Changing a habit often takes longer than we think. Habits, AKA reinforced behaviors, die hard.
Here’s what happened when I changed the location of my dogs’ eating areas for the first time in about five years.
Background: Mat Training
Like lots of trainers, I teach my dogs to station on mats. I have rubber-backed bathmats at strategic places around my house. When a new dog comes in, I immediately reinforce her for getting on a mat, and then for staying there. Puppy Clara was already getting on mats in her first hour at my house. The photo below is from her second day.
But I have written before that I am not great about training stimulus control and I am lazy about criteria. I forget whether I have cued things or not. For instance, if I verbally cue my dogs to get on their mats, they are supposed to stay there until released. I pay them well for this behavior. But they also “offer” mat behavior frequently. One will get on a mat, and since this is virtually always a desirable behavior, I will toss her something. Five minutes after a dog gets on a mat I can’t always tell you whether I cued it or not. So if she wanders off, there is no consequence. Ahem.
Anticipation
Our usual mealtime procedure is that my dogs get on their mats in the kitchen and wait while I prepare their meals.
Then when the meals are ready (food toys loaded), I release the dogs and take the meals to three different places in the house so everyone can eat without interruption or worry. When Summer was still in this world, she ate in the front room. Clara has always eaten in the den. And Zani has always eaten right there in the kitchen a few feet from her mat.
So, given my sloppy training, what happened over the years was perfectly predictable to a professional trainer or behaviorist, but it took me a long time to figure out what was happening. In fact, I wrote a post in 2014 complaining about it before I grasped the whole situation.
What happened first is that Zani started leaving her mat before I was finished with the process of loading food toys and would start wandering around the kitchen. This drove me nuts, though I did grasp that it was likely my fault. In the earlier post, I attributed it to the attractiveness of sniffing in the kitchen, and that I hadn’t reinforced her strongly enough for staying put. True, but that was only part of the story.
I didn’t figure the whole thing out until Clara started releasing herself as well. Did Clara wander farther into the kitchen? No! She trotted away from me, into the den. This bothered me a bit, but I didn’t intervene. My thinking was that at least she wasn’t coming farther into the kitchen, so I didn’t mind if she went into another room. (Sloppy trainer!)
Are you getting the picture? I kid you not: after that, Summer started breaking her mat stay as well. But she headed for the front room.
I finally got it. Zani wasn’t leaving her mat just to sniff for crumbs. She was leaving her mat in anticipation of where the big reinforcement was going to happen. She ate her meals in the kitchen, so she was going to that location. Likewise, Clara and Summer were going to wait where their meals would appear.
It took me an embarrassingly long time to realize this. We tend not to think of meals as reinforcement for a behavior. (At our peril!) I was thinking of the treats I gave the dogs on the mats as solely reinforcing mat behavior. And kind of vaguely thinking of their meal as a final reinforcer for that behavior. However, what was the meal reinforcing even more strongly? Leaving the mat to go to the usual eating area. (If that isn’t a good argument for treating richly in position, I don’t know what is!)
Changing Locations and Reinforcement History
But then I changed things up on Zani and Clara. In mid-October of this year (2017), Zani had a spell of intervertebral disc disease. It was not safe for her to be eating out of her usual food toys (either a Kong Wobbler or the saucer-shaped Nina Ottosson dog treat maze). Too much movement. But I didn’t want to switch to a plain bowl because my dogs don’t think they have eaten when they have a meal out of a bowl. I settled for the snuffle mat. I wanted to make it easy and comfortable by elevating it. The steps in my den were perfect for that. So I switched Zani to the den. And to prevent traffic jams, I moved Clara all the way to the front room.
So here is the question: How long did it take Clara and Zani to learn their new meal locations? When did they start running to the new spots (after breaking their stays)?
After 30 meals (15 days) of eating in the new location, Clara was still going to her old location while I finished preparing her meal. But she was going more slowly and tentatively. And when I walked into the other room with her food toy I didn’t have to call her as I had for the first week or so. She was watching and would come running as soon as she saw me head that way.
So her behavior was in the process of changing, but she had not adapted to the new location. That sounds like a lot—30 meals in the new room and she was still going to the old room! But compare it to the number of times she had eaten in there. My rough estimate is that she had eaten more than 3,650 meals in the den. So she had had less than 1% of that number in her new location. Thirty meals don’t sound like so many anymore. That’s the power of reinforcement history. And if there were meals for her available concurrently in the two locations, where would she likely go first? The matching law would tell us: the place she had eaten 3,650 times before.
Zani’s behavior was similar. She had not yet switched to her new location, although she followed me there more readily than she had when we first started.
Habits
This behavior on both of their parts reminded me for the hundredth time that behaviors with a reinforcement history stick around. If your power goes out, how many times do you still flip the light switch when you go into another room? Lots, right? I think the longest my power has ever been out was four days, and my habit of flipping switches was still strong on Day 4. I would be slightly surprised when nothing happened. “Oh yeah! That doesn’t work anymore!”
So when your dog has had a behavior that worked for him for his whole life, keep that in mind. Dogs have behavioral habits just as we do. If the dog continues to perform the old behavior now and then, even though you have trained something new, he is not “blowing you off.” He likely has a long-term habit to change. Remember Clara and how she continued an unreinforced behavior twice a day for two weeks.
Changing Antecedents
There was another joke on me. After starting this post, I decided to film Clara still running to her “old” eating location so I could do a before and after video when she finally changed over. However, when I put up a camera, dammit if she didn’t stay on the mat and wait for a cue. For the first time in a couple of years, she stayed on her mat without releasing herself. I knew putting up a camera tended to change my dogs’ behavior, but I hadn’t realized how much!
In the movie, you can hear me release her with “OK!” or a comment about supper EIGHT times. She doesn’t budge until she sees which way I’m headed with her food. (Speaking of stimulus control: my dogs are not required to move when I give a release cue. Some people teach it that way but I haven’t yet in my training life. So although Clara’s behavior was extremely unusual under the circumstances, she wasn’t technically failing to respond to a cue.) She stayed stock still every time I had the camera out and was filming. When I stopped bringing out the camera, she reverted to running to her old eating area. Stinker.
Link to the movie for email subscribers.
Related Posts
- Herrnstein’s Matching Law and Reinforcement Schedules
- Three Behaviors I’m Tempted To Punish and Why I Don’t
- Dogs Notice Everything
- 16 Behavioral Cues That I Didn’t Train (But Are Still for Real)
- Stimulus Control, Or Lack Thereof
Copyright 2017 Eileen Anderson
I’m pretty much a novice, but I think I know what you are talking about…. I used to work a job where I did call… On the day that my dog taught me how quick he picks up on my cues… I’d already worked a full day of work, but I’d hurried home to let him go to the bathroom before I had to run back to work to do my call shift. I was in a rush, and I was hurriedly trying to encourage him to get his job done which at that moment was pooping… and I’d noticed that every time I said, “Go pee”, he’d lift his leg. Poor guy was getting a humongous leg work out with my worry and repetition of “go pee, go pee, go pee”. I felt bad enough that I was not giving him a fun playtime, and although it was cool that I’d unintentionally taught him to lift his leg on command. My worry was not helping him poop.
Oh man. And here I have tried to teach three dogs to pee and poop on cue, and not very successfully. I taught two halfway to pee on cue (I think they kind of got the cue) but nobody to poop. Good story!
While staying at hotel I noticed that several times during my stay, when exiting my room I was turning left, even though the elevator was to the right. Besides being a directional dyslexic (I don’t know how I got anywhere without GPS) I wondered why, even after several repetitions of leaving my room was I STILL immediately turning the wrong way? My guess was that at the last hotel I stayed at, I needed to turn left and I had already developed a ‘habit’. I meant to keep track of the number of times I exited my next hotel room and which direction I needed to turn, and whether I ever got it wrong. I forgot to keep track, but I did notice that I was paying more attention when I walked out the door and managed to turn in the right direction for the elevators. It turns out it was to the right.
Oh man, I have some spatial dyslexia and particularly in hotels. It’s such a surprise when those elevators are not where they are “supposed” to be.
Oh yes. Dogs in my house eat in their crates with the doors shut. I leave Nina’s door open (she’s the slowest eater I’ve ever had, & even my cat can back her away from her food), & then go back & release the others when Nina comes out. Nina has been–mostly–the only dog for about a year now, but since I want to maintain the behavior I continue to crate-feed.
About 4 months ago Nina was put on meds for her separation anxiety; one pill a day at suppertime. She will not consume any pill voluntarily, no matter how I present it, so I stuff it down her throat in the kitchen. For the first month or so, I had to call her back every night to get her pill. (I didn’t worry about calling her for something she doesn’t like much, since dinner follows immediately.) For about the next 2 months, she ran to her crate, then ran back to the kitchen without being called. For the last month, she spins around in the kitchen, takes a couple of steps towards the dogroom, & then turns her attention back to me. She’s clearly figured out that no food moves until the pill has been pushed down that llooonnngg Flat-Coat muzzle.
Oh that’s very much like our situation! And the interim steps are so interesting, aren’t they? I can so relate to those half-way moves that the dogs do in the middle of the change.
While reading this it sounds to me like the only cue they get when you’re ready prepping food is your body movement. Is that right?
I know the blog is about reinforcement history and habits, and I think you’re spot on, as usual. But one of the biggest lightbulb moments I have ever had in dog training, was when I understood the magical powers of rewarding the release on cue. It sounds like they didn’t associate their dinner with staying on the mat, and that could be because you didn’t give a clear secondary reinforcer, like a release cue or a “yes”.
Hi Agnes,
Thanks for the comment and kind words. I’m not sure I understand you correctly, but I do habitually (haha) give a verbal release cue. I first give it in the video at 0:27 (“OK”). But…over the years they started anticipating the release, and I didn’t do anything about that. They would often all be gone before I thought to release them. When I do use it they are typically eager to jump off the mat and into the next activity.