The Limits of Clicker Training

A leash correction or correction from a e-collar is just that, a correction, and regardless of how many fancy ways you say it, a correction is negative and the word itself means just that.
***a·ver·sive   
adj.   Causing avoidance of a thing, situation, or behavior by using an unpleasant or punishing stimulus, as in techniques of behavior modification.***

 

clarification,,,in operant conditioning  punishment is in the behavior of the receipent.  In the case described,, the pops were not aversive/punishment because the target behavior did not decrease or stop.

if you've really told the dog what *to* do first, then an aversive like a scat mat or bark collar is the only humane thing to do--for everyone. You've got to be able to communicate completely with a dog, and I don't think it's OK to limit yourself to +R when there are a small handful of situations that really do call for well-placed and well-executed +P.

So to stick with the original intent of the post, I think that's a (small) limitation of clicker training.

I gotcha. I've talked about this with my friends in the past - I call it the slow band-aid versus the quick band-aid. You can peel the band-aid off so slowly that it doesn't hurt, or you can grit your teeth and zip it off. Both get the band-aid off equally well, but the quick band-aid will hurt. I'd tend to go the first way with my kids, unless time were of the essence for some reason.

ETA quote again [8|] LAte nights are getting to me!

A leash correction or correction from a e-collar is just that, a correction, and regardless of how many fancy ways you say it, a correction is negative and the word itself means just that.

 
It has nothing to do with fancy wording, it is just basic operant conditioning classifications.
 
Positive means you are applying something; negative means you removing or withholding something.
 
Punishment means you are reducing a behavior; reinforcement means you are increasing it.

I like the idea of a sticky with just the basic scientific terminology such as houndlove posted.
 
Then there is a clear, common oint of reference. []
 
 

So are we saying that giving a dog a leash pop using a choke, a pinch collar, or by e-collar is not negative to the dog? 

Yes, if behavior continues to occur in the situation in which a "commonly labeled" punishement is applied, then it was not an aversive.  Some thing is reinforcement or punishment based on its impact on the behavior.
 
It is only negative if the behavior that occurred immediately prior to the pop, decreases in frequency or disappears completely (over time).  There are plenty of dogs who have "collar corrections" who still pull their people all over the place.  It is only punishment or an aversive if it stops behavior.  If kids get yelled at but still misbehave then the yelling did not function as punishment/an aversive.

But there are those who believe that distracting, redirecting, shaping, or manipulating the dog's behaviors (symtoms of instability, frustration, or confusion, IMO) is all you need to do, and that "social learning", leadership, fulfillment by design, has little or no value.

It depends on who you are talking to, and their personal choices and/or beliefs. []

 
Who are "those"? I'm just curious if you are addressing any particular people, or making a general comment about "clicker trainers". Because certainly just because people clicker-train overt behaviours doesn't mean they don't use any of the above concepts as well! In fact most clicker trainers I know use ALL of the above!
 
And as for teaching what "NOT" to do in dogs, can someone provide an example? It's awfully hard to address such a broad topic when there are going to be, literally, a hundred different answers depending on what you are trying to teach. So for those who have asked this question, why not provide an example or two and us clicker trainers can tell you how we'd deal with it. Because in a lot of times, an incompatible behavoiur (or a DRO), paired with extinction and P-, are more than sufficient for working on just about any behaviour in dogs (and most other species).
 

  So are we saying that giving a dog a leash pop using a choke, a pinch collar, or by e-collar is not negative to the dog? 

There is no question that these things are aversive to the dog - aversive being that they cause some sort of pain, intimidation, fear, or frustration. In your everyday language, yes, these things would be considered "negative" = bad, unwanted. However, in the actual behavioural terminology, these are all called Positive Punishers (that is, if they actually decrease a behaviour. If they don't decrease a behaviour then it's simply manhandling, not a punisher).
 
Positive = applying something
Negative = removing something
 
Perhaps you should read up some more on your operant conditioning terminology so you're on the same page as those who are discussing the quadrants. []
 
Kim

  So are we saying that giving a dog a leash pop using a choke, a pinch collar, or by e-collar is not negative to the dog? 

There is no question that these things are aversive to the dog - aversive being that they cause some sort of pain, intimidation, fear, or frustration. In your everyday language, yes, these things would be considered "negative" = bad, unwanted. However, in the actual behavioural terminology, these are all called Positive Punishers (that is, if they actually decrease a behaviour. If they don't decrease a behaviour then it's simply manhandling, not a punisher).

Positive = applying something
Negative = removing something

Perhaps you should read up some more on your operant conditioning terminology so you're on the same page as those who are discussing the quadrants. []

Kim,  You hit the nail on the head - excellant post.[sm=clapping%20hands%20smiley.gif] 

I just wanted to add to what mrv said, that if you aren't impacting behavior the technical terms "punishment" or "reinforcement" don't apply, but that doesn't mean that these things have no impact on the dog. It just means they haven't had an impact on that particular behavior. But if you're collar correcting a dog up one wall and down the other and it is having absolutely no impact on whatever unwatned behavior you're applying corrections to, it may not be a "positive punishment" but it may be nagging or in extreme situations, abuse. And it may cause more reactivity, more fear, more distrust, or more just in general riled-up-ness. And if you're throwing treats at a dog but that too is not impacting the behavior you're trying to modify, again that's not technically "positive reinforcement" but it can impact the dog in other ways, the dog may become spoiled, demanding, whatever. Though in general the fallout from poorly applied positive reinforcement is generally less dangerous for both dog and handler than the fallout from poorly applied positive punishment (I'd rather be around a spoiled dog than a fear-aggressive dog any day of the week). That's why, in the hands of non-professionals, I'd rather see the former being screwed up than the latter.

Then again aversive might not be the right term either, depending on the context in which it is used. Because the term aversive implies (from aversive conditioning) that the animal will work towards avoiding it.

How about, instead of "There is no question that these things are aversive to the dog ", we say:
There is no quesiton that these things are stressful to the dog.

It's probably more accurate than using the term "aversive", although then again stress is such a broad term, depending on the context in which you define it. *sigh* The joy of semantics.

Kim

There is no question that these things are aversive to the dog - aversive being that they cause some sort of pain, intimidation, fear, or frustration. In your everyday language, yes, these things would be considered "negative" = bad, unwanted. However, in the actual behavioural terminology, these are all called Positive Punishers (that is, if they actually decrease a behaviour. If they don't decrease a behaviour then it's simply manhandling, not a punisher).

Positive = applying something
Negative = removing something

Perhaps you should read up some more on your operant conditioning terminology so you're on the same page as those who are discussing the quadrants.

Kim

Thank you, that makes it nice and clear.

And to respond to Houndlove's post, you mention a really important point (to me), which is that +R is less damaging when incorrectly applied than +P incorrectly applied. There is a place for +P in dog training, hopefully it plays as small a role as possible, but like you said - fear aggression from improper +P is much worse than the inconsistencies from improper +R.

LOL. I often have to remind adoptive families that stress can result from pleasant experiences too - stress being simply arousal which has a negative effect on well being.

Sometimes you have to give in to the limitations of language and we'll have to accept that terms like "aversive" and positive and negative have technical as well as common meanings. It would simply be helpful to footnote which we intend when we have technical discussions. For instance, "I don't think punishment is the most effective approach to fear aggression - by punishment I mean the common definition." More writing but it would save all the posts and cross posts trying to clarify respective meanings. Yes, just call me Pollyanna . . .[]

LOL. I often have to remind adoptive families that stress can result from pleasant experiences too - stress being simply arousal which has a negative effect on well being.

 
That was my point, that stress occurs in all facets of life, so to use the broad term would be not entirely correct either. [] Negative stress vs. positive stress? Good stress vs. bad stress? Then what defines those terms?
 
Guess on diverse sites like this we'll have to play "walking dictionary" and define the words that might be used in a number of contexts. I'm just so used to others knowing the meaning behind what I'm saying when I say it, when there is no confusion. But when people aren't aware of the use of certain terminology, that is when confusion arises.
 
But it's an interesting thing to keep in mind, so that I can begin to realize when people have no idea what I'm truly talking about.
 
Kim

LOL. I often have to remind adoptive families that stress can result from pleasant experiences too - stress being simply arousal which has a negative effect on well being.

I've always wondered about statements like "don't do X because it will cause stress for your dog." Well, depending on the dog and he circumstance, stress can come from so many positive AND negative things that I think it is hard to say that "stress" should be avoided at all costs. Lucy shows some signs of stress shen we are clicker training a new behavior and she is trying to figure out what I want. Does that mean training is bad for her? Learning almost anything new - even if it is something we really enjoy can cause stress. I LIVED for my hunter/jumper lessons - but there was definitely stress involved when I was mastering the very skills that made riding enjoyable for me.