First I would like to say this is a really great forum you guys have put together here with tons of great information.

Now I will get into my problem. I have just went through a 6 week state course for my dual purpose narcotic/patrol GSD. He is a German import from a very good line, and I went to a national known training kennel for my class. In week 3 of the class my K9 began to start false alerts. He is a passive indicator, so he started to just sit on blank cars. We worked with him by giving him a "no" command and then the search command again. He would usually come off the sit, but sometimes I would have to clap my hands or make some other noise to break him from his concentration on the indication. This problem of false indications started to get better once we introduced more blanks.

He would false on cars, and rooms still, just occasionally though. I thought the problem was corrected once I got back from training, but now it seems like it has gotten worse. My trainer states the K9 may be keying off what I am doing. I am not touching the area I am searching, and I keep a good pace because when I slow down he seems to false more.

My question is what can I do to cut down on the false indications? I do not give corrections when he false indicates, rather I just call him off. I have seen when he is in odor and indicates correctly, it is pretty muchg impossible to call him off the find without giving him his reward. Thanks for your help.

I would first start setting up problems where the dog has to work extended periods before he finds something. For example, I would set up 20 vehicle and not find anything for at least 15 vehicles. If the dog responds on a wrong vehicle, don't even say "NO", just repeat your search command and keep moving. You should not know where the target(s) are. Have someone that understands the problem and tells you immediately when the dog responds incorrectly.

If, as your trainer says, the dog is keying off of something you do, not knowing where the target is absolutely essential in correcting the problem. If you don't know where the target is, your behavior during the search will never become a cue to the dog. It really shouldn't matter whether you touch where search, speed up/slow down etc. Properly trained, the dog is conditioned to respond to the odor, not your actions. That's why it's crucial you don't know where the targets are when you train.

Also ensure you don't condition the dog to "numbers" of targets or the "timing" of targets. If the dog gets accustomed to finding a target every 20 seconds during training, it can almost become ingrained in them to do exactly that. Drug dogs are usually trained differently than we work. During training there is generally a effort to get as many targets as possible, rather than concentrating on just training for situations. How often have you had a blank search during training, that you did not know was blank. The results of that can also be very telling. I would also make sure the dog is not responding to something other than the drug odor. Have you conducted negative tests to make sure the dog isn't responding to plastic bags, or human odor etc. These are just some of the areas I would first look at.

DFrost

Thank you for the reply. I will give you some insight on our last training group session. We set up 8 blank cars, and then the last car was where the find was. My dog sat on three of the eight cars the first run through. On the second run through they told me to speed my pace up and he sat on one blank. I work with two master trainers through NAPCH, and they keep telling me to correct the dog verbally with the "no" command when he sits. Is there a reason not to do this, and just give him the search command?

During training I have worked with various negative training aids i.e. food, toys, duct tape, and dirty clothes. I have to agree with your perspective on it should not matter what I do, the dog should have the conditioned response to the odor. However it seems like the two trainers are telling me that either I am not moving fast enough or he needs more repetitions. Well even when I moved really fast he still sat once, and in addition to this he is not searching as throughouly when I am moving at that pace. I am trying to stay consistent with my vehicle searches that I run the dog around the car twice, just like I would on the road.

I know he recognizes the odor well, I am just trying to figure out what is going through his head when he false indicates. Could it be maybe he just wants his toy and just sits to get it?

This is just do frustrating because he is doing good in all other areas, but I want a reliable dog and will work however hard I need to so that he is at his best. Any other help would be appreciated.

Did you know where the targets were on any of the runs. Did you do any blank problems, not know there were no targets. Your trainers have the advantage of seeing the dog and you work. When talking about a trained dog, not one in training, it's my opinion response issues are often caused by;

1. predictable handler behavior, caused by the handler knowing where the targets are.

2. A timing issue. the dog used to lots of targets in a short time.

3. The dog responding on something other than the drug odor, a proofing problem

4. The dog is just confused and not exactly sure what is wanted out of him.

When your working does the dog ever miss the odor? Do they have ever direct you back to the target.

I don't tell the dog no because I don't make a big deal out of it, I simply tell the handler to keep moving, the handler repeats the search command. In my program handlers have a very small vocabulary - "find it", "check here" and "good boy/girl".
The dog of course gets the primary reinforcement when he properly responds to the target. Ignoring a behavior is a form of correction.

DFrost

I have done both, where I have known where the targets are and where I have not. But mostly I am the one setting the hides lately to get more reps with him.

1)predictable handler behavior, caused by the handler knowing where the targets are
-I have been very careful to not doing anything different when running the dog. I do not touch the vehicle or walk behind him when he indicates on the source.

2) A timing issue. the dog used to lots of targets in a short time.
-I am not sure about this I try to vary it. I vary the number of cars involved, include different items to search, luggage, boxes etc, and a scratch wall.

3. The dog responding on something other than the drug odor, a proofing problem
-I don't think this is the problem, as he has false indicated on cars, rooms, etc.

4. The dog is just confused and not exactly sure what is wanted out of him.
-He recognizes the odor and searches freely when he is off lead and given the command. I am confident that he recognizes the search command and what he is asked to do. He very rarely misses the odor, only on very hard small finds.

I continue to think that he is false indicating simply because he wants his toy. The only thing that is saving me recognizing a false positive is that I can call him off his sit. Plus he gives no head snap like when he passes a source. When he is in odor and indicates he will not break the sit until he is rewarded with his toy.

I am puzzled by this daily, and can't really get any solid answers on how to work or fix the issue other than repetition. I am just not seeing any change in him. I have been suggested to hook him to a long lead and have someone give him a correction when he false indicates. I am not sure how I feel about trying that yet.

In addition I was looking through some older posts where someone stated "go to a primary reward system and start over, the problem goes away." Are they talking about going back to the scratch wall with someone paying the reward from behind the wall?

"But mostly I am the one setting the hides lately to get more reps with him."

In my opinion, that's the most probable cause of your problem. You're got an odor contamination (yours) probably doing too many targets and not enough blank areas (get more reps) and you know where the targets are, which leads to exactly this type of problem.

Try to follow my reasoning about hiding your own targets.

You agreed with me that the actions of the handler should not affect how the dog works, ie slowing, speeding up, stopping etc. With that in mind, when you work a dog where you know where the targets are located, you will not allow the dog to exhibit certain behaviors in areas where you know a target in not located. Behaviors, such as intensive sniffing, lingering in specific area. Speeding up is ok in an area you know there is no target, but yet you would not allow that, albeit subconsciously, in an area where there is a target. By not allowing or allowing these behavior, because you know where the targets are, you are unable to understand what the dog is doing when you train and you don't know where the targets are.

If I were training this dog I would:

1. Ensure the dog knows the odors.

No - train the odor

yes - start varying the length of training and start increasing distance of search with less targets.

Unless a dog is really messed up, I don't like going back to boxes or scent board. To me it's like sending a high school graduate back to kindergarten.

DFrost

I know for sure the dog knows the odors. In training he consistently found the sources in various unknown wall hides, luggage etc.

When you say increase distance with less targets can you provide an example? Are you saying run a car 3 or 4 times if it's blank, and then move to one with a hide?

Unfortunately our group only trains twice a month so in between that I am placing the hides. I use plastic gloves when I place my hides out.

I see where you are coming from about not allowing the dog to exhibit certain actions. I have noticed once he starts to slow up and start intense sniffing I try to keep him moving when I know it's a blank, versus when I know the hide is there I let him go.

"When you say increase distance with less targets can you provide an example? Are you saying run a car 3 or 4 times if it's blank, and then move to one with a hide?"

No no, I'm saying do 15 - 20 blank cars. Never more than one target per vehicle. I train handlers to do the vehicle in one pass, two is permitted but discouraged. If you have the training areas to do rooms, do several blank rooms. sometimes the search is more important than the number of targets a dog gets in a training session. Also don't discount the benefits of open area training, stuff just thrown in the bushes. If you have a training group, each of you set up a problem. If there are three of you in the group, you will always be able to run 2 blind problems. You can work it out. It's in our K9 Manual and the General Order; if a handler hides his own targets he can not count it as part of his required training hours.

DFrost

Something I've experienced, and I'm by no means an expert in the field, so take this with a grain of salt. My dog will indicate on lingering odors if I reuse a hide so I have to try to go out of my way to mix up my training areas so I'm not hiding in the same area twice in a 48 hr period. This also means I need to be careful to avoid porous materials that might hold the scent for longer periods. Are you using cars he is familiar with and has trained with recently?