I have a friend whose mom is an asshole.
She’s emotionally abusive, in the kind of way where she’ll send dozens of text messages about what a terrible child she brought into the world, and then the next day pretends nothing happened and asks about getting dinner.
To my friend’s credit, she’s tried to be the adult and reason with her, talk her down, ask where the anger and outbursts are coming from, but it never seems to get them anywhere. Her mom gets into one of these moods and starts being mean about something, they get into an awful fight, friend ends up crying, her mom pretends nothing happened, and then they repeat in 1-4 weeks.
It’s been this way for the last 8 years of their relationship with no sign of stopping.
But I know you came here for dog training, not stories about familial conflict, so let’s go there.
This is Pepper:
She’s a nine-month-old ball of fur and energy my girlfriend and I got in January of this year.
Mostly, she’s well behaved, but she’s also still a puppy. She’ll destroy a piece of underwear if she gets it, doesn’t totally understand people don’t like getting their shoes chewed on, and she can get over-excited and accidentally nip you while playing tug of war.
But, hey, she’s super cute and does stuff like this so we forgive her:
When we got her, I knew I wanted to make sure she was very well trained. I grew up with a family dog who I loved dearly, but who didn’t get along with other dogs and would look at you with a “that’s cute” expression if you tried to tell him to “come” or “heel.”
Between February and July, Pepper went through three courses of training with Shelby Semel in NYC (socialization, basic, intermediate) and has gotten pretty good at all the important obedience stuff like sit, down, stay, come, look, bed, go home, heel, don’t be a dick to other dogs, etc.
To be clear: I’m no expert in dog training. But from going through these classes and speaking with Shelby and some of the other trainers, I’ve gotten a good sense of a few core principles underlying all dog training, and how you can apply them to teach your dog almost anything you want.
Where these principles get more interesting, though, is how they apply to humans. We like to think we’re rational, self-controlled, free-willed individuals, but for most things, we’re about as automatic and reactive as Pepper is when she sees a sock fall off the dresser.
By understanding these principles, we can not only train our pets but also shape the behavior of people around us.
Maybe even my friend’s mom.
There are three principles that underlie 90% of good dog training:
Let’s say you want your dog to stop bolting out of the door as soon as you open it. If you open the door, your dog starts to bolt, and you quickly shut it and yell “NO!” and try again, your dog has no idea what’s going on and will just get confused. It’s too big of an ask all at once.
Instead, you work them up to doing what you want them to do. First, maybe you just ask them to sit next to the door and then you say “yes” and give them a treat.
Then, you have them sit by the door and you just touch the doorknob. If they stay seated, great, reward them. If not, take a step back to just having them sit and try again.
Once they’re cool with you touching the doorknob, you start turning it just a little bit. If they stay seated, reward, if not, take a step back.
Eventually, you can get to the point of your dog sitting next to the door while you unlock it, open it, step outside, and then invite them to join you. But you have to build up to it, you can’t do it all at once.
As Pepper got older, she developed an extremely annoying habit of losing her shit every time I picked up my shoes to take her for a walk. She’d start jumping on me, nipping at my pants, pulling on my shoes, trying to grab my socks, it made every adventure out of the house difficult and annoying.
To fix it, I slowly worked her in a more productive direction. I sat on the couch, put my shoes on the coffee table, and didn’t touch them, I just waited for her to calm down a bit and then asked her to sit.
When she did, I rewarded her. Then, I’d pick up one sock. If she jumped up, I put it down and made her sit again. Eventually, she figured out she should stay sitting, and I picked up the sock and she stayed seated so I rewarded her.
I broke down putting my shoes on into about as many steps as possible and rewarded her for staying calm every step of the way. That looked something like:
That meant 10 stopping points for rewards and praise while putting my shoes on for a few days, but she quickly figured out what the game was and now I can just put my shoes on with maybe 1 or 2 stops, and when she starts going crazy I can just look at her and she’ll go “okay fiiiiine” and sit down and wait.
Dogs (and really all animals) learn best when you focus on rewarding what you want them to do, instead of punishing them for what you don’t want them to do.
A perfect example of where most people fail at this is when their dog pees inside the house. They might have read in some book like Puppies for Dummies that it’s a good idea to find their dog, drag them by their collar to where they peed, and then push their face in it while saying NO!
This is stupid for a number of reasons. First, your dog has no idea what “no” means in this context. No resisting you pulling on their collar? No hanging out wherever you brought them over from? No sniffing my own pee? Also, what does “no” mean and why are you mad and can we play fetch now?
Second, even if they do make the connection that peeing in the house = angry owner that doesn’t necessarily mean they’ll pee outside, it might just mean they’ll try to hide it from you. They’ll pee in a corner and then slink around and hide from you.
Third, you’re making your dog scared of you, so next time you look pissed and are moving quickly towards them, they’re most likely going to try to run away because they don’t want to sniff their own piss while wondering why you’re in such a bad mood (again).
So instead of negatively reinforcing the behavior, you don’t want, you positively reinforce the behavior you do want. When your dog pees outside, reward them, give them a treat and praise and get super excited about it. It tells them that this thing they just did is good and makes you happy and gets them lots of treats and attention, so they’ll want to do it more.
One way to think about the difference between positive and negative reinforcement is a messy bedroom. Your bedroom only has one or two “clean” states, but it has infinitely many “messy” states. So if you picked a possible state of your bedroom at random, it’s much more likely to be a messy one, and it takes more work to reach a clean state than it does to reach any messy state.
With positive reinforcement, you’re focusing on the clean state. This is the way the universe should be, so thank you for doing that. But with negative reinforcement, you’re focusing on all of the different messy states, and there are infinitely many things your dog could try in response to negative reinforcement without successfully guessing at what it is you actually want.
Imagine trying to organize someone’s apartment, but they only told you where things don’t go. That’d be miserable. You’d be begging them to tell you what they actually want, all the while they’d be shoving your face in the book you put on the wrong shelf yelling “NO! Bad human!”
Humans are somewhat good at generalizing experiences. If I smash my plate on the floor after finishing brunch and the manager gets annoyed at me, I’m going to have a decent sense at dinner that night that someone will probably get annoyed if I smash a plate again.
Dogs are a little more stupid. Yours could be amazing at sitting for you when you’re facing a certain direction in your apartment, but try it in a different room, let alone outside, and they’ll just look at you like “huh?”
To help them generalize the behavior, you need to progressively increase the difficulty through the “Three D’s”:
Distance is how far away they are when you give them the command. Duration is how long you make them hold the command before being rewarded. And distraction is how much other stimuli they can handle while still listening to you.
Taking commands in your quiet living room where it’s just you and the dog is a different beast from taking commands outside in central park surrounded by other dogs and people. Pepper is super obedient when she’s home, or in a small area without too many other people, but put her in a big open area with tons of distractions and she may as well have partial amnesia.
And while it’s annoying, it’s fine, we just need to keep working on it. Where some owners go wrong is they assume their dog is being “bad” because they won’t listen in other environments, when it’s a totally normal thing since they haven’t properly trained their dog to handle it yet.
To fix it, you work on the three Ds. You give them the command from one step, two steps, five steps away, rewarding them each time they successfully listen to you. You make them hold the position for one, two, thirty seconds before getting rewarded. You get them to keep paying attention to you while your partner drops socks on the floor next to them, and reward them for the focus.
If you understand all three principles, you can teach your dog just about anything. For something simple like sit, all you have to do is:
And something more complicated like “get your leash” turns into a string like:
You get the idea.
We can apply all of the same principles to changing the behavior of people around us, and we’re already doing much of it whether we know it or not.
Unfortunately, we mostly do the opposite of everything we should be doing. I didn’t touch on it too much in the dog section, but each of the three principles has a dark side, a way that you can leverage the same principle but in a negative way that ends up hurting your progress towards your goal.
And I think our tendency to do that, to get in our own way by doing the opposite of what the principles suggest, is what leads to much of our frustration with other people.
I wasn’t great at school. I was lucky that I was smart enough to show up and wing it for the test and do just well enough to get mostly Bs, but I was never a straight-A student.
In middle and high school, this was a big source of tension between me and my parents. I knew every report card was going to lead to a discussion of “why didn’t you do better,” and after a few of those I stopped caring about the discussions and treated them as the price of doing business.
I could either spend all semester grueling over work I didn’t care about for a chance at not having that conversation, or I could accept that I was going to have to deal with that conversation and spend my time playing video games instead. It wasn’t a hard choice.
Reflecting on that from the lens of this article, though, I think a big issue was that I never felt good about getting good grades since so much focus was on the bad ones. There were usually a few classes I would do exceptionally well at because they were interesting, but when I brought home the report card the conversation was rarely “oh wow great job in Economics!” it was “why didn’t you get an A in math?”
It’s like expecting your dog to do a perfect 30-second Stay the first time you use the command, and then punishing them when they don’t do it. “Why didn’t you stay sitting for longer?” “Screw you, human, I’m going to go chew up your underwear.”
We make this mistake a lot, though, every time we expect other people to immediately rise to our expectations and level of competency.
Let’s say your significant other never puts their clothes away and leaves them lying around the bedroom. You might ask them nicely, “hey would you mind putting your clothes away instead of leaving them on the floor?” and that’s a great place to start, but what you do next is the most important part.
If they’re good about it for a few days, but then you come home and see clothes on the floor, you can’t lose your shit and say “I thought I asked you to pick up after yourself!” To them, they were improving at it just like you asked, but now you’re mad at them anyway. Now they’ll be much less motivated to keep getting better at being clean because on a subconscious level they’ll feel like you’ll probably end up yelling at them again for it at some point anyway.
But if you thank them and mention how much you appreciate it whenever they do put their clothes away, you can slowly build up the habit over time as they get better and better about cleaning up. No one is going to radically change their habits overnight, and it’s not fair to expect them to, so we have to help them incrementally improve until they reach the goal you’re aiming for.
There’s a big problem with positive reinforcement as we’ve been discussing it that needs to be addressed though: it works for bad things, too.
Whenever someone is acting like what would be described as “an asshole,” it’s usually because that behavior was reinforced in the past.
That asshole manager who yells and bullies his direct reports into doing what he wants? It’s positively reinforced by people obeying him and doing what he’s bullying them into doing.
The asshole friend in the group who flirts with her friends’ boyfriends in front of them? It’s positively reinforced by the attention the guys give her.
Your asshole college buddy who gets a little too drunk at bars and ends up having to be taken care of all the time? You or someone else giving them all that attention is positively reinforcing that getting blackout drunk is a great way to feel loved.
And my friend’s mom from the intro? Every time her daughter responds to the insults by getting into a fight with her, it positively reinforces that being an asshole is a way to get attention from the child she rarely talks to.
So what do we do about existing bad behavior that's been positively reinforced in the past?
The first answer is: don’t reward it.
If your partner’s friend is flirting with you in front of them, ignore them.
If your buddy keeps drinking too much, put them in an ambulance instead of your bedroom.
If someone is being mean to you to try to get attention, ignore them, aggressively if necessary (by leaving, putting in headphones, etc.)
It might seem like you’re doing nothing, but by giving someone zero reinforcement for a behavior they're doing to get the attention they've gotten in the past, you’ll reduce its frequency over time. Any negative reaction will only encourage them.
But sometimes that isn’t enough, which is where things get tricky. How do you un-reinforce a behavior that someone has been rewarded for in the past, and where you can’t just ignore it or walk away? The best answer I’ve found is to interrupt the process by breaking the script they’re trying to play out.
Remember when I mentioned that Pepper will sometimes get way too excited and nippy? You can’t really ignore her when she does that, because it’s self-reinforcing. She’ll just keep jumping around nipping at you. Instead, we interrupt the process by yelping loudly like a hurt dog and backing away, which triggers her “I broke the rules of play” instincts and calms her down. You can pretty much do the same thing to humans.
If you have a friend who tends to put you down in front of other friends, explicitly saying things like “whoa that was really mean,” especially in front of other people, will usually interrupt their script. They’re no different from a dog: they think they’re playing, and by giving them a cue that they broke the rules of play, they’ll adjust their behavior. Especially if you keep repeating the interruption every time the behavior comes up.
In any situation where someone around you is behaving in a way you don’t like, you have to ask yourself how you’re reinforcing them acting that way. If your boyfriend, parent, boss, customer, is continually pissing you off, then you have to take some responsibility for shaping whatever behavior is annoying you.
And it wouldn’t hurt to get a dog to practice on.
Then consider joining the 19,000 other people getting the Monday Medley newsletter. It's a collection of fascinating finds from my week, usually about psychology, technology, health, philosophy, and whatever else catches my interest. I also include new articles, book notes, and podcast episodes.