Predictability Changes Everything

WEBVTT
00:00:08.480 --> 00:00:12.960
Hello, hello, and welcome back to the Everyday Trainer Podcast.
00:00:13.119 --> 00:00:16.399
My name is Meg, and I am a dog trainer.
00:00:16.800 --> 00:00:24.480
We are picking back up where I left off, and we're breaking down another common study.
00:00:24.960 --> 00:00:29.920
This week we're talking about the predictability and e-collar training.
00:00:30.160 --> 00:00:31.199
You know the drill?
00:00:31.440 --> 00:00:34.799
Grab yourself a tasty drink and meet us back here.
00:00:41.359 --> 00:00:44.000
Hi guys, welcome back.
00:00:44.399 --> 00:00:49.520
I took a little bit of a breather from the podcast and just kind of the internet in general.
00:00:49.679 --> 00:00:51.840
Um, I had to go touch some grass.
00:00:52.079 --> 00:00:55.280
I went and lived in the forest for a couple weeks.
00:00:55.439 --> 00:00:56.240
It was amazing.
00:00:56.399 --> 00:00:58.640
10 out of 10, definitely recommend.
00:00:58.880 --> 00:01:05.840
And it also made me realize that it really is these phones that are kind of making us miserable.
00:01:06.079 --> 00:01:13.920
So here's your sign to maybe take a little break for um the internet and go frolic in the forest with your dogs, maybe.
00:01:14.079 --> 00:01:15.280
You know, you know.
00:01:15.519 --> 00:01:23.280
Um, so all that to say, I'm probably a little bit rusty, but we're here, we're doing it.
00:01:23.439 --> 00:01:30.799
And this week I'm breaking down another study that is commonly used in the industry.
00:01:31.040 --> 00:01:35.280
So last week we covered the science of motivation.
00:01:35.519 --> 00:01:39.359
This week we're covering the science of predictability.
00:01:39.599 --> 00:01:45.200
So if you're new here and you don't know, I am a dog trainer who utilizes tools.
00:01:45.359 --> 00:01:52.079
I use e-callers, I use prong callers, I'm mostly, I would say, reward-based.
00:01:52.239 --> 00:02:03.359
I would consider myself a true balanced trainer and that I'm using both rewards and also punishment to teach behaviors and also stop unwanted behaviors.
00:02:03.599 --> 00:02:16.319
And if you spent any time on the internet, you have probably realized that there is a pretty big debate amongst dog trainers and methodologies.
00:02:16.800 --> 00:02:24.319
And there's like a camp of trainers who will label them as I'm doing air quotes here, force-free.
00:02:24.400 --> 00:02:28.000
Then there's the balance people, then there's like the compulsion people.
00:02:28.159 --> 00:02:31.199
Not very many people are gonna like self-identify as that.
00:02:31.439 --> 00:02:51.840
Um, but there's a camp of trainers who don't believe in using tools, and they're often citing the quote science, saying that the science doesn't actually back up the use of punishment and tools in dog training and that it causes a lot of stress to our dogs.
00:02:52.080 --> 00:02:57.599
So the study that I'm gonna break down this week is the Shulk study.
00:02:57.919 --> 00:03:09.599
If you spent any time in the e-color debate, you've probably seen the Shulk study get cited, usually by somebody who is trying to prove that tools are inherently harmful.
00:03:09.919 --> 00:03:12.639
But here is what it actually found.
00:03:12.960 --> 00:03:21.120
Shulk and his team took 14 laboratory beagles and split them into three groups, each getting stimulation in a completely different way.
00:03:21.360 --> 00:03:28.800
One group got stimmed, and this is the e-collar stim, the instant they touched a decoy prey.
00:03:28.960 --> 00:03:33.840
So the consequence was immediate and directly tied to their own action.
00:03:34.159 --> 00:03:39.840
A second group got stimulated for blowing off a recall command, they already knew well.
00:03:40.560 --> 00:03:49.840
The third group got stimmed or shocked completely at random with zero connection to anything the dog was doing.
00:03:50.159 --> 00:03:54.639
And the results really tell the whole story.
00:03:54.879 --> 00:04:01.280
The random group, the ones who had no way to predict or control what was happening, showed the biggest spike in cortisol.
00:04:01.360 --> 00:04:09.840
And the group that got stimulated the instant they touched the prey, the most predictable and controllable setup, barely moved that needle.
00:04:10.000 --> 00:04:12.400
Their stress response was very minimal.
00:04:12.639 --> 00:04:19.279
This is the part that gets left out every time someone waves this study around as proof that e-collars are harmful.
00:04:19.519 --> 00:04:23.839
The study doesn't say that the tools cause stress.
00:04:24.160 --> 00:04:27.759
It says unpredictability causes stress.
00:04:27.920 --> 00:04:36.879
And this was actually brought up in that plus one study, the one that I brought up last week, and have made a couple posts on social media about.
00:04:37.279 --> 00:04:40.720
So it says unpredictability causes this stress.
00:04:40.800 --> 00:04:52.720
And a dog that clearly understands what triggers a consequence and has a real way to control or avoid it does not show meaningful stress markers.
00:04:52.959 --> 00:04:56.879
So that's not an argument against e-collars.
00:04:56.959 --> 00:05:03.839
That's an argument for exactly how good balanced trainers use the e-collar.
00:05:04.079 --> 00:05:18.560
We build predictability first, make sure the dog has full clarity on the cause and the effect, and the tool becomes communication instead of this unpredictable stress maker.
00:05:18.800 --> 00:05:26.800
So I did a little bit of deep dive into this study, and I wanted to see, first of all, when was this study conducted?
00:05:26.959 --> 00:05:33.040
It was actually conducted in 2001, but it wasn't published until 2007.
00:05:33.279 --> 00:05:37.199
They took about six months to conduct this experiment.
00:05:37.759 --> 00:05:41.839
And I had I had quite a few questions as a dog trainer.
00:05:42.000 --> 00:05:44.639
You'll have to excuse the dogs in the background.
00:05:44.720 --> 00:05:47.519
I'm at the facility and we're doing our night potties right now.
00:05:47.600 --> 00:05:53.519
So if you hear scratching or barks, that's what's going on in true dog trainer fashion, right?
00:05:54.079 --> 00:06:05.519
So I wanted to see what kind of e-callers were using in this study, how they were being used, all that stuff.
00:06:05.680 --> 00:06:11.920
So what I found is they used a Teletact Micro 3000.
00:06:12.079 --> 00:06:17.040
It had a transmitter, receiver, caller, and then a dummy caller.
00:06:17.120 --> 00:06:19.040
So that was like the control group.
00:06:19.279 --> 00:06:26.720
And the adjustable intensity level was zero to five, five being the strongest.
00:06:26.959 --> 00:06:41.360
And what really stood out to me when I was doing the research on all of this is every single trial in this study was run at a level five, which is the maximum intensity setting.
00:06:41.439 --> 00:06:46.240
And this is the highest possible pulse the device could deliver.
00:06:46.399 --> 00:06:49.680
The pulse was under one millisecond.
00:06:49.759 --> 00:06:50.959
They're calling it pulse.
00:06:51.120 --> 00:06:52.399
We call it STEM.
00:06:53.120 --> 00:07:06.319
And the researchers explicitly state they chose level five to test the dog's reaction under the highest electric pulse and as such, the worst condition possible.
00:07:06.639 --> 00:07:14.319
So that last point right there is the one that actually matters the most.
00:07:14.720 --> 00:07:21.600
This wasn't a study on typical responsible e-color use at a working level.
00:07:21.759 --> 00:07:41.040
It was a worst-case scenario stress test at the max intensity on dogs with no prior handler relationship, no proper conditioning process, and no, and this is for like the random group, no fair way to predict or control the outcome.
00:07:41.360 --> 00:07:55.279
So even under those deliberately harsh conditions, the group that could clearly associate the stimulus with their own action, which is that group A, showed the lowest stress response.
00:07:55.600 --> 00:08:05.759
Even at the most extreme intensity setting on dogs with zero conditioning, predictability was still the deciding factor in welfare outcomes.
00:08:06.000 --> 00:08:12.240
That's actually a very strong point for e-callers.
00:08:12.480 --> 00:08:18.399
So I wanted to look at how we're actually measuring the dog's stress.
00:08:18.800 --> 00:08:25.279
Because if you guys don't know, I do have a background in biology.
00:08:25.680 --> 00:08:32.240
I spent pretty much all of my college years reading scientific articles.
00:08:32.399 --> 00:08:35.440
So I do know how to kind of break all of these down.
00:08:35.600 --> 00:08:49.840
And I also know that whenever we're creating experiments, as the person who's structuring this, you kind of go in with some sort of prediction of how you think the data is going to show.
00:08:50.159 --> 00:08:57.039
And you create the experiment around that, either kind of proving or disproving your prediction.
00:08:57.200 --> 00:09:06.799
So I wanted to see how they're actually measuring all the little things in this because it's very easy to just say, you know, oh, these tools cause stress.
00:09:06.960 --> 00:09:09.919
And then you get into the details of like how they actually measured things.
00:09:10.000 --> 00:09:13.600
And it's like, um, that's kind of a very generalized statement.
00:09:13.919 --> 00:09:26.080
But unlike the PLOS study, which I covered last week, I really thought that this was a well-done study, to say the least.
00:09:26.320 --> 00:09:30.879
So I asked, How are we measuring the cortisol here?
00:09:31.039 --> 00:09:35.039
So they measured it through saliva, not blood.
00:09:35.120 --> 00:09:37.519
And this was their collection method.
00:09:37.759 --> 00:09:43.279
They put a small amount of citric acid in the dog's mouth to stimulate saliva production.
00:09:43.360 --> 00:09:48.320
Then they'd collect the sample from the dog's cheek using a cotton swab.
00:09:48.480 --> 00:09:55.759
The samples were immediately spun in a centrifuge at 4,000 RPM and cooled to, you know, 20 degrees Celsius.
00:09:55.919 --> 00:09:57.039
None of that really matters.
00:09:57.200 --> 00:10:07.919
And the lab staff doing the measurements were blinded, which means they didn't know which group was which whenever they were measuring everything, which is solid.
00:10:08.080 --> 00:10:08.720
Good job.
00:10:08.960 --> 00:10:10.080
So timing.
00:10:10.240 --> 00:10:24.159
The samples were taken at set intervals, five samples per test, spaced five minutes apart, with the first sample taken 10 minutes after the stem from the e-collar was applied.
00:10:24.399 --> 00:10:29.200
And they did this timing because cortisol doesn't spike instantly.
00:10:29.360 --> 00:10:32.879
It takes several minutes to show up in saliva after a stressor.
00:10:33.039 --> 00:10:38.799
So waiting 10 minutes was really the right call to actually catch the response.
00:10:39.039 --> 00:10:58.399
And they measured saliva instead of blood because they noted that saliva cortisol correlates highly with plasma cortisol, just at a smaller concentration, roughly 4 to 11% of blood levels, depending on which prior study you reference.
00:10:58.720 --> 00:11:14.320
Saliva sampling is far less invasive and lets you repeat measurements over a short period without the stress of repeated blood draws, which would have contaminated the stress readings that they were trying to measure.
00:11:14.559 --> 00:11:20.399
And while they were doing this study, they did also measure heart rate simultaneously.
00:11:20.720 --> 00:11:26.639
So I was like, okay, well, what were the heart rate results?
00:11:26.960 --> 00:11:35.759
And the heart rate told a very similar story to cortisol, but the picture was kind of messy and not as clean cut.
00:11:36.080 --> 00:11:47.279
The overall pattern was on the days that the STE was applied, the random group showed an increase in their peak heart rate.
00:11:47.600 --> 00:11:59.840
Group A, which is the aversion group, the one that they stemmed on the prey, and group H, which is the here group, showed the opposite.
00:12:00.080 --> 00:12:02.960
Their peak heart rate actually decreased.
00:12:03.200 --> 00:12:23.200
And that is kind of interesting and a very meaningful distinction because earlier research found that animals given a real chance to predict or avoid a stressor tend to show a drop in heart rate, while animals with no way to see if it's coming show a spike.
00:12:23.360 --> 00:12:26.399
That tracks exactly with this study here.
00:12:26.559 --> 00:12:32.960
So group R had no way to associate their own behavior or a known signal with the STEM.
00:12:33.440 --> 00:12:39.679
Their bodies just stayed in this heightened, reactive, stressed state.
00:12:40.320 --> 00:12:47.120
So we'll kind of break down like the statistical significance across like the things that they measured.
00:12:47.279 --> 00:12:54.320
And this is where it gets a little nuanced across most of the specific data comparisons.
00:12:54.480 --> 00:12:59.039
There weren't significant differences in the standard heart rate measures.
00:12:59.279 --> 00:13:19.440
There were significant differences that showed up in a couple of specific data sets, one comparing peak heart rate values, and another comparing the recovery heart rate and peak to recovery ratio, specifically between groups A and groups H and R.
00:13:19.759 --> 00:13:29.120
So all that is really saying that there's no real significant differences in heart rate between any of the groups.
00:13:29.360 --> 00:13:45.519
In this study, the researchers flagged the heart rate as kind of like, eh, this is kind of messy and is not really giving us a good thing to go off of because the dogs were also like physically moving, they're chasing prey.
00:13:45.759 --> 00:13:52.720
So a heart rate spike could reflect the physical exertion just as easily as stress.
00:13:53.039 --> 00:14:02.399
And that overlap between motion and emotion makes heart rate a much less clean signal than cortisol in this particular study.
00:14:02.639 --> 00:14:12.639
So that gives you the breakdown of kind of how you know study was done and why we studied cortisol specifically in this paper.
00:14:12.799 --> 00:14:18.000
I also wanted to understand who was conducting the study.
00:14:18.159 --> 00:14:20.960
And I was pleasantly surprised with this one.
00:14:21.120 --> 00:14:23.360
The study was conducted by E.
00:14:23.679 --> 00:14:24.879
Schulk, J.
00:14:25.519 --> 00:14:33.679
Stichnoff, and a few other people, published in Applied Animal Behavior Science in 2007.
00:14:34.320 --> 00:14:42.879
So these people were affiliated with the Department of Animal Welfare and Behavior at the Veterinary College of Hanover, Germany.
00:14:43.039 --> 00:15:02.080
The research itself was carried out at a laboratory that I cannot pronounce because it is German, which is an animal facility affiliated with the University of Heidelberg, which supplied the 14 beagles from the breeding stock.
00:15:02.240 --> 00:15:16.320
Also, which is nice to note, the study was approved by the animal welfare officer of the Veterinary School of Hanover and by the German governmental authorities under the German Animal Welfare Act.
00:15:16.480 --> 00:15:22.080
The funding for the study came from Hans and Helga Maus Foundation.
00:15:22.240 --> 00:15:24.399
So I was like, well, who the heck is that?
00:15:24.559 --> 00:15:29.840
Because it's very important to understand who is funding these studies.
00:15:30.000 --> 00:15:47.679
So this charitable foundation specifically funds research and activities in animal welfare in the state of Baden-Wurttemberg at the veterinary school in Hanover, which is the same institution where Schulk and her colleagues were based.
00:15:47.840 --> 00:15:50.960
Okay, so let's get into the results of the study.
00:15:51.200 --> 00:15:57.519
Group A, which is the aversion, this is the one that the dogs are stemmed for touching the prey.
00:15:57.759 --> 00:16:08.960
Shock tied precisely to touching the prey, fully predictable and controllable, the smallest cortisol increase, 22 to 31% above baseline.
00:16:09.120 --> 00:16:16.559
This was even lower than the increase seen just from being leash restrained during the preliminary tests.
00:16:16.799 --> 00:16:17.120
Okay.
00:16:17.679 --> 00:16:27.600
Group H, which is the here group, so dogs that had a here command, shock for ignoring a known recall command.
00:16:27.759 --> 00:16:40.159
This was a moderate increase up to about 114% in absolute cortisol and up to 100% 160% in relative cortisol.
00:16:40.320 --> 00:16:45.759
And then the group R, which is our random group, the dogs that were just stemmed randomly.
00:16:46.159 --> 00:16:52.159
The shock was delivered arbitrarily with no way to predict or control it.
00:16:52.480 --> 00:17:12.319
This one by far had the highest increase in cortisol, up to 336% in absolute cortisol and 328% in relative cortisol, higher than both groups across nearly every data set.
00:17:12.480 --> 00:17:16.960
Group R had the highest cortisol levels of the three groups.
00:17:17.039 --> 00:17:21.839
And four weeks later, after this study was done, they did a post-test.
00:17:22.240 --> 00:17:34.559
Group R still had the highest cortisol level of the three groups, and group H, uh, their cortisol still hadn't returned to pre-experiment baseline.
00:17:34.880 --> 00:17:40.319
Group A's levels were the lowest consistent with the main test.
00:17:40.960 --> 00:17:52.799
So their conclusion was dogs who could clearly associate the shock with their own actions and therefore predict and control it did not show considerable or lasting stress.
00:17:53.039 --> 00:18:00.799
Dogs who couldn't do that, group H and especially R, showed real sustained stress responses.
00:18:01.039 --> 00:18:22.079
The paper's own stated takeaway was that poor timing and unpredictable application of high-level shocks carries a real risk of severe, persistent stress, and that use of these devices should require proof of proper training and be restricted to specific appropriate situations.
00:18:22.319 --> 00:18:38.720
And the last line right there, that is worth keeping at kind of the front of our minds since it's literally the author's own recommendation and it supports my argument of tool usage.
00:18:38.960 --> 00:18:46.079
We need to know what we're doing with the tool before we just slap it on our dog and start using it.
00:18:46.240 --> 00:18:46.559
All right.
00:18:46.640 --> 00:18:50.480
This brings me into learned helplessness and that connection.
00:18:50.720 --> 00:18:57.519
Learned helplessness is what happens when an animal learns that their behavior has no reliable effect on their environment.
00:18:57.680 --> 00:19:01.440
They just stop trying and they completely shut down.
00:19:01.680 --> 00:19:05.119
This is what the random group and the shulk experienced.
00:19:05.359 --> 00:19:09.680
It is also what poorly applied punishment produces in dogs.
00:19:09.839 --> 00:19:18.000
A dog who has learned that nothing they do changes what happens, they stop engaging with whatever we're doing.
00:19:18.160 --> 00:19:33.359
So when we have bad timing with our punishment, when our dog isn't able to predict why they're being punished, that is what's causing lasting negative effects on dogs and causing these spikes in cortisol.
00:19:33.519 --> 00:19:37.279
Now I've been a dog trainer for the past eight years of my life.
00:19:37.519 --> 00:19:45.839
And in the beginning, I wasn't really using tools that much because I didn't really understand the timing of them.
00:19:46.000 --> 00:19:47.519
I didn't want to mess up dogs.
00:19:47.680 --> 00:19:55.599
I remember I was learning from a trainer friend of mine, and she had a client dog, and I wanted to shadow her to see how she was using the e-collar.
00:19:55.680 --> 00:19:56.799
I was very new to it.
00:19:57.039 --> 00:20:00.400
I was just experimenting on Lucy, my own dog.
00:20:00.960 --> 00:20:04.880
But I was very, I didn't even want to like touch the remote.
00:20:04.960 --> 00:20:05.920
I was like, oh my gosh.
00:20:06.079 --> 00:20:10.640
I shadowed her and she handed me the dog and she was like, here's how we use the e-collar.
00:20:10.720 --> 00:20:13.519
And I was like, nope, I'm gonna hand it right back over to you.
00:20:13.759 --> 00:20:18.000
Because I was petrified of messing up a dog.
00:20:18.240 --> 00:20:43.839
Now that I understand these tools and have spent the last, you know, almost decade of my life training dogs, it's so obvious for me, which I don't think this is really represented in science, but it's so obvious for me after the thousands of dogs that I've trained, how helpful these tools are in dogs' well-being.
00:20:44.240 --> 00:20:49.440
Because I don't really get the dogs that come in with bad e-collar training.
00:20:49.680 --> 00:21:04.240
We do every now and then, they're kind of sprinkled in there, and you know, you'll put an e-collar on the dog and you just see them kind of shut down, and you're like, oh, you have had a bark collar or an invisible fence before, or you have been to another board and train program.
00:21:04.720 --> 00:21:17.920
I don't really get that many of those dogs, but what I do get a lot of is people who are petrified of using tools because they have been villainized online.
00:21:18.400 --> 00:21:32.160
They're watching all of these force-free trainers content, telling them that they're gonna ruin their relationship with their dog, that they don't need to use these tools, that they can solve all their problems without these tools.
00:21:32.319 --> 00:21:45.680
And so a lot of the clients that we're getting are coming in with these super anxious, fearful, reactive or aggressive dogs that are honestly running these people's lives.
00:21:46.000 --> 00:21:54.400
Or we get the people who are coming in and saying, I don't know what to do if this doesn't get fixed.
00:21:54.720 --> 00:21:56.240
I have to give the dog up.
00:21:56.480 --> 00:21:59.759
I see that a lot more than bad e com.
00:22:00.880 --> 00:22:05.599
And this isn't to say that bad e-color training isn't out there.
00:22:05.759 --> 00:22:10.799
I'm sure that there are plenty of people that are using their e-colors incorrectly.
00:22:10.880 --> 00:22:21.680
I'm sure there are plenty of dog trainers that are using their e-colors incorrectly and are just like blasting the dogs and the dogs don't know why they're getting corrected or they're starting off at super high levels.
00:22:21.920 --> 00:22:27.039
But for a majority of balanced trainers, like we are just not using e-colors in this way.
00:22:27.200 --> 00:22:34.880
And if you've listened to my previous e-color podcast, I break down the stages of e-color training.
00:22:35.039 --> 00:22:42.480
First and foremost, we teach everything on leash and teach, you know, new behaviors using food.
00:22:42.559 --> 00:22:53.119
If a dog is food motivated or play, we build up the behaviors first, teach the dog what the leash means, teach the dog what sit means, what place means, all of this stuff first.
00:22:53.279 --> 00:23:00.160
And then we layer the e-color on top of that at the lowest level that the dog feels the sensation.
00:23:00.400 --> 00:23:03.759
We teach the dog how to turn that stem off.
00:23:04.000 --> 00:23:09.920
We're basically treating it just like how we teach the leash, how we teach the dog to give into leash pressure.
00:23:10.160 --> 00:23:10.960
Same thing.
00:23:11.119 --> 00:23:18.720
And it's not until the dog understands the e-color and how to turn it off that we use it for punishment.
00:23:18.960 --> 00:23:40.319
All of these studies that are citing e-color use and punishment and training, and I'm saying all of these studies, I'm sure that there are some that are not using e-colors at a high level, but these studies, the ones that I'm reading, they are not using e-colors in the way that good balanced trainers are using e-collars.
00:23:40.480 --> 00:23:59.359
And so it's really skewing the data and skewing people's perception of these tools because we're using these tools in as worst-case scenario in studies when as a dog trainer, I'm not blasting dogs on the highest level pretty much ever.
00:23:59.599 --> 00:24:08.319
There might be a dog here and there that we have to correct, you know, one time at a high level, but the dog knows where the correction is coming from.
00:24:08.480 --> 00:24:12.000
They know what behavior is eliciting this correction.
00:24:12.400 --> 00:24:26.480
The research around these tools is not a good representation of good e-color training and also the benefit and welfare of animals.
00:24:26.640 --> 00:24:41.119
Like there has to be some sociological studies done that's like, okay, yes, we see a spike in cortisol, but also there are plenty of other moments in a dog's life where cortisol spikes.
00:24:41.359 --> 00:24:51.440
You know, I'm sure if we studied reactive dogs and took samples of their saliva after being reactive, their cortisol is spiked.
00:24:51.599 --> 00:25:14.240
And so who's to say that a, you know, correction from an e-collar is that much worse than allowing a dog to exist in a state of anxiety and you know, spiked cortisol all day, every day, because we haven't trained them, because we have no way of telling them, hey, you don't have to do this.
00:25:14.400 --> 00:25:15.759
You don't have to be this way.
00:25:15.920 --> 00:25:17.440
Like there is another way.
00:25:17.680 --> 00:25:29.359
Just communicating with them in that moment and disrupting those behaviors, in my mind, that is so much more aversive than a lot of these tools.
00:25:29.519 --> 00:25:43.440
That is so much, it's so aversive to let a dog exist in that state and just stand there and watch them and wait for them to hopefully bring themselves back down so that we can reward them.
00:25:43.759 --> 00:25:47.920
I think that is that is what is aversive.
00:25:48.000 --> 00:25:54.640
And that is really what I would love to see studied is okay, let's study the other side of things.
00:25:54.799 --> 00:25:57.279
Let's study the force-free side of things.
00:25:57.519 --> 00:26:03.519
Let's measure cortisol spikes in dogs after they're, you know, reacting.
00:26:03.680 --> 00:26:11.200
And then let's, you know, take them to a good balance trainer and then let's measure their stress levels when they're no longer reacting.
00:26:11.359 --> 00:26:14.000
Like, let's do these types of studies.
00:26:14.079 --> 00:26:21.119
Let's study good e-collar training, good balance training instead of measuring worst case scenario.
00:26:21.440 --> 00:26:39.519
And, you know, last week's, I'll say last week, even though it's been like a month, but last week's episode, the PLOS one study, that is like stringing dogs up, kicking dogs, yelling at dogs at a high rate per training session.
00:26:39.599 --> 00:26:44.000
Almost every training session, these trainers are doing that with these dogs.
00:26:44.240 --> 00:26:50.799
And we're saying that that is getting lumped in with good balance training.
00:26:51.279 --> 00:26:52.799
It's just not the case.
00:26:52.960 --> 00:27:24.480
And anybody who uses these tools understands that you see the dogs that come into you for training, you see the owners that are struggling with their dogs because they don't want to tell them no because they're scared that they're gonna ruin their relationship, they're scared to use these tools, they're scared to put structure, they've been told that you can do this without aversives, you can do this, and I'm sure you can.
00:27:24.720 --> 00:27:33.839
I'm sure you can, but why allow a dog to suffer in that state for that long?
00:27:34.160 --> 00:27:38.559
So that's that's the stuff that I want to see.
00:27:38.799 --> 00:27:43.599
And there's actually a paper that kind of breaks this down a little bit.
00:27:43.920 --> 00:28:00.799
A 2026 paper in Frontiers in Veterinary Science documented that the professional dog training landscape has become politicized with training method choices tied to moral identity and ideological affiliation rather than evidence.
00:28:01.039 --> 00:28:09.039
And this paper flagged real methodological problems across the existing literature.
00:28:09.359 --> 00:28:24.480
Correlation, study designs that cannot establish cause and effect, convenience samples from schools that are already committed to one method, and expectancy bias in observational coding.
00:28:24.559 --> 00:28:33.920
So, what that means is that the people who are observing these studies already have a bias on this stuff because that is who is doing this research.
00:28:34.079 --> 00:28:45.519
And is also why I really liked this Schulk study because you could tell that it was not going in with this initial bias.
00:28:45.599 --> 00:28:52.720
Like they really just wanted to study the effects of these tools and the predictability factor into this.
00:28:53.920 --> 00:29:00.319
So here's what I think the research would need to show to be fair.
00:29:00.720 --> 00:29:20.160
A study comparing reward-only training versus balance training as actually practiced, which means full behavior foundation first, then e-color introduced unknown behaviors at the lowest level possible with a clear off signal.
00:29:20.720 --> 00:29:28.559
And until that study exists, we are comparing apples to very different kinds of apples.
00:29:30.000 --> 00:29:37.680
So the research is limited and is often not studying what balance trainers actually do.
00:29:38.960 --> 00:29:41.680
Let's talk about what we actually do.
00:29:43.759 --> 00:29:55.519
The e-caller is never a teaching tool in skilled balance training, it is simply a communication tool that gets layered over behaviors that the dogs already know.
00:29:55.920 --> 00:29:58.880
And we can use these tools as punishment.
00:29:59.119 --> 00:30:04.960
There are plenty of studies out there that show that punishment works in stopping behaviors.
00:30:05.200 --> 00:30:07.200
We can use the e-collars as punishment.
00:30:07.279 --> 00:30:10.880
I'm not here to say that we never dial up over a certain level.
00:30:11.039 --> 00:30:16.160
I think that is very misleading as well.
00:30:16.480 --> 00:30:21.519
But if the dog doesn't know the behavior, the e-collar does not come out.
00:30:21.920 --> 00:30:29.359
You cannot layer a consequence, you cannot layer the stim over something that the dog doesn't know yet.
00:30:30.240 --> 00:30:36.640
So we build the behavior completely with positive reinforcement, with our leash.
00:30:37.920 --> 00:30:44.000
The dog sits, the dog comes, the dog goes to place, they walk on a loose leash.
00:30:44.160 --> 00:30:51.839
They have to do these things first many, many times, and then we can layer the tool over.
00:30:52.079 --> 00:31:03.359
We have to proof the behavior across different environments and distractions, and the behavior needs to exist under real world conditions.
00:31:04.559 --> 00:31:10.400
First, we find the dog's working level on the e-collar, and this is the lowest level the dog notices.
00:31:10.559 --> 00:31:14.880
This is not the level that the dog reacts to with discomfort.
00:31:15.119 --> 00:31:16.880
This is just the dog noticing it.
00:31:17.119 --> 00:31:18.559
And then we can apply the stim.
00:31:18.880 --> 00:31:20.319
We give the known cue.
00:31:20.400 --> 00:31:23.519
The moment the dog performs the behavior, the stim stops.
00:31:23.759 --> 00:31:26.640
So the dog learns, I can turn this off.
00:31:26.799 --> 00:31:27.599
I know how.
00:31:27.759 --> 00:31:29.039
I do the thing.
00:31:29.279 --> 00:31:34.319
This is how we avoid that dog shutting down, going, oh my gosh, I don't know how to turn this off.
00:31:34.400 --> 00:31:36.480
So I'm just gonna shut down because I don't know what to do.
00:31:36.640 --> 00:31:38.319
That learned helplessness.
00:31:38.559 --> 00:31:46.720
And we're gonna repeat this until the dog is offering the behavior to turn off the stim before you ever need to cue it.
00:31:46.960 --> 00:31:52.240
So now they're working on the e-collars themselves.
00:31:52.559 --> 00:31:55.680
So why does this remove the stress?
00:31:55.920 --> 00:31:58.000
It goes back to our schulk study.
00:31:58.079 --> 00:32:04.960
The dogs who were not stressed were the ones who could predict the stim and control the outcome with their behavior.
00:32:05.200 --> 00:32:10.400
That is exactly what this type of e-collar training creates.
00:32:10.640 --> 00:32:17.279
The dog has full predictability, full control, and a clear known behavior to execute.
00:32:17.680 --> 00:32:22.480
The collar does not become a source of anxiety because the dog is never confused.
00:32:22.640 --> 00:32:29.359
They know what it means and they know what to do and they know how to turn that stim off.
00:32:30.319 --> 00:32:31.839
I can't speak.
00:32:32.720 --> 00:32:43.839
So a dog in a balanced training session with a properly introduced e-collar is not a dog who is tense or flat or avoidant.
00:32:44.079 --> 00:32:52.720
These are dogs who are switched on, they are paying attention, they are moving confidently because they understand the rules of the game.
00:32:52.880 --> 00:32:54.799
There is no ambiguity.
00:32:55.119 --> 00:33:01.839
Contrast that with the dogs in the research studies who do not know what the stim means or how to make it stop.
00:33:02.000 --> 00:33:13.200
Of course, those dogs are stressed, of course, their cortisol rises, the methodology creates uncertainty, and the dog's nervous system is responding accordingly to that.
00:33:13.440 --> 00:33:18.240
Now, here is the part that I find most interesting.
00:33:18.640 --> 00:33:26.160
What this methodology reveals about the gap between what research studies and what experienced trainers know.
00:33:26.480 --> 00:33:30.319
So, this is what the science has not figured out yet.
00:33:30.559 --> 00:33:35.119
The entire framework described above has not been studied.
00:33:35.599 --> 00:33:49.839
The research on e-collars does not distinguish between this approach and using the collar as a primary teaching mechanism or just slapping the tool on and correcting the dog when they don't know how to turn that off.
00:33:50.160 --> 00:34:06.640
And this matters because if you ran a study comparing the random group from shulk against a dog trained with this balanced approach, you would likely see dramatically different welfare outcomes.
00:34:06.799 --> 00:34:13.760
Not because one group used a color and one didn't, but because one group had that predictability and one did not.
00:34:14.079 --> 00:34:20.639
One of the most important things experienced balanced trainers understand is that the dog needs a clear way to turn the stem off.
00:34:20.800 --> 00:34:23.280
This is not just a training preference.
00:34:23.440 --> 00:34:25.840
It's honestly animal welfare.
00:34:26.000 --> 00:34:33.039
Now there is a camp of trainers that don't believe in conditioning the e-color quite like I do.
00:34:33.360 --> 00:34:39.280
Um, they don't do the low-level e-collar stim, layering over, teaching the dog how to turn it off.
00:34:39.519 --> 00:34:51.920
I believe, and I don't really know because I have not learned from these trainers, but I believe that they still are teaching behaviors first and then holding the dogs accountable to whatever it is.
00:34:52.079 --> 00:34:59.199
I don't know that they do it with obedience or if it's just like behavioral situations.
00:34:59.360 --> 00:35:03.679
So for example, we layer e-color over place.
00:35:04.000 --> 00:35:20.639
I know that the other camp of e-collar trainers don't do stuff like that because their belief is that we're just teaching the dog to tolerate higher levels on the e-collar, and they would rather the e-collar just be a correction itself.
00:35:21.360 --> 00:35:25.760
So I don't necessarily know that they do it with obedience.
00:35:25.920 --> 00:35:30.480
I believe that they do, but it's only after the dog understands like what the behavior is.
00:35:30.639 --> 00:35:34.559
So the e-color in that sense is also still predictable.
00:35:34.719 --> 00:35:40.719
So it's still not what a lot of these research articles are showing e-color training is.
00:35:40.880 --> 00:35:51.199
Even if you don't do that conditioning stage, the dogs are still taught behaviors and know why a punishment is happening or why a punishment event is happening.
00:35:51.360 --> 00:35:51.920
I would hope.
00:35:52.000 --> 00:35:52.719
I think so.
00:35:52.960 --> 00:35:53.679
I don't really know.
00:35:53.840 --> 00:36:04.960
I'm gonna need somebody else to reach out to me and uh let me know if you are a part of this camp of trainers or have learned from them or know what they're doing.
00:36:05.519 --> 00:36:34.159
But, anyways, I guess my takeaway and my opinion on all of this is I hadn't up until this point, I hadn't really looked into a lot of the science because the science in dog training, I just always knew it was a little biased because the people who are preaching the science are the people on the other side of the fence from me.
00:36:34.320 --> 00:36:36.880
And I'm not anti-force-free.
00:36:37.039 --> 00:36:38.719
Like, do your thing, you know.
00:36:38.960 --> 00:36:57.280
If you don't have any issues and with how I train dogs, like I don't have any issues with how you train dogs, and I'm never gonna be the person that has to like bash somebody else's methodology to make myself feel better.
00:36:57.599 --> 00:37:00.960
All I can do is just be better, right?
00:37:01.119 --> 00:37:16.719
And show the proof in the dogs that I train and show how happy they are and the lives that they get to live, which also kind of brings up a point of what does your world, what does your dog's world look like after the e-caller?
00:37:16.880 --> 00:37:18.159
Is their world bigger?
00:37:18.320 --> 00:37:26.079
You know, and I've had quite a few clients of mine who came to us who are like, we're not gonna use tools, like this is how we want to do things.
00:37:26.239 --> 00:37:27.679
And we work with them.
00:37:27.920 --> 00:37:34.639
We totally work with them, but I see their dogs and they continue to struggle.
00:37:35.039 --> 00:37:39.119
And, you know, they're bringing their dog to class and their dog is being super reactive.
00:37:39.280 --> 00:37:56.000
They don't have any control of over their dog, they sometimes have to leave class because there's not enough, you know, distance because their dog is so reactive, they can't talk with us after class in the parking lot because their dog is being so reactive to other dogs.
00:37:56.239 --> 00:38:06.960
Meanwhile, the clients who did adapt to using these tools, their dogs are progressing so much faster.
00:38:07.280 --> 00:38:09.920
And the dogs are chill now, right?
00:38:10.000 --> 00:38:19.440
Like they're not popping off at other dogs, they're not super stressed, they're not spinning during the training session and being just like completely unhinged.
00:38:19.599 --> 00:38:21.519
So their world opens up.
00:38:21.679 --> 00:38:31.519
They can hang out with us outside in the parking lot, they can take their dogs to a coffee shop, they can do more group classes with us, like they can do so much more.
00:38:32.320 --> 00:38:35.760
So for me, it's just a no-brainer.
00:38:36.079 --> 00:38:43.280
And I think if you spend enough time in this space learning from people who are doing good balanced training, it's a no-brainer.
00:38:43.440 --> 00:38:47.760
It's like these tools open up your dog's world.
00:38:48.239 --> 00:38:56.960
And I would rather correct my dog and let's say I'm using the e-collar as a punishment to stop a certain behavior.
00:38:57.280 --> 00:39:14.400
I would rather my dog have a moment of discomfort, a moment of stress, so that they can have a lifetime of more freedom, more adventures with me, more coffee shop visits, more park visits.
00:39:14.639 --> 00:39:18.079
I get to bring them to my friend's house because they're well behaved.
00:39:18.239 --> 00:39:22.960
Like trained dogs get to go everywhere.
00:39:23.280 --> 00:39:35.920
And I would hate for you guys as dog owners to limit your dog's world because of the fear-mongering that happens around these tools.
00:39:36.079 --> 00:39:51.199
You know, I do think that it's so important to actually break down the science and look at what people are referencing when they're saying e-collars cause stress and e-collars, you know, do all of this damage.
00:39:51.360 --> 00:39:52.639
Well, let's look into it.
00:39:52.800 --> 00:39:57.760
Let's look into what is actually being measured here.
00:39:58.000 --> 00:40:01.920
I hope that this has opened your eyes a little bit.
00:40:02.239 --> 00:40:04.880
Um, there's there's tons of science out there.
00:40:05.039 --> 00:40:13.440
I'm gonna keep breaking it down, but those are really the two main studies that are focused on e-callers right now.
00:40:13.679 --> 00:40:21.599
Um, if you are listening to this episode and you know of any other studies that you would like to send me, send them my way.
00:40:21.679 --> 00:40:22.719
I'll read them over.
00:40:22.880 --> 00:40:27.039
We'll break them down in this podcast and go over kind of what I think.
00:40:27.280 --> 00:40:30.719
The takeaway from this is do your own research.
00:40:30.880 --> 00:40:32.639
Don't take my word from it.
00:40:32.880 --> 00:40:42.079
I will put the sources, I'll put the sources to the articles at the end of the podcast.
00:40:42.159 --> 00:40:46.239
I think they're also cited in the last episode that we did.
00:40:46.320 --> 00:40:49.599
So if you guys want to read them for yourself, you totally can.
00:40:49.840 --> 00:40:52.639
A few little housekeeping things, virtual shadow program.
00:40:52.960 --> 00:40:59.360
If you guys are interested in becoming a dog trainer or you are a dog trainer and just want to fine-tune your skills a little bit.
00:40:59.599 --> 00:41:02.559
That is starting on Monday.
00:41:02.800 --> 00:41:04.159
Very exciting.
00:41:04.480 --> 00:41:07.760
And I'm going back into the wilderness.
00:41:07.840 --> 00:41:09.440
We're doing some adventure board and train.
00:41:09.599 --> 00:41:19.840
So if you guys are interested, if you're on the west coast and want to do a board and train with me, I have some spots open for the end of this year and the beginning of next.
00:41:20.000 --> 00:41:21.199
Thanks for being here.
00:41:21.519 --> 00:41:22.320
Miss y'all.
00:41:22.480 --> 00:41:23.760
We'll see you next week.