Cliker ?s

Tricks, obedience, behavior, and more.
User avatar
Semper Bull
Matured Bully
Posts: 450
Joined: Fri Feb 12, 2010 9:42 pm

Cliker ?s

Postby Semper Bull » Tue May 07, 2013 11:50 am

So I took the kids to the zoo yesterday and the keepers were doing some training with the elephants and utilizing clickers. They would have the elephant lift its front leg then the back leg and click each time she did what was asked but did not reward until after a series of the movements was done.

My questions: the clicker is a marker and a signal that a reward is coming, correct? Should you instead ask for a series of movements; ex. front leg up, back leg up and once the series is done then click and reward? Were the keepers possibly using the clicker as marker that the correct movement had been done and not a prelude to the reward?

Now, later in the training the elephant was asked to put its head on the ground and was clicked and rewarded after she had done so.

For application to my dogs, I understand that you can shape behaviors and have done so. I have done a series such a sit, down, sit, down, sit and click and reward after the series is done. I do not click each time the correct action is done since the dog knows what the commands sit and down are.

User avatar
Amie
Moderator
Posts: 17463
Joined: Wed Oct 24, 2007 7:10 pm
Contact:

Re: Cliker ?s

Postby Amie » Tue May 07, 2013 12:38 pm

The clicker means "that thing you did that very instant you heard the click is what will get you a reward." Reward doesn't always mean treat, it just means something the learner wants.

The cool thing (well, one of the many many cool things) about clicker training is that animals start to respond to cues as if they were rewards. So one way to train a chain of behaviors is to start with the last behavior and pair it strongly with the reward (by rehearsing it thoroughly and reinforcing it enthusiastically) and then that behavior itself is associated with good things. In essence, if you want an animal to perform a string of behaviors that are ABCDEFG, you would start by heavily reinforcing G, and then when that behavior is solid, you start teaching F, then ask for G, then reward. So the animal starts to think "I did F! Now I get to do G which will get me a treat!" And then when they've got that, you teach E, and "reward" with F & G which lead to treats. And so on and so forth until the animal goes "woo-hoo, she asked for A, that means I get to do B, then C, then D, then E, then F, then G - AWESOME!!!" Except, you know, in a more elephanty tone.

User avatar
Semper Bull
Matured Bully
Posts: 450
Joined: Fri Feb 12, 2010 9:42 pm

Re: Cliker ?s

Postby Semper Bull » Wed May 08, 2013 12:06 pm

Thanks for explaining that Amie. I figured that it was something like that since the elephant did get a reward at the end of the series and was very willing during the training.

User avatar
UnconventionalLove
Bully Lover 4 Life
Posts: 1131
Joined: Mon Dec 22, 2008 12:08 am
Location: Kalispell, Montana

Re: Cliker ?s

Postby UnconventionalLove » Thu May 09, 2013 9:01 pm

From what I understand this is used with whales and dolphins as well. There are different "schedules" of reinforcement that can be used to get results. :)


Return to “Training and Behavior”

Who is online

Users browsing this forum: No registered users and 8 guests