Do You Always Reward?

[poodlefan]

I'm training him to see most as big toy ejectors.

:D pf I found the Susan Salo jumping method helps to build a lot of value for jumps - and/or SG

Yes, that's the way we'll go Barb. Just got to get off my butt and do it.

[fuzzy82]

With intermittent rewards, do you mean that you occasionally ask the dog to do something, and give no reward whatsoever, no praise or anything? And then next time you will reward, and work your way up to maybe rewarding every 5th time or something like that?

I usually reward with treats, but sitting before I open doors is rewarded with the door being opened. I don't make them sit if they don't want to go through the door, but maybe I should, and that would then be an intermittent reward? Sometimes they enjoy going through the door (going outside) and sometimes they don't (going inside).

Sitting until released at the front door is rewarded by going for a walk, but I have been meaning to introduce sitting at the door while I go through it and the dog is left behind inside, so that she doesn't actually get to come for a walk every single time she sits at the door...

And what is a secondary reinforcer, for example? What do you use?

Edited November 11, 2010 by fuzzy82

[Aidan]

With intermittent rewards, do you mean that you occasionally ask the dog to do something, and give no reward whatsoever, no praise or anything? And then next time you will reward, and work your way up to maybe rewarding every 5th time or something like that?

When we reward something every n times, that is a "fixed ratio schedule". What you would be doing in your example is increasing a ratio schedule, n gets larger over time.

A "variable ratio schedule" is where you vary the number of responses before you reward. When people say "intermittent reinforcement" I believe they are usually referring to a "variable ratio schedule". A variable ratio schedule is where n varies over time, you might reward every 2nd, third, tenth, 4th, 7th etc responses. I'm explaining the technical terms because they are well defined, and given that we've gone down this path understanding how they are defined might help.

but sitting before I open doors is rewarded with the door being opened.

That would be a fixed ratio schedule of 1:1, every response that meets your criteria is rewarded. Another name for this is "continuous schedule of reinforcement".

And what is a secondary reinforcer, for example? What do you use?

A secondary reinforcer is anything that provides a reinforcing consequence that isn't a primary reinforcer. Primary reinforcers are things like food, water, comfortable temperatures, safety and sex. Secondary reinforcers are things that take on reinforcing properties by becoming paired with primary reinforcers; so we have things like praise, clickers, and money (for people).

Edited November 11, 2010 by Aidan

[Tapferhund]

Just want to get others thoughts on this.

I always give a reward if my dog obeys a command. It might not neccessarily be a high value reward such as food or a toy, but I do always give verbal praise at least. With recalls I always give a high value reward.

I have been told that it is optimal to only reward intermittently. does this mean you can't use verbal praise somtimes?

I think it just becomes automatic to give verbal praise every time.

[dark angel]

Just want to get others thoughts on this.

I always give a reward if my dog obeys a command. It might not neccessarily be a high value reward such as food or a toy, but I do always give verbal praise at least. With recalls I always give a high value reward.

I have been told that it is optimal to only reward intermittently. does this mean you can't use verbal praise somtimes?

I think it just becomes automatic to give verbal praise every time.

As mentioned previously giving praise every time will result in sloppy actions to commands.One cant praise/reward a sit thats slow or a drop thats sloppy,yet inadvertently praising a dog even verbally for every command done only reinforces that.It also weakens that invisible line where a slow sit a month ago is now acceptable by the owner.

I know there hasnt been one mention of a correction .If a dog is taught a sit and knows that command and then he starts to become sloppy a correction is due.

But then again i guess it depends what the owner wants from the dog and how fast he/she wants the sits stands drops recalls to be

[corvus]

In my mind, a correction is not due when it's not necessary. You get what you reward. If you reward sloppy sits, then how is it fair that the dog gets punished for doing sloppy sits? It's easy enough to just shape back to what you want by introducing criteria and just not rewarding sits that don't meet that criteria. You just have to be careful about frustration.

[poodlefan]

Just want to get others thoughts on this.

I always give a reward if my dog obeys a command. It might not neccessarily be a high value reward such as food or a toy, but I do always give verbal praise at least. With recalls I always give a high value reward.

I have been told that it is optimal to only reward intermittently. does this mean you can't use verbal praise somtimes?

I think it just becomes automatic to give verbal praise every time.

Sadly this is not that common. The number of times I have to cue handlers to verbally reward their dogs in class is legion.

[bedazzledx2]

What she said

In my mind, a correction is not due when it's not necessary. You get what you reward. If you reward sloppy sits, then how is it fair that the dog gets punished for doing sloppy sits? It's easy enough to just shape back to what you want by introducing criteria and just not rewarding sits that don't meet that criteria. You just have to be careful about frustration.

[fuzzy82]

Just want to get others thoughts on this.

I always give a reward if my dog obeys a command. It might not neccessarily be a high value reward such as food or a toy, but I do always give verbal praise at least. With recalls I always give a high value reward.

I have been told that it is optimal to only reward intermittently. does this mean you can't use verbal praise somtimes?

I think it just becomes automatic to give verbal praise every time.

Sadly this is not that common. The number of times I have to cue handlers to verbally reward their dogs in class is legion.

Agreed, I know people with dogs and they keep telling the dog to do something, like lie down quietly at their feet rather than investigate something that is worrying or interesting, and doesn't get so much as a 'good dog', just gets completely ignored while the owner keeps playing in their phone or chatting to other people.

I tried suggesting making things rewarding for the dog, and pointed out that running around investigating stuff is more rewarding than lying quietly at the owner's feet (which the dog did when asked, but jumped up a minute later and went back to what it had been doing) and the dog will do what is most rewarding, and she got so offended and started a rant about how she rescued the dog from a previous friend of hers and had done her best with it, and who am I to "criticise" what she's doing with it. Attempting to tell her that it was just friendly advice was pointless. And this is a really close friend of mine and we normally get along really really well.

Edited November 17, 2010 by fuzzy82

[corvus]

Sometimes people naturally use a clumsy kind of R- to get their dogs to behave. Grump at them until they do the right thing, then ignore them. Of course, do that to a sensitive dog and you systematically shut it down so it's afraid to do anything.

[poodlefan]

Sometimes people naturally use a clumsy kind of R- to get their dogs to behave. Grump at them until they do the right thing, then ignore them. Of course, do that to a sensitive dog and you systematically shut it down so it's afraid to do anything.

Got one of these in my agility class at the moment... very frustrating to teach. Dog is super high drive but rapidly learning to ignore handler and self reward.

[Aidan3]

If a dog is taught a sit and knows that command and then he starts to become sloppy a correction is due.

But then again i guess it depends what the owner wants from the dog and how fast he/she wants the sits stands drops recalls to be

I'd be looking at why the trainer was so confused about when to reward and when not to in this case.

[Paul Bright]

When starting training with a new command, I reward with treats, As the dog/puppy learns the command and later on masters it, I reward verbally or with a pat. I start out with treats as this easily and surely catches their attention.