Drug-sniffing dogs are cheating cheaters hoping for steak

[Fletch]

Everything we "know" about drug sniffing dogs is apparently poppycock.

The Mind of a Police Dog - Reason Magazine

TL;DR version: The dogs are more interested in pleasing their handlers than finding drugs, so they tend to cue on the handler's body language. If the handler thinks there might be drugs, the dog is likely to ignore its own nose and signal anyway. Researchers have done some pretty solid experiments to back this up.

 

[E5RANGER375]

Everything we "know" about drug sniffing dogs is apparently poppycock.

The Mind of a Police Dog - Reason Magazine

TL;DR version: The dogs are more interested in pleasing their handlers than finding drugs, so they tend to cue on the handler's body language. If the handler thinks there might be drugs, the dog is likely to ignore its own nose and signal anyway. Researchers have done some pretty solid experiments to back this up.

I believe this. when my GSD (who has no training as a drug dog, but still the right breed) knows im mad at her, she will tap the door handle with her nose (her signal to go pee & very pleasing to me since she doesnt pee on the floor, lol), but when I get up to put her out she doesnt wanna go. they got to my dog

 

[Kirk Freeman]

Oh, we've known that faith in Officer Chompy is utter rubbish for some time now, or as I call sniffer dogs "the phrenology of the 21st century."

Problem is that thanks to DoJ Drug War money the dogs are widespread and the prosecution has so much riding on the dog testifying correctly so courts are unwiliing to hear evidence of fabrication (the 100% success rate, the "no fail" rule, and my favorite, now they are claiming dogs smell pills even when not trained to do so), even after I point it out on every single cross. We keep swinging.

I think a steady drip is what will finally opens the eyes of judges and defenstrate Officer Chompy from the courtroom.

 

[shibumiseeker]

I spent a few years helping train SAR dogs with some of the best folks in the business. The two hardest problems are:

1. Having the handler trust the dog. The dog is right more often than not by far.

2. Having the handler NOT lead the dog to alert falsely.

Among the best of the best the tests are to have the dogs be able to operate without being able to see or hear the handler. Like many things though, 90% of the teams (dog and handler) out there are not up to that level and range anywhere from sort of ok to dismal. It takes constant, near daily training, to keep a team up to snuff and very, very few handlers will train to that level.

I've seen teams do amazing things. My exwife's FEMA level 1 dog could find me anywhere on my 150 acres of woods and trees where I work outside all day long, in 15 minutes or less. I literally could not hide from this dog.

I've watched cadaver dogs find bodies under 10' of swampy muck. I've seen SAR dogs find survivors deep in rubble. I've watched drug dogs in double blind tests find trace amounts of drugs double sealed in mylar.

Those are the exceptions though. 99% of the teams I've seen couldn't find their asses reliably.

 

[dross]

I spent a few years helping train SAR dogs with some of the best folks in the business. The two hardest problems are:

1. Having the handler trust the dog. The dog is right more often than not by far.

2. Having the handler NOT lead the dog to alert falsely.

Among the best of the best the tests are to have the dogs be able to operate without being able to see or hear the handler. Like many things though, 90% of the teams (dog and handler) out there are not up to that level and range anywhere from sort of ok to dismal. It takes constant, near daily training, to keep a team up to snuff and very, very few handlers will train to that level.

I've seen teams do amazing things. My exwife's FEMA level 1 dog could find me anywhere on my 150 acres of woods and trees where I work outside all day long, in 15 minutes or less. I literally could not hide from this dog.

I've watched cadaver dogs find bodies under 10' of swampy muck. I've seen SAR dogs find survivors deep in rubble. I've watched drug dogs in double blind tests find trace amounts of drugs double sealed in mylar.

Those are the exceptions though. 99% of the teams I've seen couldn't find their asses reliably.

Yet, these dogs are authorized to override the Fourth Amendment. It's almost kind of a Bizarro World thing, isn't it?

 

[Kirk Freeman]

It's almost kind of a Bizarro World thing, isn't it?

Money is not bizarre; it is the essence of the War on Drugs. Too much money riding on the furry little paws of Officer Chompy. To question dog sniffs is to begin to question the entire War on Drugs.

It will take time for sniffer dogs to be seen in the proper light and perhaps an extraordinary case. E.g. look at how high and holy fingerprints were held until the Brandon Mayfield Debacle. Heck even years after Brandon Mayfield I had the tech from Lowell (ISP) tell a court that "fingerprints have never been wrong". She said she made never heard of Brandon Mayfield (obviously, or just did not want to hear).

 

[rambone]

 

[Ramen]

Awww!

 

[Fletch]

I believe this. when my GSD (who has no training as a drug dog, but still the right breed) knows im mad at her, she will tap the door handle with her nose (her signal to go pee & very pleasing to me since she doesnt pee on the floor, lol), but when I get up to put her out she doesnt wanna go. they got to my dog

If I hold up a treat, one of my dogs will go through his entire range of tricks with zero prompting, hoping that one of them will be the magic action that releases the treat from my hand. It's not difficult at all for me to believe that service dogs do the same stuff.

 

[E5RANGER375]

Ive never liked the fact that they can bring out dogs and just throw out your rights based off of an animals response. I would bet a full paycheck on it, that somewhere in the country theres at least one department who teaches their dogs to hit on every car, or atleast has a signal to tell the dog which one they wanna search (probly one of those speed trap towns in GA). no way i will ever support ONLY a dog indicating on a car as reason to search.

 

[ljadayton]

If I hold up a treat, one of my dogs will go through his entire range of tricks with zero prompting, hoping that one of them will be the magic action that releases the treat from my hand. It's not difficult at all for me to believe that service dogs do the same stuff.

This is exactly true! My dog will offer up a sit as soon as she thinks I have a treat in hand. Anyone heard of the "famous" counting horse Hans? Early last century. Could do math..even when his owner wasn't there. The trick was he was reading the body language of whomever was standing by him. They would tense up as he got closer to the correct answer and relax when he reached the right answer. Once they set him up where he couldn't see anyone, he couldn't count anymore. Horses and dogs primarily communicate through body language. Take away the handler and how useful are they?

 

[ljadayton]

Ive never liked the fact that they can bring out dogs and just throw out your rights based off of an animals response. I would bet a full paycheck on it, that somewhere in the country theres at least one department who teaches their dogs to hit on every car, or atleast has a signal to tell the dog which one they wanna search (probly one of those speed trap towns in GA). no way i will ever support ONLY a dog indicating on a car as reason to search.

It'd actually be pretty easy to teach a dog a signal like that. A touch in a certain spot, a low innocent sounding word, any number of ways. I have 2 seperate commands for my dog to leave something alone, depending on whether she has her mouth on it yet or not so why not a highly trained drug dog?

 

[E5RANGER375]

It'd actually be pretty easy to teach a dog a signal like that. A touch in a certain spot, a low innocent sounding word, any number of ways. I have 2 seperate commands for my dog to leave something alone, depending on whether she has her mouth on it yet or not so why not a highly trained drug dog?

ill pay you to teach my dog

 

[ljadayton]

ill pay you to teach my dog

What exactly am I teaching her ....

 

[jsx1043]

Having been a successful explosive detection canine handler, I can tell you that the science is there. The relationship between a detector dog and it's handler is completely different than that of a normal dog owner. Being a handler is a tough job and shibumi is right - it takes constant training of both the dog and handler to maintain proficiency. Emotion and frustration travels both ways up and down the leash and a handler can cue a dog, but usually those are worked out in training and is usually a rookie handler mistake. A lot of it has to do with the drive of the dog and how hard the dog wants to work.

My first dog, a 77 pound german shepherd, was a beautiful dog, but dumber than a box of rocks. He didn't have the drive to to be a successful dog. My second dog, a black lab/border collie mix, was CRAZY about working and did a great job. In fact, I couldn't cue her on anything (and I tried, to see if I was giving things away) becuase she cared about finding the aid more than pleasing me.

And really? I care a lick about something the Chicago Tribune has to say? Of course thy're going to post something negative about a good practice that all the liberal lawyers want to get rid of so they can get their douschebag clients off. The DoD stated in the article that they could not create technology as efficient and sensitive as the K-9 nose. They didn't say that the whole thing was a farce. And as for the the test they performed in KKKalifornia? It was rigged from the get-go. A real test NEVER tells any idicators to the handler for just that reason. A real search is not perfomed that way (there are no red papers to give it away) and the test was set-up for failure.

FAIL.

 

[Fletch]

Having been a successful explosive detection canine handler, I can tell you that the science is there. The relationship between a detector dog and it's handler is completely different than that of a normal dog owner.

Kindly note that the article explicitly defends bomb-sniffing dogs, particularly those deployed in war zones. RTFA.

And as for the the test they performed in KKKalifornia? It was rigged from the get-go. A real test NEVER tells any idicators to the handler for just that reason. A real search is not perfomed that way (there are no red papers to give it away) and the test was set-up for failure.

The results of the test were telling. There were NO actual scents in the test area. The dogs did not just signal on the marked areas, but on the unmarked areas as well. But they were twice as likely to signal on the marked ones, indicating handler bias.

The reason this is significant is because handler bias creeps into law enforcement. Example: LEO pulls over a vehicle packed with young males aged 18 - 22, all wearing distinctive clothing. He suspects drug gang, goes to the dog, and the dog signals, allowing a probable cause search. There's no red papers, but the LEO's bias affects the dog's judgment. That's the issue at stake here.

 

[Keyser Soze]

Ahhh I get it...So dogs are really good at smelling explosives, tracking humans......but they just cant smell drugs..........Makes perfect sense to me

 

[Fletch]

Ahhh I get it...So dogs are really good at smelling explosives, tracking humans......but they just cant smell drugs..........Makes perfect sense to me

Duck a little faster, and you could miss the point entirely next time.

This is NOT the message of the article, which you obviously didn't read.

 

[Compatriot G]

We wouldn't have to worry about whether or not the dog was hitting because of body language cues from the handler if we just eliminated the real problem. The War on Drugs is a complete failure. I realize drugs will never be legalized because there is too much money involved in the whole thing, though.

 

[E5RANGER375]

What exactly am I teaching her ....

sicke'm .... balls (not the throwing kind)