I think I’ve figured something out.
I continue to see the concept of choice bandied about the positive reinforcement-based training world. It can be a code word for a setup that includes negative reinforcement. “I’m going to do something physically unfamiliar or unpleasant to you and you have the choice of staying here and getting a piece of food or leaving and being relieved from whatever it is I’m doing.” I’ve suggested that this is not a laudable kind of choice; as trainers we can use our skills and take our time so that the dog doesn’t want to leave in the first place.
It can also refer to human-centric preference tests, many of which are subject to extreme bias.
But here’s my new realization. I think we have grabbed hard onto the concept of choice because it seems like the opposite of force.
- Instead of pushing the dog’s butt down into a sit, I don’t. The dog now has a choice.
- Instead of restraining the dog for nail trims, I don’t. The dog now has a choice.
- Instead of pulling the dog away from the fire hydrant by his leash and collar, I let him sniff, or I give a cue for another behavior that I will strongly reinforce. He has a choice.
But there is a semantic mismatch here. Force and choice are not opposites.
Force has to do with our actions. In force-based training a human might push, pull, loom over, shock, scare, or drag a dog to get or stop behavior. Those are all concrete, describable actions. The human performs them.
Choice has to do with an internal state of the dog. We conject about it. We pat ourselves on the back for “giving him a choice.” But “making a choice” is not an observable behavior; it’s an internal event. We see the behavior that follows it.
Not only that, but whatever internal state that might exist when we explicitly “give the dog a choice” may also exist when we are not doing this. It doesn’t depend on us. We don’t have to be the center of the “choice universe” for dogs.
Shock collar trainers often say the dog has a choice, and they are correct. The dog, once it understands the system, can “choose” to endure the pain or can “choose” to perform the behavior that turns the painful stimulus off.
I’m not equating positive reinforcement based training with shock training. I’m pointing out that the presence or absence of choice is not the difference between them.
The Opposite of Force
So if force refers to a human behavior, what is the opposite? We can’t say “not using force.” That’s a dead man behavior. What human behavior/s are the opposite of using force on an animal?
Force on our part limits and constrains the animal’s choices, besides often causing pain or fear. Therefore:
I’m going to suggest that in a practical way, the opposite of using force is to proactively remove physical barriers (when safe) and provide lots of simultaneous opportunities for positive reinforcement and enrichment.
This is an undramatic thing for us to do. It means noticing what our dogs like and providing opportunities for them to do it. It means being flexible enough to work around them if a dog suddenly discovers something fun that was not part of our plan. It means not demanding a dog’s attention when she is happily doing something that doesn’t involve us. And it means the opposite, too: giving dog opportunities to be close to us in the ways they prefer.
That last one—not demanding the dog’s attention—can be hard. My rat terrier, Cricket, slept in the bed snuggled up to me her whole life, even after she had advanced dementia. During this time, young Clara slept in a crate right next to my bed. Summer slept in a crate on the bed. (She had been aggressive to Cricket in the past.) Zani was free to be where she wanted, which was usually somewhere on the bed.
After Cricket died, I did away with the crate on the bed and put no constraints on where Summer and Zani slept. I decided not to let Clara out of her crate at night right away. She had always slept there with no apparent frustration. I wanted to give the other dogs a chance to develop new routines. They had seniority. I assumed they would sleep with me.
But they didn’t. Neither Summer nor Zani slept on the bed with me for several months after Cricket died. I missed Cricket and was lonely. A couple of nights I closed the bedroom door with the dogs inside just to have some company, but I felt bad. That was against my beliefs.
So you could say that after Cricket died I “gave the dogs choices.” But let’s operationalize that. What I actually did was to remove barriers (keep the baby gate and other doors open, remove and open up Summer’s crate) and make sure there were lots of comfortable sleeping places all over the house.
Finally, I let Clara out. She got on the bed with me and never looked back. The other dogs eventually came back and have their own quirky sleeping habits.
Two other examples:
- I leave my back door or doggie door open when I’m home and the weather allows. The dogs can come and go. I have to put some limits on this because of safety and social concerns (neighbors) but I do it when I can. You can say I “give them a choice.” But what I am doing is leaving a door open and thus providing simultaneous access to multiple forms of reinforcement and enrichment.
- Clara prefers to drink fresh, running water, an apparent carryover from her feral days. When we have been out and I pour some water from a plastic bottle into a bowl, she likes to drink out of the stream as I pour. If I stop because there is an inch of water already in the bowl, she’ll lap a couple of times, then bump my hand with the bottle until I pour again. This is a pushy behavior but no harm done. I always arrange it so Clara can drink out of the bottle if she wants. I didn’t have to teach her this. I just had to pay attention.
The Fine Line
All this choice talk causes me to worry on behalf of people who are new to training and maybe even new to having a dog as a family member. Giving a dog too much freedom too soon is such an easy mistake and we may be encouraging it in the wrong places. My dogs would love to be underfoot when I cook in the kitchen and wait for me to drop a crumb, but it’s not safe. Hence I have trained them to stay on mats. They would love to run around snatching items from visitors, but I have taught them alternatives. They would love to chew up my furniture and hey—peeing feels good wherever you do it. But I have taught them to chew their own stuff and to pee outside.
Training dogs to live with us involves limiting choices, especially at the beginning. There is no way around that. I think the way to mitigate it is to give them as many opportunities for reinforcement and enrichment we can within the confines we set. There are more limits with puppies or dogs new to our household; we can relax them as the dogs mature, habituate, and learn through training how to thrive in a human household.
Even though it limited her choices in the short term, one of the best things I ever did for my household and for Clara was to keep her crated at night as a youngster. I was consistent. Letting her be unconfined at night before she was house trained and before she had learned to leave my stuff alone would have created many problems. The tippy situation with the older dogs would have made it dangerous. And letting her out before she was ready would have broken both our hearts (permit me that small anthropomorphism) when I was later forced to crate her again for another long period.
Providing Enrichment
So you can train your dog “yes and no” (although it’s much trickier than most methods allow for). You can set up husbandry methods planned around the dog leaving periodically, and call that “giving the dog choices.” Or you could, much less dramatically, observe what your dog likes throughout his life and give him opportunities to do it. You could provide multiple concurrent sources of enrichment. You could notice when he expresses a preference in his own way and honor it when you can. You could practice self-control on your own desires to influence your dog to pay attention to you or stay with you when it is not necessary.
Force is something humans do to dogs. Setting up an enriching life and training with positive reinforcement are the opposites of that. Those are the behaviors we humans can do so our dogs can make choices.
Text and photo of Clara copyright 2017 Eileen Anderson
Photos with pugs and the pugcat copyright 2017 Blanche Axton. Thank you, Blanche!
Thank you, Eileen. Another thought-provoking article from you that has set me thinking and got me examining more closely how I live with my dogs. It’s so easy to slip into demanding –in the nicest possible way — simple behaviours that please us but maybe don’t please the dog. For example, some nights my Terrier doesn’t want to come up to bed with me and it’s a struggle for me to accept this. Your suggestion that we practice self-control over our desires will be ringing in my head next time the wee boy elects to spend the night on the sofa.
Thanks Carole, I think that’s the hardest one for me, too. That time period after Cricket died and everybody else decided to sleep somewhere else in the house was really hard! Not only did I miss having them with me, but it was so the opposite of the scenario I had in my head! Why, why, why?? I’ll never know, but we made it through and they came back of their own accord. Zani still has a strange little schedule but shows up in the wee hours sometimes…. Thanks for the comment! I never know when I put these out in the world how they will resonate. Appreciate your response!
As always, I appreciate the way you think about enrichment in terms of smaller, day-to-day interactions that can be made more dog-friendly rather than just big-picture changes like sniffing walks and puzzle toys.
I think I’ve mentioned before that your bootleg reinforcement article led me to notice that my dogs LOVE smelling my mail. Sniffing the junk mail has become one of the highlights of their day, especially for Indi. It’s 100% free for me and it turns something that used to be an irritating/disappointing obligation into something interesting that I can share with the dogs. Haven and Pan are broad-strokes sniffers who seem to take the whole thing in as one combined experience, but Indi really details it — he sniffs every inch of the mail, especially around seams where a hand would have smoothed down the tape.
When I come home from training appointments with client dogs, I’ve gotten into the habit of sitting on the floor for a minute or two so my clothes are accessible to the dogs. Indi obviously takes great joy in the opportunity to thoroughly sniff my clothes where I have touched the other dogs. If something comes in my house, as long as it’s dog-safe, I will almost always invite the dogs to smell it — groceries, packages, and they really love things from thrift stores. It has given me a whole new appreciation for how crucial scent is for their experience of the world, so I go out of my way to make sure they get access to novel smells during the course of our daily routine. (Our sniffing walks are the same route over and over due to safety issues).
I wish I could give Haven access to more novel sights. She is a “window warrior” and we live on a property with dozens of feral cats nearby, so it’s not healthy for her to have unrestricted access to windows (much less a yard), but she loves patrolling the windows as much as Indi loves sniffing. She could happily perch on the back of a sofa and watch the street all day every day if the cats wouldn’t be a reactivity nightmare.
That’s so cool how that that kind of sniffing is working out with your dogs. Clara is the one who taught me the value of letting them sniff everything that comes in the house. The others are moderately interested but she is passionate!
That’s too bad about Haven and the cats. No easy fix for that one.
Thanks so much for your comment. I’m glad the post resonated.
Good article, Eileen…not just because my crew are included!
Thanks Blanche–and I knew just where to go to find pics of “enrichment outside the box!”. Love your photos.
Hi Eileen,
Thank you, as always, for your insiteful and thought provoking blog.
For me, the real joy in ‘training’ animals is not so much getting them to do what I want to do but in learning to understand what they are trying to tell me.
A few nights ago I was trying to rug my horse Bobby. He is usually pretty good at standing still for me to do this, but that night he kept on backing up every time I went to put the coat over him.
I thought, maybe the rewards weren’t high value for enough, so changed the carrots (good value) to the store bought grain cubes (high value, probably because they have molasses in them).
Nah, didn’t work.
By this time I was beginning to feel a bit frustrated. But then I realised, maybe he didnt want his coat on (it was early evening and not particularly cold). So I said to him, ok, that’s fine, and left it.
Later that night, when I was starting to get cold I went out and asked him if he wanted his coat on. He stood there and let me put it on him!
Rebecca
“Learning to understand what they are trying to tell me” Yes! And they tell us so much if we watch and listen. That’s very cool about Bobby. I bet he’s glad you listened!
The sleeping in the bed thing is always interesting to me. When Annabelle came to me, I was shocked she had NO interest in sleeping on the bed. I spent a ton of time bending my brain to explain why that might be. My bed is on the floor so that my dogs can access it easily so it wasn’t a height issue, etc…I only gated off the stairs at night to keep the elderly safe…so she had the beds in my home office, my bed and several dog beds in my bedroom including a dog bed beside my bed to sleep in…she chose the dog bed by my bed…from the get-go. I would occasionally put her on the bed to see if she’d stay, but she never did….she’d stay for 10 minutes, but never actually lay down to sleep on it…she’d get off and get into HER bed and be happy there.
I just had to shrug and accept that that was where she preferred to be. The reason ultimately didn’t matter…she was most comfortable there and had other options…there are 6 dog beds on my second floor…she chose the bed next to me, but not with me.
Gosh, I miss that old lady.
One of a kind, that one. Summer’s choices are the most interesting in my house. She spends the time from when I go to bed until I turn the lights off next to me in the bed. When I reach for my glasses case or turn off the lights, off she goes to her favorite crate like clockwork. Just for fun I’ve tried to interrupt her, to no avail. But sometime around 4-5 am she gets back on the bed again, at the foot. She uses her ramp for all this coming and going.
I seem to have to learn over and over again to listen to what my dogs are telling me. Recently Sophy started playing keep-away when it was time to be lifted into the car – I thought it was travel sickness that had put her off, but it turned out to be a sciatica-like flare up, which meant that being picked up hurt. Another thing to watch out for. Fortunately she improved rapidly thanks to an excellent osteopath – life got complicated when she wanted to go on a proper walk, which meant using the car, but equally definitely did not want to be lifted into the car…
I do try to give my two choices, but often life is a series of compromises. I choose where we walk, Sophy chooses which direction we take when we get there. How far we go is often a matter of negotiation – Sophy can be very definite if she thinks we should go on, or need to turn back early because it is about to rain! She is equally definite about preferring to eat on the rug in the sitting room, whether or not she wants me to carry her up the stairs (her back is much better, but at bedtime she sometimes stops half way up for a lift), when and what to play, etc, etc. Poppy, on the other hand, is a born follower, and gets rather anxious and upset if asked to make any decisions, which is probably just as well – two dogs wanting to go in opposite directions on every walk might be a tad difficult to accommodate!
(Happy moment – Sophy just jumped up onto the sofa! She has been self limiting and avoiding jumping since her back hurt, and asking me to lift her instead, so this is a sure sign she is not only much better, but is also regaining her confidence. And the fact that I am delighted to see her up there reveals another set of choices my dogs have about which piece of furniture or dog bed to snooze on…)
You are making such great points. First, great that you realized your dog was in pain when her behavior changed. And I love that you pointed out that for some of us (Poppy and me too), sometimes choices aren’t welcome at all. They take energy and can make me nervous too! I’m glad Sophy is doing better. And hooray for lots of beds. I find out interesting things about my dogs from their bed choices! Thanks for the comment.
HI Eileen!
I just had a spare moment to catch up on your blog and was blown away to see your last set of posts, including the one which includes your detailed response to my comment (“I failed to falsify…twice!”). I’m too late to comment directly there, but I really enjoyed reading your analysis and rationale for continuing to give a low value reinforcer to your dogs when they come into the house: enrichment, and reinforcement opportunity. I think we can all agree that that is a great thing to do for our dogs.I also couldn’t agree more with you that enrichment and reinforcement, are opposed to force in animal training. When we pair these two aims (providing the most enrichment and reinforcement we can for the animals in our care) with commitments to be thoughtful about how and when to intrude on our animals spaces, activities, and desires, I think we have a good shot at giving our animals a lot of a good thing: freedom.
Now, I teach philosophy of behavior, so I’m very aware of how slippery that concept is, and getting into the nitty gritty of it here would be a self-indulgent exercise, but I will say that when it comes to technologies of control and influence (e.g. training) the freedom I think we can offer our dogs is not at all sullied by a deep history of differential reinforcement; rather it often depends on it.
To take an obvious example, my own lovely, once-feral dog’s ability to come with me out in public depended on a sizeable amount of desensititzation and counterconditioning to the sights and sounds (and beings) that are part of the urban landscape, as well as countless hours of training to teach her how to interact smoothly with those same things and beings. If I had offered my dog a “choice” when I first brought her home, to stay home or go out, she absolutely would have “chosen” to stay home , to avoid the sounds, smells, sights, and interactions with the scary human world. When I decided to work on this, I was aiming to change her preferences. And I did, and now she “chooses” to come on walk with me, “chooses” to greet (some) people and (some) dogs, “chooses” to go into the previously scary woods to smell and roll and frolick. By creating a deep history of reinforcement, I overrode her very very strong preferences, in the hopes that the life I could give her would be a richer one. But I did so by operating on her preferences directly! And this is how good training works, right? We change motivation, we change desire, we stack the deck so that our animals choose behaviours that are socially acceptable, safe, delightful to us and to other humans, etc. etc.. Or, to say it as elegantly as you did “as trainers we can use our skills and take our time so that the dog doesn’t want to leave in the first place.” The aim of doing this, at least in my book, is not just to make the animal into a delight to us, but to then allow the animal as much freedom and enrichment as we possibly can. I teach my dog a reliable recall in order to give her more off leash freedom. In order to be able to let her chase that smell deep into the woods, knowing that I have a way to ask her to come back if I deem it really necessary. Or, as you put it, (and I agree!) ” the opposite of using force is to proactively remove physical barriers (when safe) and provide lots of simultaneous opportunities for positive reinforcement and enrichment.” In other words, the opposite of force is very often…training.
I think the problem with the resurgence of the concept of choice is not that it is somehow mysterious, or happens “inside our dogs’ heads”, or is unfalsifiable. After all, it might fail the behaviorist’s test, but it doesn’t fail the ethologist’s, and I would know how to train without both sets of concepts. I think the problem with “choice” as I hear it often in the dog training world, is that it is often a substitute for freedom for our animals, real freedom. Freedom FROM (fear, pain, physical restraint, etc.) and freedom TO (explore, sniff, communicate, be left alone and to seek interaction, to have conspecific relationships, etc. etc.). Real freedom isn’t total freedom, and there’s nothing noble in allowing our dogs to choose to be uncomfortable for food, affection, or other reinforcements.
Also, final note– as someone who works with neuroscientists on issues of consciousness and therefore of free will, I would take all the anti-free will stuff with a HUGE grain of salt. It sounds good in popular science, but it often relies on definitions of free will that are untenable, as well as the testimony of the participants to establish the order of events, and nothing we have seen justifies the rejection (OR accepting) of free will as a scientifically proven concept. We simply do not have good experimental design for these issues!
Thank you again for such an amazing and thoughtful set of posts! And my apologies for the endless comment!!!!
Laika, if I wait until I can do justice with a reply to this wonderful comment, it might be a long time! So just thank you so very much. You make some great points, and I promise to take the neuroscientists’ stuff with a grain of salt! (I mostly bring it up to make a point about the misconceptions people have about behaviorism, which is perhaps naughty of me. But the neuroscience is interesting in its own right as well.) Thanks again. I so appreciate your thoughts and your kind words.