Everyday dog training methods and their effect.

Curbside Prophet said:

Anytime you talk about emotional responses you have to consider thresholds. Excitement, over threshold, is problematic when that excitement leads to involuntary behavior that is not compatible with what you want. It's also problematic because you can't get to operant behavior (sit at greeting guests, for example) until the respondent behavior (excitement) is addressed. Excitement leads to jumping on guests, which to most of us is not compatible with what we want.

Click to expand...

Okay, that makes sense - though it seems more correct to say the problem behavior is the jumping, not the excitement, or the call it having too low a threshold (I know most owners won't say that, but those conducting the study could "translate" it), but maybe I'm wrong. If excitement is an emotion, it shouldn't be listed with behaviors, imo. Everything on that list follows this (they are all behaviors or behavior chains).

Besides, you can't really reward or punish an emotion, correct? (Since it's not an operant behavior, I believe you taught me this ) So it seems even more out of place on a study about the results of reward/punishment/"miscellaneous" -based methods. They seem to have made that distinction with the separation-related behaviors being on the list, but not separation anxiety (the emotion) being listed.

sparkle said:

" Traditional techniques have used mainly aversive stimuli, either in the form of positive punishment (application of an aversive stimulus in response to an undesirable behaviour ) or negative reinforcement (removal of an aversive stimulus leading to an increase in the performance of a desirable behaviour) (Lieberman 1999). The use of aversive stumuli in training may have a negative welfare effect implications: it is thought to cause suffering (beerda et al 1997). possibly poses health risks (through increased levels of physiological stress), and has been found to be related to aggression towards other dogs (Roll & Unshelm 1997).


The key word is ***** MAY****** have.

which does not equate to ****WILL**** have.

Click to expand...

Not to mention that it seems to avoid/ignore a whole leg of the P side of the scale, Negative punishment (i.e. withdrawing something the dog would find/is finding to be rewarding in response to a behavior). This seems only oriented around P+ since it's focused on aversives, but that's only telling half the "punishment story" imo.

Curbside Prophet said:

As ineffective as negative punishment is, I'd ignore it too. I don't think the results would demonstrate anything of interest studying negative punishment other than it sometimes works and it sometimes has no effect whatsoever. I believe all the average dog guardian needs to understand is that positive reinforcement works very well, and positive punishment can exacerbate the problem at hand, and offers little (if any) to no long term benefit.

Click to expand...

Is negative punishment any more ineffective than positive punishment? I mean, if P+ can actually create problems or make them worse, how much worse off is P- methods?

While I agree the R+ almost always works on dogs, I wouldn't say P- is any more ineffective than P+, except on dog by dog basis. I mean, if I get up and leave for 10 minutes because Wally refuses a command, then I come back and give that command again and get instant and eager compliance - was P- ineffective?

I would say it would be MORE effective than me smacking Wally on the butt or pulling his hair until he yelped, etc, and what negative side effects would me getting up and leaving create - other than it just not working? (And then, if he did it in response to the aversion, wouldn't that be closer R- instead?)

Maybe Wally isn't the "average dog" (what would that be, anyway - just curious?) - but P- is rather effective on him if only to "tell" him he failed to earn the reward.

Of course, P- works when the dog understands the requested behavior - but that would also seem to apply to P+ since no punishment can actually gain behavior.

harrise said:

Ok, I'll ask. ¿How would anyone here go about explaining this to John/Jane Doe dog owner? Arguing in learning theory lexicon is one thing, getting a frustrated wit's end dog owner on board is another.

Click to expand...

Is it really that hard? I think it's just the terms aren't often used, but people use these in all kinds of ways, imo, they don't go "oh I'm P+ you because you didn't get your work in on time!" they just do whatever.

It's not necessary to use the terms, but I don't think it would be all that hard to explain it in "everyday dog owner" language.

Curbside Prophet said:

I think if you've had a long, mutually beneficial relationship with a dog, P- can be effective, especially if used in conjunction with a no-reward-marker. Absent of that long, mutually beneficial relationship it's a waste of good training time. My opinion only.

Click to expand...

That would make sense - the dog would actually have to care that you left him behind/ended interaction with him.

But don't most owners have some kind of good relationship with a dog, other than the utterly clueless and the utterly hostile type owner? Wouldn't most owners have dogs that would actually care if their people stopped interacting with them?

Curbside Prophet said:

Actually, I'd say P+ works better for known behaviors than P-. Say you want sit (a known behavior). Which is more effective? If I sit!, and the dog sits (P+), or walk out the room for a failed cue, I return, and the dog sits? IMO, the one that saves me time is more effective, but I have to define my criteria for you to understand that point.

Click to expand...

Interesting you mention a strongly given command (I'm guessing that's what you were intending - i.e. a harsher/louder/very firm tone of voice), because that used to freeze Wally up and start him to shaking. That's why I went with P- type approaches instead.

Now, I can actually do that - somewhat - but it still shakes him. He gets visably more worried, starts moving slower/looking away (calming signals).