What happens when you ask your dog to do something they don’t care for? We are not all perfect trainers, plus sometimes we are forced to compromise. Let’s say you’ve worked on teaching your dog to get her nails trimmed and teeth brushed but suddenly she has an ear infection and needs ear drops. You haven’t gotten to ear handling yet.
So you get out massively high-value treats and start following your dog’s ear drops with ham, Romano cheese, or whatever your dog finds amazing. But you didn’t have time to condition this gradually, starting with approach, then ear handling, and working up to ear drops. So you are doing something weird and unfamiliar, plus your dog’s ears are sore.
But you have worked a lot with your dog, and she doesn’t run from the room when you get the drops. She lets you do it, and you do it with the least restraint and most kindness possible.
What might you notice over time about her behavior as you prepare for this procedure when you compare it to other things you do with her?
She is probably not going to be eager to get into position. She will likely have high response latency to your cue.
Response latency, or reaction time, is:
The time that elapses between the onset or presentation of a stimulus and the occurrence of a specific response to that stimulus.Online APA Dictionary of Psychology
Dog trainers usually refer to response latency as just plain latency, but there are other types of latency, both in psychology and applied behavior analysis, so I’m making the distinction.
Note that response latency does not include the time it takes the animal to perform the behavior. But doing the behavior slowly is another good thing to pay attention to.
Generally speaking, we want response latency to be low, zero even. We hope to see the dog start to respond immediately after our cue. And not only that, to quote my friend and wonderful trainer Lori Stevens, we want to see the joy. We don’t just want fast; we want happiness and enthusiasm. And with positive reinforcement training, fast and happy often go hand in hand.
It’s possible to get high latency for reasons other than a reluctant or worried animal. A common reason is that we haven’t worked hard enough on cue recognition or proofing. Or perhaps the behavior is challenging, and it takes the dog a bit of time to get ready. Or there may be a competing reinforcer. Bob Bailey wrote a great article about latency years ago for the Clicker Solutions Yahoo group. You should check it out if you are having latency problems because of training mishaps, or are interested in this little-discussed training issue. Just keep in mind that his latency protocol is not designed to address an untoward emotional response.
When a dog is uncomfortable or afraid of something, high latency often roughly means, “I know what you want but I don’t wanna.” This is tricky territory because it’s risky to say the dog “knows what you want.” It’s such a common mistake for newbie trainers or force trainers to believe that and then assume any lack of response is because of stubbornness or other “character flaws” in the dog. If we are using reinforcement well and the dog doesn’t respond or responds slowly, it’s generally because we haven’t been clear. We haven’t trained the dog to fluency (which is a better description than “knows what you want” anyway). But I’ll go out on a limb for my particular examples and say that I think Zani knew exactly what the circumstances predicted. She just didn’t have good feelings about it.
Eva Bertilsson and Emilie Johnson Vegh
I credit the lovely Swedish trainers Eva and Emilie for starting a conversation about latency that a lot of us needed to hear. We talk so much about body language and learning what our dogs’ behavior can tell us. We study stress signals and observe our individual dogs and their palettes of behaviors. But as far as I know, response latency as an emotional indicator was rarely discussed before Eva and Emilie started speaking about it. (Correct me on this if I’ve missed earlier discussions of it.)
They have a whole protocol they developed after noticing how important latency was, first in an agility setting. I’m not qualified to discuss that, and it’s not what I’m talking about here. My focus is more basic. I want to show you this one piece. I want to show you the difference between a low latency and a high latency version of the same behavior.
Mounting the Klimb for Husbandry
I have one of those cool platforms called a Klimb next to my bed. I got it so Zani would have an easier time getting on my bed, which is high. But it’s such a convenient thing that we use it for general training, physio exercises, and husbandry. Lots of fun stuff happens on the Klimb. And one not-so-fun thing: face holding for teeth brushing.
I brush Zani’s teeth every night. She is going on 12 and gets plaque easily. And now with the pandemic, I want to avoid unnecessary vet visits more than ever. So, although I’m not always great at being consistent about things like this, I have forced myself to be good about brushing her teeth every night. (And it’s helped immensely, by the way.)
I did a decent job with our beginning toothbrush practice. I found a video I liked, made myself a task list, nicely split out, and practiced with Zani. I didn’t work hard to develop a positive response to the toothbrush itself since she wasn’t bothered by it. I just got her used to different manipulations of her head and mouth and gradually introduced the brushing. It went fine.
But in the meantime, I had one of those husbandry mishaps going on where I had to go ahead and do something even though she didn’t like it. I won’t go into the whole unfortunate eyedrops story here. It, too, started out fine, but took a bad turn.
And after that, she started some mild avoidance maneuvers when I brushed her teeth. Not because of the actual brushing, but because I have to approach her head with my hands in a way similar to eyedrop application.
So the outcome is this.
- She is happy to get her toenails trimmed on the Klimb.
- She is not happy about getting her head positioned to get her teeth brushed on the Klimb, but she will cooperate.
And I can show you her attitude to both activities without showing the activities themselves. I can show you her response latency and her behavioral speed when mounting the Klimb.
One more bit of background! In both clips, you’ll hear me give a verbal cue. Now Zani is a clever little dog, but she isn’t great at verbal cues. That is not surprising, given her genetics, which include hound and terrier, breeds that were historically meant to work independently. But she reads situations superbly. When we are going to work on toenails, it’s usually during the day. I take her into the room by herself, pick up my headlamp and the trimming tools, and invite her onto the Klimb. But I always brush her teeth late at night, before going to bed. Clara is in the room with us. I do my going to bed routine, then get the toothbrushing gear.
These two situations are super easy for her to discriminate. They are screamingly different to her observant little self.
I go into these slightly personal details to clarify that the words I say to her in the two different situations are not important. She’s reading the situation loud and clear. Her responses to my invitations to get on the Klimb are very different because she knows before I cue her up there what we are going to do.
Latency Rundown from the Movie
The first time I invite Zani to get on the Klimb for a nail trim, she is caught off guard, but then so enthusiastic that she jumps all the way onto my bed instead. And the time between my verbal invitation and her response—the latency—is 2.0 seconds. The second time, she actually anticipates my cue, but then waits. Then I give the verbal cue, and her latency is 0.6 seconds.
When I invite Zani to get on the Klimb to get her teeth brushed, her first response latency is 4.8 seconds, and the second is 5.3 seconds. Remember, those are the gaps in time before she moves in response to the cue. And her responses themselves are both slow. She takes several seconds and in one case a false start to actually get on the platform. Recall that in one of the nail trim videos she mounted it (from several feet away) in less than a second.
And note her body language. I slowed down part of the video so we can see her body language as she reluctantly approaches the Klimb for teeth brushing. There are some more “I don’t wanna” signals.
By the way, I’ll send a free PDF of my book to the first person who comments (here on the blog, not on social media, email, or message) to identify what I did in the video that was unconsciously applying negative reinforcement. I didn’t catch it until I watched the clips. (All blog comments are moderated, and there is often a delay before I approve them. But they are queued in order of receipt, so I’ll be able to tell who was first.)
Congratulations to Camille Asmer, who pointed out that I was repeating my verbal cues when Zani was hesitant. In my own words, I was adding pressure to the situation, which I immediately released when Zani got on the Klimb.
Am I Going To Leave Things Like This?
The focus of this post is to show some examples of latency and attendant body language. I’m providing some video of my dog who is not happy to show up for some husbandry procedures that I must do. Am I going to leave things like that? Of course not.
One thing that is important to me in training is to get my dogs not only comfortable with, but happy about the stuff I need to do. Zani is relaxed and waggy during nail trimming, which took a bit of work on my part. There are several ways to approach the teeth brushing issue. I could teach her to bite a stationary dowel so I could take a hands-off approach. Or I could just work in lots of gradual face approaching and handling that doesn’t predict ouchy eyedrops, which will probably be the way I go.
But that’s another post.
Copyright 2020 Eileen Anderson