1 2 3 4
The click doesn't have to be associated with a food reward, though food rewards work very well with clikers because ... it must be of high value to the dog, the dog must want it, and it must always be delivered.

Right. Doesn't matter if it's food, a thrown ball, running to the handler to play tug- the click signals that a reward is coming.
If you want to fade clicker, you fade the number of times you use it and increase the chain of ... 6, then 12 before getting clicked. But the click still ends the behavior and the dog still gets the treat.

Yep!I wanted, btw, to clarify the comment I made about clicking several times rapid-fire while the dog (Rocsi, in case you hadn't guessed) is running the weaves- IF she had stopped when I clicked the first time, I'd have rewarded. Her response to the click, however, was to accelerate towards the end; so I clicked again for the increase in speed, and for the forward drive. It's also important to note that I was working with guides on the weaves, and that she was NOT working for food, but for a thrown tug toy.

When I reward for sucessfully completed weaves (with or without a clicker), I throw the toy as the dog exits the weaves. So, on hearing the click, she continued driving to the end to get her reward. That was her choice, and I wasn't about to MAKE her pop the poles if she chose instead to go to the end. If she had stopped on hearing the click, I'd have thrown the toy where she was, of course.
Not at all the same thing as clicking in the middle of the course for the A-frame, then expecting the dog to do five more obstacles.
Some people will, say, ask their dog to do an agility course, and click for a correct aframe, click for ... using the clicker as a keep going signal. But in doing so, they are diluting the power of the tool.

Among other things, if you ask for several different behaviours, it dilutes the message sent by the first click- the dog will most likely become confused as to exactly what it was clicked for.
If I want to reward mid-course, I stop, reward, and then continue.
If you click a good weave entry but then expect the dog to complete the weaves before getting a treat, what happens if the dog pops out at the 5th pole? Do you still treat?

Yes- the dog still has to be rewarded for the behavior you marked.

If
you do treat, you have now just rewarded a bad behavior through the primary reinforcer (treat/toy).

Um, no, you haven't. The dog is being rewarded for what it was doing at the moment you clicked, not for what it did afterwards- it's not all that different from clicking for a good entry, then allowing the dog to pop to get the reward. It's still best not to ASK for anything more after the click, though, because as you say it dilutes things.
So clicking and a quick reward (and there's all kinds of debate as to whether where you deliver the reward is important) is key to successful clicker training.

Yep.
you do treat, you have now just rewarded a bad behavior through the primary reinforcer (treat/toy).

Um, no, you haven't. The dog is being rewarded for what it was doing at the moment you clicked, not ... It's still best not to ASK for anything more after the click, though, because as you say it dilutes things.

Acutally there's huge raging debate on clicsport right now about this, between the people who take your position and the people who say that placement and delivery timing of the primary is very important. I tend to sway toward your side. One of the things that got the conversation going is that early on I clicked Cala for speed on the dogwalk, and didn't care if she hopped the contact to get to her treat. Big discussion ensured as to whether delivering the primary (the treat) after popping actually would reinforce popping. I don't think so, but some pretty big names in operant conditining do (and other big names don't. People disagree. Go figure!)
Um, no, you haven't. The dog is being rewarded for ... the click, though, because as you say it dilutes things.

Acutally there's huge raging debate on clicsport right now about this, between the people who take your position and the ... think so, but some pretty big names in operant conditining do (and other big names don't. People disagree. Go figure!)

AHA!!! So people disagree, eh? Then you must ALL be DOG ABUSING THUGS!!!
I think it is probably a good sign when people disagree on the particulars. How brains and motivations and communications work are still not well known enough to have everything be cut and dried. Until it gets to that point, absolute statements on something that has a lot of aspects to it is more of a red flag to me. That's where taking the things that have been learned and applying them to see what works in your specific complex mix of circumstances and personalities and motivating and communication factors makes sense to me. It is also why it is not very productive for people to assume something that has worked for others won't work for you if you haven't tried it yet.

Paula
"Paula talks tough, and she wears vicious lipstick, but she lacks the depth of hate that I have spent many years cultivating." The Avocado Avenger
Many shepherds are not very food motivated. Toys or balls can usually be their turn on. The only disadvantage is ... reps in. I have seen Shepherds learn to respond to food - just depends on what you want to do.

Yep. I sat on the floor handfeeding Java his kibble for every meal for 2 weeks to build food interest. There are times now when I wish I hadn't :-(
Lynn K.
Acutally there's huge raging debate on clicsport right now about this, between the people who take your position and the people who say that placement and delivery timing of the primary is very important.

I agree with Gail Fisher's differentiation between "training with a clicker" and "clicker training". In my humble opinion, lure and reward is actually still at the heart of the training if position and delivery timing of the primary is very important. (Not that there's anything wrong with that...)

But using the clicker as a true "event marker" is quite different. When it really becomes a bridge between the precise behavior it captures (think photo caught by the click of a camera shutter) signaling that the primary has been earned, "payday" can be more time and space independent.

One way to make sure you can use the click as a true event marker is to practice some "Doggy Zen" (a la Shirly Chong), wherein in order for the dog to get the treat, she has to avoid the treat. Simply started, hold out a treat in a closed hand and click the dog for looking away from it. I did this with my food-crazed retrievers simply so as not to get constantly mugged for food. What a pleasure it is for me to be able to put a bowl of treats down in plain sight and have them concentrate on their actions instead of slobbering over the bowl!
Gail calls this (yes, it's in her article in The Clicker Journal issue with Shammie on the cover ;->) "getting the food out of the trainer's hand and out of the dog's mind", and instead... " usung the clicker to provide information to the dog in a clear, unambiguous and precise manner".

How do you know if the dog is getting this "freeze frame" message? Gail says if you click early on sits, the dog should start trying to hover with butt in the air at the spot the click came. I don't know about that, but I do know that if you have a clicker savy dog, you can shape really cool things because subtle muscle shifts and postural alignments will be repeated if clicked, even in the absence of prompt in-place treat delivery.
And you can also chain and back-chain much easier with dog who has had a steady diet of event markers and variable reinforcement as opposed to meticulous ">placement and delivery timing of the primaryplacement and delivery timing of the primary<"!! LOL!!
Susan Fraser, owned and trained by
BeBop a Lu SheBop SH, Shamma Lamma Ding Dong MH,
Semper Choo Choo Ch'Boogie, and Gris Gris Gumbo Ya Ya http://mypeoplepc.com/members/chinchuba/AuH2OK9s /
Gail calls this (yes, it's in her article in The Clicker Journal issue with Shammie on the cover ;->) "getting the food out of the trainer's hand and out of the dog's mind"

There's a correlation to this in attention work when food has been used to teach attention. It's always enlightening to have people stand in front of their dogs with food in their hand at their face, then give their attention command while moving that hand to the side at shoulder level. Tells you pretty quickly whether the dog really understands attention or not :-)
Another similar one is to hold treats out from your sides on a recall and see if the dog maintains attention on your face or not all the way in. Funny how many opportunities we get to go back and do some more work :-)
you have a clicker savy dog, you can shape really cool things because subtle muscle shifts and postural alignments will be repeated if clicked, even in the absence of prompt in-place treat delivery.

Yes! Sometimes what you want from a click is to make the dog conscious of something they did unconsciously.
Ever seen a dog do a whole obedience run through, get clicked once at the completion, and then run from the ring for a big fat celebratory primary reinforcer orgy? Now, THAT (imo) is proper ">placement and delivery timing of the primary<"!! LOL!![/nq]
Damned right! I recently got a whole bunch of exercises from Dawn Jecs on getting away from the primary reinforcer. Things like leaving treats on a chair and running away from the dog, then both running back to the chair to reward after the dog caught up to you. It was enlightening to me to see how much motivation was created by getting the primary reinforcer away from me, and how that increased motivation has held up well over a couple of months time.
Lynn K.