers do not mean you are getting
a quality dog. They are merely a birth certificate. Even puppy
mill pet shop pups have AKC papers" - Bob Maida

 
 
 

Clicker training, philosophy/methodology question

Post by Mary Heale » Wed, 14 Jun 2000 04:00:00


Quote:

> ...In my classes, we have about 8 weeks to turn family pets into
> well behaved family members, and clicker training would probably take way too
> long :)

Actually, I've found that clicker training goes very quickly, once the
dog and the human have the basics down, than a lot of other methods.

As it's been explained to me (and makes sense, FWIW) is that the
'command' or the label for a behavior is the very last thing to be
applied.  IOW, the trainer's instruction to say "come", then chase the
dog down and lure it, is not appropriate 'clicker-based training'.

Quote:

> .... I've had a lot of fun watching my
> dogs catch on almost immediately when I do things like train them to
> heel by praising them *only* when they drift into the position I want
> them to walk in, and going silent in mid-syllable if they range out of
> that specific position while we are walking.

There's your operant conditioning.

Quote:
> ...it seemed to me that offering food as a
> training reward would result in a dog whose motivation would be "I'll do
> it IF you have food,"

With my dogs, the attitude seems to be "I'll do it because you MAY have
food, and because you're going to make me do it anyway".  Enlightened
self-interest, really.

Quote:
> This trainer, on the very first day, had all of us let loose our totally
> uncontrolled and untrained puppies and dogs (we were in a very large
> room, not outside, thank goodness), and then told us to say the word
> "come", once, then scramble all over after our dogs (who were busy
> chasing each other with wild delight) trying to shove a food treat under
> their nose, and lure them back over to wherever we were when we said
> "come"... then click the clicker, and give them the food.

I.  Don't.  Think.  So.  I think I'd modify a behavior from the "Choose
to Heel" crowd and find something terribly interesting off in a corner
of the room, by myself.  When my dog wandered over to see what was so
fascinating, THEN he'd get a reward.  No command.  Using a command at
this stage is just silly.

Quote:
> I have a pretty good education in psychology, including operant
> conditioning,

Does the activity you describe above seem like any kind of operant
conditioning?  Of the dog, that is - seems like the humans are being
conditioned pretty thoroughly!

Quote:
>...it's been my experience that you never give a command you
> can't enforce,

That's one reason for not issuing commands until the critter is
volunteering the desired behavior on a consistent basis.

Quote:
>... you don't start training an animal in an
> area of maximum distraction.

I haven't had any trouble with this.  I actually had a lot more trouble
with Sam, using more traditional methods, but that's probably because he
and I were both new to everything and, well, I was pretty damned dull.

Quote:
> She did say it takes longer (I can see this already).

Nah.  I saw a team of people take a dog and teach it to alert a sleeping
person to a cry of "Fire!", lead the person to an exit, then lead the
person to an alternate exit.  Took all of an hour to teach the dog this
rather complicated behavior.  (Of course, there was no judgement on the
dog's part - no real fire, the 'sleeping' person was shamming, and the
dog was responding entirely to verbal and positional cues.)

What I would do, if it were me, is teach the dog that 'click' indicates
some kind of reward.  (You didn't mention any such introduction to the
clicker.)  Then, I'd get my clicker and my cookies (or ball, whatever)
and the dog and I'd hang out somewhere that the dog *will* be somewhat
distracted, but safely enclosed.  And I'd wait.  When the dog looked at
me - click! <treat>.  NO COMMAND.  That's how it starts.  Gradually,
increase the requirements for reward - from a look to a step, to two
steps, to wandering past, to coming to you.

An alternate strategy is to use a form of backward chaining.  Train and
reward for a straight sit in front of you.  Then change distances and
angles until the dog can find "front" from anywhere.  Then call for a
'front' (or 'come' - whatever) when the dog is moving.  There's your
recall.

It is at THIS point (when the dog has learned the behavior, and
understands what is required) that you can fairly add compulsion.
Methinks your trainer doesn't truly understand the methods she espouses.

Quote:
>...Dogs are not robots. I
> don't think they are *ruled* by "routine".

You really must meet Noah sometime.  Lovely dog, I adore him, but he IS
ruled by routine.  He must be put abed by 8 pm or he starts fights.  He
must have his morning scritch upon the bed or he sulks.  Knowing "what
happens next" is terribly reassuring for him.  

--
Mary H. & the Ames National Zoo:  Raise a Fund ("Regis");
ANZ Sam-I-Am, CGC; ANZ Noah Doll, CGC; ANZ Babylon Ranger;
felines, finches, fish and Guinea pig

 
 
 

Clicker training, philosophy/methodology question

Post by Elizabeth Nai » Wed, 14 Jun 2000 04:00:00


Quote:

> Before I even met this trainer, I talked with her on the phone, and
> expressed my concern that it seemed to me that offering food as a
> training reward would result in a dog whose motivation would be "I'll do
> it IF you have food," and "I'll do it because YOU can please ME by

no more than a correction trained dog is motivated ONLY to escape
pain/discomfort/irritation/whatever.  Whichever path you take, the
destination is a place where you work together because, gosh, it's just
what the two of you DO.  ;-)

I do wonder a little about the trainer.  What I consider to be a big plus
of clicker training as opposed to food training is that once the click is
conditioned, you need not have food or other reinforcement on your person.
And you can use different things for the "treat" if that's appropriate for
your dog.  So if she is establishing the click early and well, there
should be no need to fiddle with fading the food at all -- food is not
reliably on the trainer's person, very early in the game.  So it ought to
be easier to "fade" than, say, the leash and collar in "traditional"
training!  If she needs to fade food as a separate step, then I would be a
little doubtful that she is doing what I'd consider clicker training.
(Always the disclaimer, IF I've understood correctly what you wrote that
she said.)

Also, in practice, the dog doesn't seem to do a value comparison the way
we think a human would.  But then again, a big part of sucessful clicker
training is thinking ahead and having enough control over the environment
so that the dog doesn't get unplanned reinforcements!

As for the trainer's own dog, this is the same with any method:  you get
the training you want or are willing to settle for.  It is quite possible
to clickertrain a reliable down stay and quite possible to teach a dog not
to jump up without issuing physical corrections.  The reason a
three-year-old dog does or doesn't do these things is because the owner is
willing to put up with that.  (Specifically, I bet she hasn't put any
behaviors on limited hold, and I bet she pets the dog SOMETIMES when it
jumps up or that other people do so and she's never addressed this).
Again, this does not condemn the method, but you might want to consider if
this is the trainer you want to work with.

Quote:
> This trainer, on the very first day, had all of us let loose our totally
> uncontrolled and untrained puppies and dogs (we were in a very large
> room, not outside, thank goodness), and then told us to say the word
> "come", once, then scramble all over after our dogs (who were busy
> chasing each other with wild delight) trying to shove a food treat under
> their nose, and lure them back over to wherever we were when we said
> "come"... then click the clicker, and give them the food.

She didn't have you condition the click first?  Or start with an easier
behavior without distractions, like touching a target or "Doggie Zen"?
Was there a preclass homework assignment to click and treat to establish
the connection?  (I wouldn't start off with that recall anyway, as I think
it's raising the criteria too far too fast).

Quote:
> Okay, I'll try to stop expressing my extreme dubiousness about the
> trainer, and stick to the method...

I am a great believer in the method but I'm already dubious about the
trainer myself!

Quote:
> It seems to me that, even by starting in a minimum distraction
> environment, with a colossally appealing food treat, and working your
> way up through baby-steps of gradually increasing the distraction level,
> and gradually decreasing the frequency of the food treat, whenever the
> dog hears the word "come", he is still making a decision along the lines
> of "am I that hungry? is it that tasty? or would I rather chase this

No, not really.  Just as with the collar correction there must come a time
when it's HABIT and the dog just does it.

The procedure I usually hear recommended is that you start on-leash or in
a contained, boring space without a lot of distractions.  Chasing the dog
down and showing the lure may fall under the "do anything you have to,
just get the behavior" category but it's really unnecessary.  

The reason you want to control the environment is not so you can catch the
dog and correct it, but simply to ensure that nothing other than coming to
you is particularly profitable.  To avoid giving corrections if it's a
habit, some trainers will have you tie the leash to your belt.  If the dog
is right away going to play with the other dogs, and playing with the
other dogs, and getting all kinds of dogattention and such... that's fun
stuff that isn't coming from you.  And that will undermine the recall.

Whereas if you keep reinforcing coming to you, even before you put a name
to it, and somehow the dog never succeeds in reaching anything that seems
more interesting than coming to you (or at least, not until after he's
come to you), you will get a more and more reliable recall.  ("not until
after..." I know of a terrier whose recall was perfected using
squirrel-chases as the primary reinforcer.  Now, when he sees PREY! that
used to distract him totally, he does a no-command recall -- so he can get
permission to chase.)

Regarding the dog deciding that something else is more important than
food, which should not be offered in training, examinations of wolves show
that many of them suffer broken bones in the pursuit of prey.  This has
got to be more memorable than a collar-pop, yet if the animal is making a
reasoned comparison, it is to say that "food is more important to me than
a broken bone".  If a few collar-pops or e-collar nicks can forever stop a
dog from chasing a deer or rabbit... there's something else going on here.
Something different.

Quote:
> I envision a situation where a choke-collar "do it and I praise you
> exuberantly, or else I'll enforce it and you'll have to do it anyway"
> dog, and a clicker-trained "I really like the food treats, and the
> praise, but nothing happens if I don't obey" dog, each see a rabbit and

You need to include the clicker-trained "I really like the treats, and the
praise, and the other stuff I get, and I know that there is no profit in
chasing rabbits 'cause there never has been without permission" dog.  the
correction trained dog may say "okay, here I come" but the aforementioned
terrier, after traditional long-line training and e-collar training
supervised by a professional, said "Later, babe, I'm BUSY!" (and took the
correction with terrier elan).  The dog who knows that there's no profit
in it actually doesn't think at first, just turns and comes in;  THEN he
thinks "Ok, ok, I was good, I was good, can I chase him now huh huh?  Can
we chase him together?  Let's chase him together, c'mon, let's!!".

And actually if any of the three stops to think, with the truck coming and
all, that's it.  For the true safety recall you are aiming for an
unthinking response.  ;-)

Quote:
> If your dog only sees the rabbit, and you see the rabbit AND the Mack
> Truck, and your dog sees pleasant consequences in obeying you, but
> nothing compared to the pleasure of chasing the rabbit, and no negative
> consequences of *not* obeying you, barring missing a small tidbit of
> liver or something... well... it just doesn't seem the best way to
> achieve the intended result.

See, this is what I mean:  if the dog actually mulls through all of this,
it's not gonna happen.  

Quote:
> When I asked this question of my trainer during Session Two yesterday,
> she said that negatively correcting your dog (using a ***collar for a
> correction):
> 1) caused your dog to fear you
> 2) caused the dog to label unrelated but coincidental things as "the bad
> thing" you were correcting.

You're right, not necessarily -- but these are possible side-effects of 1)
incorrect timing or overcorrecting and 2) the same as 1 + the dog's
individual nature.  A collar pop is usually directed in a way that helps
the dog not overgeneralize or decide the wrong thing, but this effect can
be seen with aversives in general.

Quote:
> I am not disputing that. I am questioning the *degree* of
> *effectiveness* and the *circumstances* of the effectiveness.
> She did say it takes longer (I can see this already).

I'm not sure it should if it's done right.  Some things go much faster
with clicker training (starting a new behavior), others go slower, and I
think it evens out.  But I could be wrong on that, would welcome input
from other clicker trainers on that.

Quote:
> And she said that "Dogs get into a routine - they are creatures of
> routine. If they get into the routine of coming when you call, then they
> will do it no matter what."

It's true, though I say "habit" where she says "routine" and someone else
might use another term.  I don't know a single one of us humans that would
COME in that tempting situation without a noticeable pause to think, the
pause that gets the example dog run over and a pause that trainers don't
like to see.

Quote:
> Habit versus a fascinating distraction? Come on! *Obedience* versus a
> distraction, I'm willing to bet on. That is, if I train a dog that IF I
> say 'come', he MUST come (not through fear, or pain, but just plain, he

But if you genuinely believe the dog is weighing the alternatives, "I can
come and be praised, or not come and be corrected" what is that if not
fear?  (I don't think it's fear or fear of pain, but I don't think the dog
is weighing the alternatives either.)

Quote:
> So can someone please explain to me, non-flamingly thank you, the actual
> psychology of the dog in the situation I'm describing? Why would a
> clicker/treat-trained dog choose to ignore a rabbit and come to you when
> you call him, if, in fact, he'd rather chase the rabbit? I'm having a

Ideally, if rabbit chasing is this dog's weakness, because rabbit chasing
has been used as a reinforcer:  you can't get to the blasted rabbit
without permission, but after you do this, there's a good chance we'll
chase it together.  If
...

read more »

 
 
 

Clicker training, philosophy/methodology question

Post by Julia F N Altshul » Wed, 14 Jun 2000 04:00:00


I read your whole post and enjoyed it.  Let me try to break it down into
issues so I can answer them or let others answers them each in turn:

1.  clicker training theory versus traditional training theory
2.  good clicker trainers versus ones that haven't a clue
3.  matching a training method with the dog it goes with
4.  matching a training method with the person it goes with

I'll get to #2 first.  Bottom line is that your clicker trainer has
learned a few buzz words and enough of the whole theory to sound good but
really doesn't have the experience or the knowledge to give that course.  
There are all sorts of mistakes in what she (he? I'll say "she") told
you.  I hardly know where to begin, but these are off the top of my head:

1.  The reward doesn't have to be food.  It has to be something, anything
the dog will work for.  It could be the chance to chase a ball.  It could
be praise.  For my dog, it's biscuits and cheezits.  Some dogs are food
motivated; others aren't

2.  She's totally foggy on the whole concept of shaping behaviors.  You
don't chase the dog down to force a behavior.  You begin by rewarding the
tiny steps that lead to a behavior.  I wouldn't begin with "come," but  
since you use that example in your post, I'll use it here.  Without
giving any command at all, you put the dog in a very small enclosed space
and wait for the dog to look at you or walk a step toward you.  You click
and treat.  Repeat 20 times or more until the dog is walking a step
toward you and expecting a treat.  Then stop clicking and treating for
that and wait for the dog to take 2 steps towards you.  Move around the
room while you do this so the dog understands that he's to walk towards
you and not a magic spot where the chair is.  Slowly move up to making
the dog come all the way to you and sit to get a treat.  Only when that
whole behavior is reliable to do you add the command COME.  There are
better explanations of this in books so I won't try to rewrite the books
here.

Basically your trainer sounds like someone who learned about clicker
training in 5 minutes and then jumped to all the wrong conclusions.  My
boyfriend has tried to mish traditional training and clicker training
together and gotten it similarly all wrong.  Never start with the
command.  Start by shaping the behavior.

On to #3-- You don't say why you were attracted to clicker training.  
Were you having a problem with a dog that led you to want to try
something new?  That's what it was for me.  Sheppe was traditionally
trained and, like your dogs, loved it.  You've never seen a happier more
motivated dog than Sheppe.  We were fairly harsh with her, but all the
leash corrections and alpha rolls in the world only made her more eager
to please.  It was like she was saying "oh, so that's what you meant,
just testing" whenever she got a negative.  Naturally we began that
training with Cubbe based on our previous success with it, but Cubbe
exhibited all the classic reasons people go to clicker training.  Even
the mildest correction got her so freaked that she'd lose control of her
bowels right in class.  She's perfectly, PERFECTLY house trained at
home.  All traditional training did was scare her so much that she
couldn't do anything much less learn.  Clicker training has been the
perfect antidote to that, and she's learning nicely with it.  Would I use
clicker training with my next dog?  Depends on the dog!  I'd be very
suspicious of any trainer, traditional or clicker, who recommends only
one method for every dog.

I'd be especially suspicious of any trainer whose own dog isn't
obedient.  C'mon, what made you go with this lady with so little to
recommend her?  Most trainers boast of obedience trials their dogs have
won or jobs their dogs have earned as therapy or guide dogs.  That's not
the only thing I'd look for in a trainer since I'd also want evidence
that the trainer can impart knowledge to her clients, but it's a place to
start.

You sound like someone with a lot of success with traditional training.  
I'd say to stick with it.

#1 is the most contentious issue.  People here tend to be in love with
their methods, and even the politest among us have trouble not finding
fault with other methods.  (It's best to killfile or ignore the less polite,
but you'll discover that quickly enough on your own.)  If anyone had
witnessed my efforts at using traditional methods on Cubbe he would have
quickly concluded that traditional methods are abusive.  And they ARE
abusive with dogs like Cubbe.  (I can say that because I was the one
trying them and giving up on them.  Cubbe's my dog.)  When I see the lady
in my neighborhood ineffectually getting her dog to do anything with food
bribes, I'd say that her method is abusive.  It doesn't work, and her dog
is going to get killed one of these days.

Good luck and get back to us after you read through a little of the flame
war that your post will invariably lead to.  We all get too little
follow-up.  What's your dog's name?

--Lia

--

 
 
 

Clicker training, philosophy/methodology question

Post by Jerry How » Wed, 14 Jun 2000 04:00:00


Hello Lauren,

The trainer you've got is a bum. Most are.

Treats are not appropriate motivators unless you're working in an
animal act and the animal depends on those treats as part of his
diet. Treats bribe the dog from the gut level... There's no mental
appreciation for the handler. You can spot a food bribed dog a mile
away. Yes, you can wean them from the bribes, but why bother starting
with bribes, when you should be working on commanding your dog's
attention through praise???

You are mistaken about your old ideas, and you are misinformed about
your ''new'' ideas.

I'm not going to spend an hour convincing you. Force as we were
taught kills millions of dogs every year. The BEST dogs will not
knuckle under to force and ''alpha'' ***. The alpha theory is
bunk... *** issues are all in the mind of the trainer, and only
surface when the trainer provokes the dog, and pushes him into a
fight or flight mode, which challenges the dog's survival instinct,
and the dog becomes aggressive, and our EXPERTS kill the dog and
blame it on bad breeding...

I've specialized in behavior problems and protection for three dozen
years, and I use NO force, no confrontation, no punishment, no
bribes, no treats, and no confinement, in training. I get strict
obedience. I have to, or someone is going to get hurt. I'm not
worried about a couple of points in the ring or field trial, I'm
obligated to turn out a safe, friendly, effective, and thouroughly
aggressive family or security K-9. So, all your other criteria and
hypothetical situations pale in comparison to my non force trained
attack dogs in homes with children and at work on high risk K-9
security sites.

Read the information you need to know and are asking about, in the
Wits' End Dog Training Method manual available for free at
http://www.moonsgarden.com/ , and I'll spend as much time answering
your questions or giving additonal advice as you like. You'll learn
to teach or break behaviors using sound sistraction and praise
techniques... You'll learn how your dogs think and learn. You'll
learn to get your dog to keep one eye and one ear on you, and you'll
learn how to make him WANT to do anything you ask.  You'll have an
instant recall in about one hour of work. And, you'll discontinue the
classes you've signed up for...  Jerry.

"Thus we should beware of clinging to vulgar opinions, and
judge things by reason's way, not by popular say." Montaigne

"*** is the last refuge of the incompetent." Salvor Hardin

"If you cannot convince them, confuse them." H.S. Truman.

;~) DRAINING THE SWAMP, AND RELOCATING THE GATORS... J>>>

"CUSTOM WILL RECONCILE PEOPLE TO ANY ATROCITY." G.B. Shaw.

"I know that most men, including those at ease with problems
of the greatest complexity, can seldom accept even the
simplest and most obvious truth if it would oblige them to
admit the falsity of conclusions which they have delighted in
explaining to colleagues, proudly taught to others, and which
they have woven, thread by thread, into the fabric of their
lives."
                                             Leo Tolstoy

Is it any wonder that the following sig file has generated more
complaints to my personal email than any other controversial
post I have made to date, bar none?:

                                            caveat
If you have to do things to your dog to train him, that you would
rather not have to do, then you shouldn't be doing them. If you
have a dog trainer that tells you to jerk your dog around, ***
him, pinch his ears, or twist his toes, shock, shake, slap, scold,
hit, or punish him in any manner, that corrections are
appropriate, that the dog won't think of you as the punisher,
or that corrections are not harmful, or if they can't train your
dog to do what you want, look for a trainer that knows Howe.

Sincerely,
Jerry Howe,
Wits' End Dog Training

http://www.moonsgarden.com/
Nature, to be mastered, must be obeyed.
                      -Francis Bacon-

There are terrible people who, instead of solving a problem,
bungle it and make it more difficult for all who come after.  Who
ever can't hit the nail on the head should, please, not hit at all.
                     -Nietzsche-

The abilities to think, rationalize and solve problems are learned
qualities.

The Wits' End Dog Training Method challenges the learning
centers in the dogs brain. These centers, once challenged,
develop and continue to grow exponentially, to make him smarter.

The Wits' End Dog Training method capitalizes on praising split
seconds of canine thought, strategy, and timing, not mindless
hours of forced repetition, constant corrections, and scolding.
                  -Jerry Howe-

 
 
 

Clicker training, philosophy/methodology question

Post by Lauren Radne » Wed, 14 Jun 2000 04:00:00


Quote:

> >What am I missing, PLEASE???

> >Sincerely, and somewhat frustrated,
> >-Lauren

> Hi :)

> Read your post and found it very interesting (and funny).

Thanks!

Quote:
>   I know
> that in the old school of dog training, there are certain breeds that take
> longer to get it than others, I imagine it's the same thing for the clicker
> training set.

Well, the one particular dog (of my four) that I've chosen to take to obedience
training, was selected by me because he definitely is... *slower*.... than my other
dogs.

My other dogs remind me of the little girl with the blond braids in the plaid
pinafore, who all through grade school, was always leaning forward way over her
desk, half out of her chair, shouting "ooo! OOO! I know! I KNOW!". Even if it was
Home Room.  Whereas the dog I'm taking to class reminds me of the guy with the
shoulder length dirty blond hair who all through high school was always*** out
behind the gym, smoking... remember him? His parents were ecstatic if he actually
got as high as a "D" in anything? Totally unmotivated... didn't ever do anything
really *bad*, no hard ***, just... never did anything really *right*, either.
Eventually ended up as a gopher in a mechanic's shop. Never understood how to fix a
car, either... just was the guy who ran and got the parts and the doughnuts.

The first night of class the trainer did have us "charge" our dogs to the clicker,
that is, click, treat, click, treat, aiming for you clicking and the dog turning
his head or coming over expectantly... as I said in my first post... two hours
later, my dog was *still* oblivious, and MUCH more interested in chasing this one
smallish black labrador puppy wearing a green harness. I remember asking with some
exasperation if I was going to have to *** him to *** to get his
ATTENTION??? Sigh.

Quote:
> In my classes, we have about 8 weeks to turn family pets into
> well behaved family members, and clicker training would probably take way too
> long :)

What you describe is the method and time frame I am used to (and <guilty look left
and right> prefer).

Quote:

> As for your trainers dog.... my students will see my dog and say " I want my
> dog to be like yours"....  and by the end of the 8 weeks, the majority of them
> are behaving quite similar to mine.  And none of them are afraid, scared,
> cowardly or broken of spirit.

In the *other* obedience class I went to, a million years ago, THAT trainer's dog
WAS terrifically obedient AND she actually HAD saved the dog's life through
obedience... the dog was sprinting across the highway after something, a big truck
was coming, and the trainer shouted "DOWN!"... and the dog dropped on the spot. She
said the truck nearly ran over her dog's toes, but her dog didn't move... and
lived.

Quote:
> If you are not happy with this class, please do not feel guilty about switching
> back to the method you have been successful with in the past.  It is not
> abusive; don't let the "postive only" crowd make you feel that way.  You know
> your dogs are happy and not afraid, that right there is proof enough:)

Thanks. Actually, I am <furtive look left and right, again> SECRETLY saying "SCREW
it" to at least *some* of what the trainer has said....  For example, for this,
their second week, we are supposed to be adding the word "sit" into the sequence,
just before the dog sits, then click, then treat. And we are supposed to be
"capturing the down", that is, if your dog happens to lie down (*EVER*, NOT!) then
you click and treat for that. I haven't asked her how this meshes with her
oft-repeated statement that "We want the dog's default behaviour to be that he sits
when he approaches you"... because (1) after all that approach-me-sit-good-dog
reinforcement, why should he do anything else? If he's hungry, he's going to do
"sits" in front of you, until he's not. Then he's going to go do anything else
(assuming he hasn't just been cavorting around in some ADD sort of way the whole
time anyway). I realize it takes patience to wait him out until he eventually lies
down, but....what's wrong with luring him there, as a hint? (2) once the dog gets
the idea that lying down is a Good Thing (TM), then a smart dog is going to run up
and *lie down* in front of you, instead of sitting, right? Because that's the
latest fun and rewarding thing he can do. Anyway, to continue our assignments for
the week, the trainer said that we'd work the hand signal for the sit into the
sequence later, as in, in several weeks. BUT (HA!) as she was saying, "And I'm not
even going to show you the hand signal yet", she *did* it. HAH!

So last night, I said the heck with it. In just one session, I trained him to sit
on the verbal command (something we are supposed to work on this week) AND went
ahead AND taught him the hand command  - which he obeys without the voice command
(!) AND I lured him and taught him to lie down, to do it with the voice command, to
do it with the voice and the hand command, and to do it with just the hand command
alone! And I made up a hand signal for "come", too... so by the time we were
done... I could let him wander away, and when he looked at me, all without me
speaking(!), I gave the hand command to "come", the hand command to "sit" when he
got there (which he was already doing anyway), AND then the hand signal to "down",
and HE DID IT ALL! More than once, too! And I could get him to toggle back and
forth between down and sit, with only hand commands, or with voice commands, even
if *I* was in a wierd position, like sitting on a chair. I ain't sayin' he's
perfect, but he's pretty good already, and after only one session! I'm tired of him
being the class juvenile delinquent.

I feel very triumphant about this. I'm a maverick, but hey, I'm getting faster and
better results than doing what SHE wants! If my dog is ready to learn at an
advanced pace, then I'm not going to slow him down! She told us she wouldn't even
show us the hand commands yet, and she said way back when class first started (a
week ago) that "luring" was "too advanced a behaviour" (for the trainers to master,
she meant) and she wasn't even going to teach us to do it until months down the
road.

Phooey.

Secretly Leading an Underground Rebellion Composed of Me and My Dog,
-Lauren

 
 
 

Clicker training, philosophy/methodology question

Post by Lauren Radne » Wed, 14 Jun 2000 04:00:00


Quote:

> As it's been explained to me (and makes sense, FWIW) is that the
> 'command' or the label for a behavior is the very last thing to be
> applied.  IOW, the trainer's instruction to say "come", then chase the
> dog down and lure it, is not appropriate 'clicker-based training'.

I think, regardless of one's training foundation and principles, that what occurred
on my first night (all the dogs loose and totally wild, while we each called the
totally unknown word "come", once, then sprinted around after them, trying to shove
a treat under their nose and lure them away), is simply bad training by any
standards whatsoever.

Quote:


> > .... I've had a lot of fun watching my
> > dogs catch on almost immediately when I do things like train them to
> > heel by praising them *only* when they drift into the position I want
> > them to walk in, and going silent in mid-syllable if they range out of
> > that specific position while we are walking.

> There's your operant conditioning.

Yes, I know. Physiological Psychology was one of my majors. I'm well acquainted
with Skinner and his principles, and other psychological principles.

Quote:

> > ...it seemed to me that offering food as a
> > training reward would result in a dog whose motivation would be "I'll do
> > it IF you have food,"

> With my dogs, the attitude seems to be "I'll do it because you MAY have
> food, and because you're going to make me do it anyway".  Enlightened
> self-interest, really.

AHA! But that's my point! In my present class there is *NO* "because you're going
to make me do it anyway" aspect.

There is no enforcement. There is no "I asked you to do it, and you didn't, so now
you STILL have to do it anyway, and I'm going to make you".  If BOTH the
choke-collar trainer and the clicker trainer "make the dog do it anyway", then BOTH
trainers are creating, as you put it, a result based on "enlightened
self-interest".

Quote:
> > This trainer, on the very first day, had all of us let loose our totally
> > uncontrolled and untrained puppies and dogs (we were in a very large
> > room, not outside, thank goodness), and then told us to say the word
> > "come", once, then scramble all over after our dogs (who were busy
> > chasing each other with wild delight) trying to shove a food treat under
> > their nose, and lure them back over to wherever we were when we said
> > "come"... then click the clicker, and give them the food.

> I.  Don't.  Think.  So.  I think I'd modify a behavior from the "Choose
> to Heel" crowd and find something terribly interesting off in a corner
> of the room, by myself.  When my dog wandered over to see what was so
> fascinating, THEN he'd get a reward.  No command.  Using a command at
> this stage is just silly.

Except that my dog ran the Kenpuppy Derby 'round and 'round for two hours without
stopping or getting tired (or even caring that when I clicked, he got food put in
front of his nose - I couldn't even *put* it in his mouth, he was so busy pushing
past me to get to the dogs he was chasing), and when we had to put them back on
leashes, intermittently, for something else, all he did was strain, front feet off
the ground, to go right back to chasing the other dogs. There is NOTHING I could
have done, or can do now, that is as interesting, enticing, motivating, and
rewarding to my dog as what he gets out of chasing other puppies. I could have sat
in the corner, cooing and making delighted little sounds to myself for the next six
days, and, as long as there was ANY other dog in the room, *I* was not going to be
a point of interest.

Of course, I agree that using a command at that stage, and in that environment, was
useless.

Quote:

> > I have a pretty good education in psychology, including operant
> > conditioning,

> Does the activity you describe above seem like any kind of operant
> conditioning?  Of the dog, that is - seems like the humans are being
> conditioned pretty thoroughly!

I think this trainer.... needs training....

Quote:

> >...it's been my experience that you never give a command you
> > can't enforce,

> That's one reason for not issuing commands until the critter is
> volunteering the desired behavior on a consistent basis.

Well, I was going beyond that. There WILL come a time when, no matter how
consistently the dog has been volunteering the desired behaviour, for whatever
reasons, he won't feel like volunteering it, directly after you've requested it..
It's my experience that if you don't make it clear that the dog has no choice
("enforcement" doesn't mean viscious, brutal, painful, it could just as easily mean
tedious, boring, tiring), then the dog learns - clearly - that DISobedience has no
negative consequences. Obedience becomes voluntary. How many of us know people with
kids where Mommy and Daddy say, "Timmy? Time for bed, please. Timmy, Mommy and
Daddy would really like it if you went to bed. Timmy, didn't we ask you to go to
bed 15 (30, 60) minutes ago? Timmy please go to bed now. It would make Mommy and
Daddy really happy and we'd praise you." Try doing that when there's a circus in
Timmy's back yard. I don't care how many times he's volunteered to go to bed, when
the little hand is on the "8", and been praised effusively for it, the circus in
his back yard far outweighs Mommy and Daddy saying "Good Boy". At some point, Timmy
will simply say, "No, I don't want to", and Mommy and Daddy can withold all the
praise they want, but that still doesn't result in him going to bed.

Quote:

> >... you don't start training an animal in an
> > area of maximum distraction.

> I haven't had any trouble with this.  I actually had a lot more trouble
> with Sam, using more traditional methods, but that's probably because he
> and I were both new to everything and, well, I was pretty damned dull.

Wow. Your dog must really be focused on you. Mine is not. I've had him only about a
month, and I got him when he was 7 months old, so I'm a relatively uninteresting
stranger, compared to the joys of Everything Else That's Going On In The
Environment.

Quote:
> > She did say it takes longer (I can see this already).

> Nah.  I saw a team of people take a dog and teach it to alert a sleeping
> person to a cry of "Fire!", lead the person to an exit, then lead the
> person to an alternate exit.  Took all of an hour to teach the dog this
> rather complicated behavior.

I'm going to bet that this was a dog who had already had a background of "clicker
training" and learning new behaviours through reinforcing progressively more
precise approximations of the desired behaviour.

I'm also willing to bet that learning such new and complicated behaviours is pretty
darned fun for this dog.

He already knew - and enjoyed - the game. AND, at that time, it WAS the most
entertaining game to be had in the environment.

Now take this dog, go to your local feed store, buy 30 half-grown chickens, drop by
the pound, pick up 6 cats, shoot by your wildlife rehab center and borrow a skunk,
three rattlesnakes, a pair of ducks, seven squirrels, and a baker's dozen of
rabbits, take a return route by the Children's Petting Zoo and stuff a dozen or so
pygmy goats in your car, and a llama or two, set them all loose in the same room,
and THEN start to teach your dog a complicated behaviour. Maybe YOUR dog would
actually pay attention to YOU, but I think MOST dogs would NOT.

Quote:
> What I would do, if it were me, is teach the dog that 'click' indicates
> some kind of reward.  (You didn't mention any such introduction to the
> clicker.)

I didn't mention it, but, yes, in fact, the evening started that way. And we
returned to it  several times after a series of disasterous puppy-chasing circuses.
And my dog was not in the LEAST interested in any of the NINE different food treats
I brought, OR any of the ones the trainer was able to provide. ALL he wanted to do
was get over THERE to THOSE OTHER DOGS.

Quote:
> Then, I'd get my clicker and my cookies (or ball, whatever)
> and the dog and I'd hang out somewhere that the dog *will* be somewhat
> distracted, but safely enclosed.  And I'd wait.  When the dog looked at
> me - click! <treat>.  NO COMMAND.  That's how it starts.  Gradually,
> increase the requirements for reward - from a look to a step, to two
> steps, to wandering past, to coming to you.

I did this with success at my own home, later. I still think the trainer's
introduction to the concept was disasterous. I also think that asking your dog to
"come" - a word he doesn't understand to begin with - on day one, in such an
environment, was nuts.

Quote:

> An alternate strategy is to use a form of backward chaining.  Train and
> reward for a straight sit in front of you.  Then change distances and
> angles until the dog can find "front" from anywhere.  Then call for a
> 'front' (or 'come' - whatever) when the dog is moving.  There's your
> recall.

The dog still has to exhibit the behaviour so you can reward it.

My dog never did.

That's my point.

Even what ought to have been simple and enormously rewarding "I click, then I feed
you! Such a deal! You don't even have to DO anything! But I bet, eventually, when I
click, you'll at least LOOK my way, in anticipation!"... absolutely no deal. And
no, I did NOT feed my dog just before going to class, either.

Yes, I overcame this, and my dog is behaving in ways that make me MUCH happier
now... BUT... there is STILL not a single motivator that outweighs him chasing down
other dogs and playing with them.

The method *I* am used to, would have *prevented* that as an option. "No, you're
attached to this leash, and just plain can't GET over there, so pay attention."
This trainer is hoping that for some bizarre reason the dog will suddenly, or
eventually tire, of this magnificent pleasure, and feel like coming over to *you*
for food or something. Meanwhile, until that time, he gets to do EXACTLY as he
pleases.

Quote:

> It is at THIS point (when the dog has learned the behavior, and
> understands what is

...

read more »

 
 
 

Clicker training, philosophy/methodology question

Post by Julia F N Altshul » Wed, 14 Jun 2000 04:00:00


I left the computer, gave Cubbe a walk, got dinner ready, thought about
it some more and realized that your post was still making me mad even
after I'd posted a long response.  So here's more:

Your question is:  "Why would a clicker/treat-trained dog choose to
ignore a rabbit and come to you when you call him, if, in fact, he'd
rather chase the rabbit?"  

Good question.  I can't climb inside a dog's head so I can't answer it
for sure, but I can try.  Here's a companion question:  "Why would a dog
who's been trained to heel on a leash with the help of leash corrections,
continue to heel off leash where he knows there's no threat of a
correction?"

My guess is that habit has a lot to do with it.  I believe both methods
work because the dog gets into the habit of doing what's commanded.

My second guess is that the dog respects the human partner as alpha no
matter how alpha-ness is established.  Some owners establish that they're
alpha with body language, by rolling the dog over and belly up (gently or
harshly).  Some owners establish that they're alpha with physical force
or by yelling or punishing somehow.  Some owners establish that they're
alpha by controlling all resources such as food, water, treats, chances
to go outside.  But the bottom line is that our dogs respect us and do
what we want because we're alpha.  As I said earlier, my alpha style with
Cubbe is extremely gentle because it's obvious that she responds to that,
but I'm still alpha.

My biggest trouble with clicker training was every beginner's mistake.  I
was trying to do way too much, too fast.  Short lessons of 5 minutes each
are the way to go unless you're both having a great time.  (Cubbe runs at
full speed for 20 minute clicker sessions now.  We stop when she gets
tired.)  Each lesson might concentrate on a single tiny shaped behavior.  
As soon as I broke the lessons into tiny mini-steps, we went much
faster.  

I recommend Karen Pryor's books for the theory end of clicker training
and Morgan Spector's _Clicker Training for Obedience_ for a more
practical guide.  I think there's a market for an even simpler and more
practical clicer training book that would give highly structured lesson
by lesson instructions for people who are truly new at the whole concept
of shaping behaviors.  There are some things we learn by understanding,
things that we have to know the theory for before we "get" them.  But I
believe clicker training is one of those things that we learn by doing.  
It all comes clear after we've been doing it for a while.  Skip the
theory to start with, and just start observing the dog while he figures
the whole thing out.

--Lia

--

 
 
 

Clicker training, philosophy/methodology question

Post by Mary Heale » Wed, 14 Jun 2000 04:00:00


Quote:

> I think, regardless of one's training foundation and principles, that what occurred
> on my first night (all the dogs loose and totally wild, while we each called the
> totally unknown word "come", once, then sprinted around after them, trying to shove
> a treat under their nose and lure them away), is simply bad training by any
> standards whatsoever.

You got that right!

Quote:

> > With my dogs, the attitude seems to be "I'll do it because you MAY have
> > food, and because you're going to make me do it anyway".  Enlightened
> > self-interest, really.

> AHA! But that's my point! In my present class there is *NO* "because you're going
> to make me do it anyway" aspect.

Either (as seems likely) your current trainer is a dipstick with no
clue, or it is possible that enforcement is being held off until the
dogs display some indication of understanding.  From your description,
it'll be a long wait.

Quote:
> ...If BOTH the
> choke-collar trainer and the clicker trainer "make the dog do it anyway", then BOTH
> trainers are creating, as you put it, a result based on "enlightened
> self-interest".

Yup.

Quote:
> ...There is NOTHING I could
> have done, or can do now, that is as interesting, enticing, motivating, and
> rewarding to my dog as what he gets out of chasing other puppies.

You might be surprised.  Or not.

Quote:
> It's my experience that if you don't make it clear that the dog has no choice ...

Once the dog understands what you're requiring him to do, noncompliance
is not an option.  Before then, compulsion can sometimes be informative,
but is most often punitive.

Quote:
>...At some point, Timmy
> will simply say, "No, I don't want to", and Mommy and Daddy can withold all the
> praise they want, but that still doesn't result in him going to bed.

"At some point".  Yes.  But not when the 'going to bed' routine is
introduced, and not if little Timmy doesn't know the difference between
'go to bed' and 'go for a ride' <g>.

Quote:
> > >... you don't start training an animal in an
> > > area of maximum distraction.

> > I haven't had any trouble with this.  I actually had a lot more trouble
> > with Sam, using more traditional methods, but that's probably because he
> > and I were both new to everything and, well, I was pretty damned dull.

> Wow. Your dog must really be focused on you.

Velcro.  I haven't been to the bathroom alone in 10 years.

Quote:
> Mine is not. I've had him only about a
> month, and I got him when he was 7 months old, so I'm a relatively uninteresting
> stranger, compared to the joys of Everything Else That's Going On In The
> Environment.

I still think you might be surprised.  One of the really consistent
things about dogs is that they usually keep track of their human(s).
Even if they look, for the most part, like they'd rather be with anyone
else, Mommy steps out the door and Puppy starts to wonder.  Especially
rescued or rehomed dogs.

I didn't used to think Sam gave a flip whether I was around or not.  He
wanted to play with the puppies, and schmooze with the people, and chase
squirrels and do all that stuff.  In my sight.  If I "disappeared", he
would frantically search for me.  When I reappeared, he'd go right back
to pretending to ignore me.  But I'd never make it out the door twice...

Quote:
> > > She did say it takes longer (I can see this already).

> > Nah.  I saw a team of people take a dog and teach it to alert a sleeping
> > person to a cry of "Fire!", lead the person to an exit, then lead the
> > person to an alternate exit.  Took all of an hour to teach the dog this
> > rather complicated behavior.

> I'm going to bet that this was a dog who had already had a background of "clicker
> training" and learning new behaviours through reinforcing progressively more
> precise approximations of the desired behaviour.

Nope.  Dog had been introduced to clickers on Friday evening, and the
routine was trained on Sunday afternoon.  What takes awhile is the
human's ability to time the clicks accurately, break the behavior into
sufficiently small parts, that sort of thing.  (Oh, that Friday evening
we had about 15 ACDs of both sexes and all ages running loose in the
training building.  Nobody tried to shape a recall, though.)

Quote:
> I'm also willing to bet that learning such new and complicated behaviours is pretty
> darned fun for this dog.

Hard to say.  ACDs can be such serious dogs.  Mine aren't, but they can
be.  One of the reasons I started clicker training with Noah is because
he's a very soft dog.  Routine reassures him, as I've said, but
conversely new things stress him.  Light's on, nobody's home.  By the
end of the weekend, he was in there working for the click like a normal
dog.

Quote:
> He already knew - and enjoyed - the game. AND, at that time, it WAS the most
> entertaining game to be had in the environment.  ...ALL he wanted to do
> was get over THERE to THOSE OTHER DOGS.

Sounds like the right reward for the circumstances.

Quote:
> ...I still think the trainer's
> introduction to the concept was disasterous. I also think that asking your dog to
> "come" - a word he doesn't understand to begin with - on day one, in such an
> environment, was nuts.

I agree with you, as long as you're blaming the trainer and not the
method.  The method's sound enough.

Quote:
> But... that's in a calm and normal environment. *IF* you released the menagerie I
> suggested above in his bedroom at 7:59, I'm willing to bet you the cost renting
> them all that he does NOT go to bed at 8PM, and does NOT start fights either.

He'll chase them out and go to bed.  He already ejects the cats, picks
on the other dogs, and he'd evict the birds if it ever occurred to him
that they'd go away.  

--
Mary H. & the Ames National Zoo:  Raise a Fund ("Regis");
ANZ Sam-I-Am, CGC; ANZ Noah Doll, CGC; ANZ Babylon Ranger;
felines, finches, fish and Guinea pig

 
 
 

Clicker training, philosophy/methodology question

Post by DOGTV.COM NETWORK » Wed, 14 Jun 2000 04:00:00


After a quick read, it is apparent that you are a prime
candidate for the Dog Game Methodology, featured in the
Brand NEw ViDeo BITE THIS!"

All of the frustrations you are having with clicker training
will be washed away.  In the Dog Game Methodology, you can
use food or not use food, you can use any kind of collar you
want.  It is open source dog training which doesn't attempt
to shackle the trainer or the dog.

It was used to set a world record for offleash training with
a 7 month old Golden Retriever, and a three year Old
Newfoundland, both of whom were completely untrained
previously and both of whom star in the video.  When you
BITE THIS! you will get much more detailed information and
education on how to train dogs not to bite INSTANTLY, and
how to train dogs offleash INSTANTLY without being dependent
on any one implement, treat or "school of thought"

Think of the way Michael Jordan played basketball.  There
was always more than one way to get to the basket. That's
what made him so great.  The ability to change strategies in
mid air, mid court or mid anything when the situation
warranted it.

That's what the Dog Game Philosophy/Methodology is all
about.

You will not find a better video for what you are looking
for anywhere else.

This is Michael
Reporting Live...
http://www.moonsgarden.com/
Free your mind and your dog will follow.

Quote:

> Hi, All,

> First of all, I apologize from the bottom of my heart if this topic is a
> "groaner" ("oh, NO, I can't believe some stupid newbie is asking THAT
> again, HERE we GO with this STOOPID thread ALL OVER AGAIN!!! Sigh...").

> That said, here's my question....

> I have had four dogs of my own I trained over a span of twenty-plus
> years - and other people's dogs, and other species of mine and others,
> that I've trained as well -  but let's stick to dogs... Anyway, the dogs
> I've trained have, for the most part, been trained with what apparently
> is now viewed as "the bad old methods" of using a ***collar. For
> example, to teach the dog to heel, you would give the command, and walk
> along and, if - after gentler methods were tried, the dog strained ahead
> or was inattentive, you would change directions abruptly, firmly
> repeating the word "heel", and the dog would get yanked and surprised,
> and figure out pretty quickly that he had better pay more attention to
> you (you crazy unpredictable human being) than whatever else was so
> interesting a minute ago.

> At the moment, I have four relatively new dogs at once, of various ages,
> and mostly to motivate myself to really get them obedience trained
> *now*, and mostly because one of them is already the size of a percheron
> horse, and three times as strong, I have signed up for a (very
> expensive) six month course of obedience training that (of course,
> mostly depending on how much work *I* put into it) is supposed to lead
> to a pretty advanced level of trained dog by the end of it all.

> This trainer espouses the "clicker" operant conditioning method of
> training.

> I am trying to be a Good Dog myself, and I am all for training a dog
> with positive reinforcement, etc.... I've had a lot of fun watching my
> dogs catch on almost immediately when I do things like train them to
> heel by praising them *only* when they drift into the position I want
> them to walk in, and going silent in mid-syllable if they range out of
> that specific position while we are walking. So it's not like I think
> you have to club a dog to death if he doesn't obey your command to
> "glosnick" the first time, when he obviously doesn't have a clue what
> you mean.

> However.

> I am having some philosophic difficulty with this trainer's methods that
> I am obediently trying to use with my dogs....

> Before I even met this trainer, I talked with her on the phone, and
> expressed my concern that it seemed to me that offering food as a
> training reward would result in a dog whose motivation would be "I'll do
> it IF you have food," and "I'll do it because YOU can please ME by
> giving me food", not "I'll do it because I can please YOU and you're the
> pack leader, and I understand the rules, and besides I love to please
> you". It also was my observation that food-rewarded dogs tended to obey
> more strongly if there was a food reward, and less emphatically - or
> perhaps not at all - if there was no food reward.

> I was concerned a dog trained this way would, in essence, base his
> degree of obeying on whether you had food, whether he was hungry, and
> whether the food was desirable enough to compete with whatever the dog
> was already interested in at the time.

> The trainer said that none of this was an issue, if you were properly
> trained in how to "fade" the food rewards from the obedience process. I
> was (and am) extremely dubious, but trying to keep an open mind.

> This trainer's *own* dog, a three-year old which this trainer has had
> since a pup, will not lie down and stay on command, but drifts away
> within a few minutes, to be coaxed and cajoled back eventually, when the
> trainer notices. Also, this dog jumps on EVERYBODY over and over, which
> I associate with a *puppy* behaviour problem, not an *** three year
> old dog (belonging to a trainer, no less).

> This trainer, on the very first day, had all of us let loose our totally
> uncontrolled and untrained puppies and dogs (we were in a very large
> room, not outside, thank goodness), and then told us to say the word
> "come", once, then scramble all over after our dogs (who were busy
> chasing each other with wild delight) trying to shove a food treat under
> their nose, and lure them back over to wherever we were when we said
> "come"... then click the clicker, and give them the food.

> I have a pretty good education in psychology, including operant
> conditioning, and more experience in training animals than most people I
> meet, and it's been my experience that you never give a command you
> can't enforce, or you train the animal it can ignore you. And it's also
> been pretty obvious to me that you don't start training an animal in an
> area of maximum distraction.

> I kept my jaw from dropping at how this trainer handled this first night
> of training, but my dog was NOT interested in ANY kind of food treat
> that I, or anybody else, could provide, because, of course, NOTHING was
> as much fun as chasing some other puppy. For two hours, no less. I
> actually lost two pounds when I stood on the scale the next day, so I
> guess there was *some* benefit, but I don't think there was any for the
> dog - at least, not in the area of obedience training.

> Okay, I'll try to stop expressing my extreme dubiousness about the
> trainer, and stick to the method...

> Here's one of the questions I've had already (it being only the first
> week), and I didn't get an answer that really convinced me from the
> trainer....

> We were supposed to spend the week (and I did) training our dog to,
> among other things, "come"... and the method was to say the word once,
> then chase the dog down and wave the food under his nose and lure him
> back over some distance away, and then click, and give him the food. If
> the dog doesn't come, and/or isn't interested in the food, and can't be
> lured... then... nothing. You just keep trying to get his attention with
> the food he isn't interested in. It's your fault for not having
> interesting enough food (or the right pitch of squeaky toy, or appealing
> enough belly rubbing techniques, or whatever).

> It seems to me that, even by starting in a minimum distraction
> environment, with a colossally appealing food treat, and working your
> way up through baby-steps of gradually increasing the distraction level,
> and gradually decreasing the frequency of the food treat, whenever the
> dog hears the word "come", he is still making a decision along the lines
> of "am I that hungry? is it that tasty? or would I rather chase this
> rabbit? Because if I don't go over there now, it's no big deal... I'll
> get that treat later when I feel more like obeying."  I am under the
> illusion, or delusion, that the 'bad old methods' result in a thought
> process more like, "Yes Ma'am! Right away Ma'am! Even though maybe I'd
> rather not, Ma'am!".

> I envision a situation where a choke-collar "do it and I praise you
> exuberantly, or else I'll enforce it and you'll have to do it anyway"
> dog, and a clicker-trained "I really like the food treats, and the
> praise, but nothing happens if I don't obey" dog, each see a rabbit and
> start to chase it. The rabbit sprints across the highway, and both dogs
> follow it. A Mack truck is bearing down on both the dogs. The
> choke-collar dog owner says, "Rover, COME!" and Rover says, "Damn! And
> that rabbit was so interesting. Okay, here I come!", while the clicker
> dog owner says, "Bingo, COME!" and Bingo says "Yeah, maybe, in a minute,
> but I am TOTALLY distracted by this WAY cool rabbit and I know for a
> fact that there is no food treat on this WORLD that is more interesting
> to me right now than chasing this rabbit!"

> It seems to me that at least one of the aims of dog training is so that
> you can save the life of your dog ("don't eat that <poison>", "come away
> from that <rattlesnake> right now", "don't move <further into
> traffic>"), and therefore, obedience training should not be a Democracy.
> If your dog only sees the rabbit, and you see the rabbit AND the Mack
> Truck, and your dog sees pleasant consequences in obeying you, but
> nothing compared to the pleasure of chasing the rabbit, and no negative
> consequences of *not* obeying you, barring missing a small tidbit of
> liver or something... well... it just doesn't seem the best way to
> achieve the intended result.

> When I asked this question of my trainer during Session Two yesterday,
> she said that negatively correcting your dog (using a ***collar for a

...

read more »

 
 
 

Clicker training, philosophy/methodology question

Post by Lauren Radne » Wed, 14 Jun 2000 04:00:00


Wow, Julia! What a refreshing reply! Thank you!


Quote:
> I read your whole post and enjoyed it.  Let me try to break it down into
> issues so I can answer them or let others answers them each in turn:

Thank you, and thank you.

Quote:
> 1.  clicker training theory versus traditional training theory
> 2.  good clicker trainers versus ones that haven't a clue
> 3.  matching a training method with the dog it goes with
> 4.  matching a training method with the person it goes with

> I'll get to #2 first.  Bottom line is that your clicker trainer has
> learned a few buzz words and enough of the whole theory to sound good but
> really doesn't have the experience or the knowledge to give that course.

She *says* she trained under someone, for whom she has great respect, who is an
expert, etc., etc... however, it is my conclusion also, that she is in over her
head when it comes to conducting a class on this stuff. At *this* point, I would
love to ask her exactly how many classes she HAS conducted, but it would be
pretty obvious to her why I am asking... my dubiousness about her and her
methods is not sufficiently suppressed that she isn't aware of it.

Quote:

> There are all sorts of mistakes in what she (he? I'll say "she") told
> you.  I hardly know where to begin, but these are off the top of my head:

> 1.  The reward doesn't have to be food.  It has to be something, anything
> the dog will work for.  It could be the chance to chase a ball.  It could
> be praise.  For my dog, it's biscuits and cheezits.  Some dogs are food
> motivated; others aren't

THANK you! In my first phone call with her, AND my first meeting, I circled back
to the "can't I reward my dog with effusive praise instead?" concept... Barring
some strange accident, I will always have the ability to voice my praise, yet I
may not always have a clicker in my hand or food available (example, both hands
full of grocery bags, "Rover *sit*!"). Also, I would rather the dog was working
to please ME, not to EAT. The odds that he will tire of pleasing me are far
lower than the odds that he will tire of eating, or tire of this particular food
taste.

After that first session ("charging the dog to the clicker", and the disasterous
"come" session <chase-chase-chase-chase-chase- wave treat - chase-chase-chase>,
I went back and tried the clicker food thing on my other three dogs. Two are
FRANTICALLY eager to please me, they ADORE me, and they are INCREDIBLY smart.
Both of their performances DEgraded when the "prize" became food, instead of my
approval. The third is very hard-headed and independent, and - although she is
an absolute *glutton* - after two weeks, I stil have not found ANY food that she
is willing to *perform* for....including when she's hungry. But praise works, at
a moderate level of reinforcement.... unless something else interests her. And,
again, my 'class' dog, neither food, nor squeak, nor praise, nor storm, nor dark
of night, catches his attention, period.

Quote:

> 2.  She's totally foggy on the whole concept of shaping behaviors.  You
> don't chase the dog down to force a behavior.

I Agree, Agree, Agree!

Quote:
> You begin by rewarding the
> tiny steps that lead to a behavior.  I wouldn't begin with "come," but
> since you use that example in your post, I'll use it here.  Without
> giving any command at all, you put the dog in a very small enclosed space
> and wait for the dog to look at you or walk a step toward you.  You click
> and treat.  Repeat 20 times or more until the dog is walking a step
> toward you and expecting a treat.  Then stop clicking and treating for
> that and wait for the dog to take 2 steps towards you.  Move around the
> room while you do this so the dog understands that he's to walk towards
> you and not a magic spot where the chair is.  Slowly move up to making
> the dog come all the way to you and sit to get a treat.  Only when that
> whole behavior is reliable to do you add the command COME.  There are
> better explanations of this in books so I won't try to rewrite the books
> here.

I agree! And in further support of your theory that she's been exposed to
someone *else* who knows how to do this, but doesn't know it well enough to
*teach* it herself, in the orientation class, she showed me exactly that
sequence of training her dog to do some new, arbitrary behaviour, which the dog
was *vibrating* with eagerness to figure out... that *was* very impressive. On
the other hand, "regular" obedience behaviours by this dog (come, sit, stay) or
dull, slow, ignored, and/or broken (ie, lying down but breaking the stay command
at will).

Quote:

> Basically your trainer sounds like someone who learned about clicker
> training in 5 minutes and then jumped to all the wrong conclusions.  My
> boyfriend has tried to mish traditional training and clicker training
> together and gotten it similarly all wrong.  Never start with the
> command.  Start by shaping the behavior.

At this point, I'd *already* rather be taking this class from *you*, than from
this *other* person. sigh.

Quote:

> On to #3-- You don't say why you were attracted to clicker training.

I *wasn't* attracted to clicker training. I wanted to *obedience* train my four
dogs, including the huge juvenile delinquent who, at 7 months weighed 64 1/2
pounds, casually ripped those spiral-into-the-ground-dog-stakes out of the
ground, and has uprooted the rooted-in-cement-blocks baseball backstop I made
the mistake of chaining him to. I wanted to obedience train him before he
remodelled my house the hard way. I'm also relatively fond of having my
shoulders continue to reside in their resident sockets.

I saw two different adverti***ts - one for free training if you had a dog from
the pound (they are), one at a pet shop. Turns out they were the same lady.
Turns out the free training was a 2-hour lecture, that wandered in all
directions and communicated very little, starting with a lengthy soapbox speech
about how pets end up at the pound. Since the only attendees were people who had
just RESCUED pets from the pound, I thought this diatribe curiously misdirected.
I have been a professional trainer (of human beings) for many years. I do have a
clue how a training presentation should go. I've also trained and shown both
dogs and horses. This woman does not impress me favorably. BUT... it *does*
provide ME with a six-month weekly motivation for ME to train my dogs, work with
them, and get their behaviours where I want them, rather than procrastinating. I
*have* picked up *some* things from her... there is SOME value in this for me...
but probably not as much as I paid, and, again, I see that my questioning of her
methods has already raised her hackles... and I predict we will be in smiling
dislike of each other well before the six months is up. BUT... I do intend to
milk this for what I can get out of it, AND it *is* creating an environment
where *I* am motivated to train all four of my dogs to be good citizens.

I was very dubious of the clicker-food method of reward when first I talked with
her on the phone, and asked her if, please, I couldn't just reward my dogs with
praise, since they seem to respond to that VERY WELL. The end of that
conversation amounted to, basically, "fine, you do it your way if you want, but
when you don't get results as fast, I get to say I told you so."   Except I'd be
willing to bet on my own juvenile delinquent dog's behaviour over HER dog's
behaviour ALREADY, and I intend to make it *really* *obvious* as the class goes
along, how VERY obedient he becomes, as contrasted to her own dog. That is, I
intend to excel her using her own methods, and slide a few of my own "hey it
works, dammit" in behind the scenes.

Quote:
>   Sheppe was traditionally
> trained and, like your dogs, loved it.  You've never seen a happier more
> motivated dog than Sheppe. ... all the
> ...corrections ...in the world only made her more eager
> to please. ...  but Cubbe
> exhibited all the classic reasons people go to clicker training...All
> traditional training did was scare her so much that she couldn't do anything
> much less learn.  ...Would I use clicker training with my next dog?  Depends
> on the dog!  I'd be very suspicious of any trainer, traditional or clicker,
> who recommends only one method for every dog.

This (although I snipped it, I hope I left the heart of it intact) was *Very*
reassuring to read.

Quote:

> I'd be especially suspicious of any trainer whose own dog isn't
> obedient.

I had about a gadzillion red flags waving before my eyes before her orientation
even started!

Quote:
> C'mon, what made you go with this lady with so little to
> recommend her?

Well... mostly to get *myself* committed and started RIGHT NOW, rather than
trying to cram a lot of research and class visits into an already packed
schedule. But... it *does* seem that the class quality degrades with every
session...

Keeping in mind what related to you about the strength of my puppy, this
trainer's recommendation for how to prevent him from pulling on the leash, is to
simply wind MYSELF up in his leash like a maypole, so that *EVENTUALLY* he is
wound up tight against my side, and redirected. Gad, after she demonstrated this
with some little weeny dog the first time, I was SOOOOOO tempted to ACTIVELY
train my dog to deliberately wind the leash low around the ankles, then
yank-like-heck!... and then ask her to demonstrate. Yes, that was intensely evil
of me. Yes, I resisted doing it. But I have two objections to this maypole
routine. One: the dog pulls and strains on the leash, and there is no
*immediate* correction, *clearly related* to the bad behaviour. The dog
pulls-and-pulls-and-pulls and eventually ends up uncomfortably wound up tight by
your leg, still pulling like crazy, or maybe circling 180 degrees until the
attraction comes back into sight, and then he goes back to
pullingpullingpulling. Two: EXCUSE ME? This dog has uprooted BACKSTOPS out of
CEMENT. How DIM is it to wind MY legs up tight ...

read more »

 
 
 

Clicker training, philosophy/methodology question

Post by Marshall Derm » Thu, 15 Jun 2000 04:00:00



Quote:
>I left the computer, gave Cubbe a walk, got dinner ready, thought about
>it some more and realized that your post was still making me mad even
>after I'd posted a long response.  So here's more:

>Your question is:  "Why would a clicker/treat-trained dog choose to
>ignore a rabbit and come to you when you call him, if, in fact, he'd
>rather chase the rabbit?"  

>Good question.  I can't climb inside a dog's head so I can't answer it
>for sure, but I can try.  Here's a companion question:  "Why would a dog
>who's been trained to heel on a leash with the help of leash corrections,
>continue to heel off leash where he knows there's no threat of a
>correction?"

 Great job Lia!

Quote:

>My guess is that habit has a lot to do with it.  I believe both methods
>work because the dog gets into the habit of doing what's commanded.

 We Skinnerian behaviorists don't use terms like "habit," to explain
 behavior because our explanations refer to environmental events,
 but "habit" is closely related to what we might offer: drill/practice.

Quote:
>I recommend Karen Pryor's books for the theory end of clicker training
>and Morgan Spector's _Clicker Training for Obedience_ for a more
>practical guide.  I think there's a market for an even simpler and more
>practical clicer training book that would give highly structured lesson
>by lesson instructions for people who are truly new at the whole concept
>of shaping behaviors.  There are some things we learn by understanding,
>things that we have to know the theory for before we "get" them.  But I
>believe clicker training is one of those things that we learn by doing.  
>It all comes clear after we've been doing it for a while.  Skip the
>theory to start with, and just start observing the dog while he figures
>the whole thing out.

 This is why many of us insist on our students learning to train real
 rats in experimental psychology courses rather than simulated rats.

 --Marshall

http://www.uwm.edu/~dermer
____________________________________________________________________

I have read rpdb for over two years. Consequently, I urge newbies to attend
to the civil and rational posts of the rpdb regulars from whom I have
learned much. They include: Amy Dahl, Diane Blackman, Janet Boss, Susan
Fraser, Avrama Gingold, Lynn Kosmakos, Bob Maida, Cindy Tittle Moore, Denna
Pace, Marilyn Rammell, John Richardson, Ludwig Smith, and Terri Willis.

       Marshall Lev Dermer/ Department of Psychology/ University of

                     http://www.uwm.edu/~dermer

        "Knowing how things work is the basis for appreciation,
         and is thus a source of civilized delight."  -- William Safire

 
 
 

Clicker training, philosophy/methodology question

Post by DogStar7 » Thu, 15 Jun 2000 04:00:00


Quote:
>Henh, thanks. And his name is "Koala".... He's a "Shiloh Shepherd"... if you
>go
>take a look at:
>http://www.shilohshepherd.org/webdocs/gallery.html

We have one of those in our club.  He's 10 months old, and I understand exactly
what you mean when you say he'd rather chase the other dogs than listen to a
clicker :)  LOL!  The one in our club is a real sweetie, but he's been quite a
handful.  He's finally starting to mellow a bit, but all he really wants to do
is play with the other dogs.

Dogstar716
Come see Gunnars Life: http://hometown.aol.com/dogstar716/index.html

"AKC papers do not mean you are getting
a quality dog. They are merely a birth certificate. Even puppy
mill pet shop pups have AKC papers" - Bob Maida

 
 
 

Clicker training, philosophy/methodology question

Post by Jerry How » Thu, 15 Jun 2000 04:00:00


Real rats are not real dogs in real families as real pets. Jerry.


(Julia F  N  Altshuler) writes:

Quote:
> >I left the computer, gave Cubbe a walk, got dinner ready, thought
about
> >it some more and realized that your post was still making me mad
even
> >after I'd posted a long response.  So here's more:

> >Your question is:  "Why would a clicker/treat-trained dog choose
to
> >ignore a rabbit and come to you when you call him, if, in fact,
he'd
> >rather chase the rabbit?"

> >Good question.  I can't climb inside a dog's head so I can't
answer it
> >for sure, but I can try.  Here's a companion question:  "Why would
a dog
> >who's been trained to heel on a leash with the help of leash
corrections,
> >continue to heel off leash where he knows there's no threat of a
> >correction?"

>  Great job Lia!

> >My guess is that habit has a lot to do with it.  I believe both
methods
> >work because the dog gets into the habit of doing what's
commanded.

>  We Skinnerian behaviorists don't use terms like "habit," to
explain
>  behavior because our explanations refer to environmental events,
>  but "habit" is closely related to what we might offer:
drill/practice.

> >I recommend Karen Pryor's books for the theory end of clicker
training
> >and Morgan Spector's _Clicker Training for Obedience_ for a more
> >practical guide.  I think there's a market for an even simpler and
more
> >practical clicer training book that would give highly structured
lesson
> >by lesson instructions for people who are truly new at the whole
concept
> >of shaping behaviors.  There are some things we learn by
understanding,
> >things that we have to know the theory for before we "get" them.
But I
> >believe clicker training is one of those things that we learn by
doing.
> >It all comes clear after we've been doing it for a while.  Skip
the
> >theory to start with, and just start observing the dog while he
figures
> >the whole thing out.

>  This is why many of us insist on our students learning to train
real
>  rats in experimental psychology courses rather than simulated
rats.

>  --Marshall

> http://www.uwm.edu/~dermer

____________________________________________________________________

- Show quoted text -

Quote:

> I have read rpdb for over two years. Consequently, I urge newbies
to attend
> to the civil and rational posts of the rpdb regulars from whom I
have
> learned much. They include: Amy Dahl, Diane Blackman, Janet Boss,
Susan
> Fraser, Avrama Gingold, Lynn Kosmakos, Bob Maida, Cindy Tittle
Moore, Denna
> Pace, Marilyn Rammell, John Richardson, Ludwig Smith, and Terri
Willis.

>        Marshall Lev Dermer/ Department of Psychology/ University of

>                      http://www.uwm.edu/~dermer

>         "Knowing how things work is the basis for appreciation,
>          and is thus a source of civilized delight."  -- William
Safire