Jealousy
Does anyone know about (or have experienced) their dog being jealous with people (not with other dogs)? My kishu has this problem. It's not aggression, but specifically jealousy. I would like to find out how to get her not to be this way. When she does get jealous of me paying attention to someone, she tries to bite them in the ankle--even if she is not in her own yard and at a neutral site.
Comments
(Don't take her on dates LOL)
As for the attention getting behavior, there was an episode of It's Me or the Dog about it. Their problem occurred mainly inside on the couch, but I think the technique they used would be effective for you too. Here's how I remember it going. Put your dog on a leash for control purposes, but you don't need to be holding it. Place her in a sit-stay and move 2-3 feet from her. Give a guest some attention for a few seconds. If your dog holds her sit-stay, leave the guest and reward her either with a treat or some affection of her own. If she doesn't hold her sit-stay, give her a correction and move her back again. Repeat this a bunch of times, slowly increasing the distance and the duration of time she is in the sit-stay.
The goal of the training is to reinforce to her that good things happen when they are patient and wait their turn for attention. Rather than ignore or correct the attention seeking behavior and not give to your guest, reward her for seeking attention in a polite manner. With this approach you aren't trying to end the jealousy, but teaching that asking nicely to be pet too is ok.
All resource guarding is based on fear. All aggression is based on fear. Even gaurding a property is a form of resource guarding and stems from fear - that is why most LGDs are not socialized at all, so they are extra fearful, and therefore guard at a higher intensity.
Never never never correct a dog for acting fearful. The only way, IMHO, you can correct this issue is via classical conditioning [more info]. You have to change your dogs physiological association with people being around you.
FEED YOUR DOG WHEN THIS HAPPENS. This is the fastest way to alter the negative connection your dog has with people being around you. The idea is that feeding her when she is doing this will reverse her physiological reaction to the situation, making it a positive thing when people come near you.
Also, you may want to read this article: "Why You Should Always Praise an Aggressive Dog" by Lee Charles Kelley
----
Would operant conditioning tell you that offering a treat to a dog that is exhibiting fearful behavior reinforce that?
There seems to be a contradiction there, and I am not sure how to reconcile these two things. Do you ignore a dog that is acting afraid so as to not reinforce the behavior, or do you offer your dog a treat to show the dog that the situation is not that bad?
I think i like the idea that you address the dog's emotional state, rather than the behavior, a little bit better. My hope is that if you started this way, then you could tell when the dog was beginning to act up, not because of fearfulness, but because he wants those treats, then you can work on the behavior. This is all an academic discussion for me right now though.
http://www.theotherendoftheleash.com/you-cant-reinforce-fear-dogs-and-thunderstorms/
"If she doesn't hold her sit-stay, give her a correction and move her back again."
The "correction" here should be a body block or a stern "NO" or "AAHHTT" at worst, not a leash jerk. If she breaks her sit-stay, divert her away from your guest by moving in front of her and blocking her approach. Then place her back a few feet away and try again.
Operant conditioning contradicts Classical conditioning in that it doesn't distract the dog when "praising" so you are just reinforcing the dogs fears.
Classical Conditioning reminds me of Victoria Stilwel's use of the 'watch me' command. Something scary comes by, 'watch me', dog doesn't pay attention to scary item, dog gets treat.
The "watch me" command both reinforces the act of looking at the owner AND classically conditions the dog to looking at the owner's face. It's a win-win and one of the things Jen and I work on most with our LGD and Blue (as well as the others, but it's less likely they will hurt a stranger).
----
On the topic of classical conditioning (but doesn't necessarily apply 100% to the OP's situation)...
There is something important to understand about food. When a dog eats they cannot be stressed.
That is why once a dog is past his/her threshold they will not eat. With fear, and classical conditioning, the technique of feeding is often referred to as "spilling" or "open bar / closed bar". The idea is that you "spill" food to the dog until the fearful event is over, and then "close the bar" (stop feeding).
----
For this situation...
In this example, IMHO, the path would be to feed the Kishu until they are eating from you comfortably, then ask them to prefer a very basic command, like sit, and reward them for that. So, as the dog approaches her threshold and gets to the point of biting I would start feeding, and once you see that they are no longer as close to their threshold I would then start operant training. It would be important to focus on asking them to do the very most simple thing they know, like sit, and reward them for it. Once you can have them sit by you when a stranger approaches w/o "spilling" then I would up the criteria, to maybe a down, or a "watch me". w/e.
The basic principle is that once a dog is close to or past their fear threshold they are no longer learning, so feeding them is not teaching them anything (because they are too close or past their threshold and therefore cannot learn), it's just altering the chemicals in their brain that allow them to come back and focus.
You may also want to increase the distance between the Kishu and the stranger. you could increase that distance to the point where they are no longer past threshold (with Kahuna this was almost exactly 6 feet - it was really interesting - we could adjusted by only inches as we reconditioned him) then you could work them (have them sit or "watch me") and praise them for working because they are able to learn.
Keep in mind that a dog doesn't generalize well, at all. Also keep in mind that you want to "feed in position". So, if you were training a dog to site, using a clicker and treats, you would want to click at the exact moment they "sit" and then feed them in that position. The click marks the moment or behavior and the food reinforces it. If you clicked when they sat, and then fed them when they were standing, the dog is not getting the full message and may be confused.
So, in the example above, you would be reinforcing the dog looking at you or eating treats from you and not reinforcing the general situation of "being aggressive when people approach my person".
I am no expert tho, this is just my opinion
Also, not to promote LCK too much, as I know there are many other trainers/behaviorists out there that are good and could shed light on my points, but here is another article he wrote on this very subject:
"Positive Mental Associations v. Reduction of Tension" Part 1 and Part 2
By Lee Charles Kelley
----
Prior to Emma's class, I was driving myself to tears trying to work Sage at what I *thought* was a reasonable exposure level, but it was really enlightening to see how far you have to break it down in to such baby steps for them. Keeping the dog successful is the key, little baby steps, short durations, long distances. Reaction Failure is not an option because it sets you back so far. It is hard to accept because you want to just take a simple walk, stand and talk with a neighbor, but it's always a "working" walk and I have to be with Sage first and foremost and secondarily with the neighbor we meet. The conversation must be aborted if I feel that Sage may be getting anxious (risk of fail), and it's never what seems like far enough to get any real exercise. So we work the exposure, and then come home and play frisbee!