Despite what we hear in popular self-help books, it is possible to change other people’s behavior for the better. And you can do it without resorting to any of the nasty methods we typically associate with influencing others — nagging, bullying, threatening, ultimatums, etc.
In fact, as a psychologist, teaching my clients how to do this effectively is a huge part of what I do for a living. Because, contrary to another popular myth, our struggles and difficulties aren’t “just in our heads.” With a little creativity, some patience, and a few principles of behavioral psychology, anyone can learn to simultaneously eliminate negative behavior and create positive alternative behavior in the people around them.
For the better part of a century, behavioral psychologists have been studying and implementing a technique called differential reinforcement. Whether you’re interested in cultivating better manners in your children, breaking your spouse of their annoying habit of leaving laundry on the floor or even training your boss to be less overbearing, learning to implement differential reinforcement can significantly improve your quality of life.
Let’s look at a case study from my experience as a therapist to see how differential reinforcement works, and then go through five steps for trying the method for yourself. Two practical examples at the end will show how you can apply it in real-life situations — for when a boss abuses email, and in teaching manners to children.
Vicky the Venter
One of my biggest professional challenges came in the form of a client I affectionately referred to in my head as “Vicky the Venter.” Vicky had a habit of beginning our sessions by immediately launching into a long retelling of all the stressful things that had happened to her over the past week. Of course, a little venting and talking about your week is normal at the start of a therapy session. But it’s rarely therapeutic and can often take up huge chunks of time that would be far more productively spent working on stated goals.
My dilemma was how to get Vicky the Venter to stop venting and begin our sessions more productively. I tried bringing it up with her explicitly, and each time, she agreed that she wanted to start her sessions more productively. But it was a habit she just couldn’t seem to break. With a little help from a clever supervisor, I implemented a differential reinforcement plan that very quickly turned Vicky into one of the most efficient, productive clients on my caseload.
As soon as Vicky started in on her venting session, I stopped asking questions or commenting on her narrative and severely limited the amount of nonverbal interaction I had with her while she was storytelling about all her weekly stress. In other words, I stopped engaging with her beyond a minimum of respect and politeness.
[S]top unwittingly feeding the undesired behavior and… feed the desired behavior in the most appealing manner possible.
Then, at the first hint of Vicky getting off the venting train and bringing up one of our agreed-upon goals for therapy, I made a huge point to be enthusiastic and approving of this shift. I commented on what a great idea that was, my non-verbals got big and positive (smiles, eyes wide open, leaning forward and sitting straighter in my chair), and I asked specific questions about what she had in mind. I laid on the positive engagement pretty heavy.
Over the following few weeks, Vicky’s venting sessions became progressively shorter. By our fourth session after implementing my differential reinforcement plan, her venting was down to a totally acceptable two or three minutes. She even commented on our way out of my office one day: “You know, I really feel like we’ve been making a lot of progress the last few weeks. It’s really encouraging!”
I couldn’t have been more pleased. Vicky went from being one of those clients you kind of dread having on your schedule to someone I genuinely looked forward to working with each week — all because of a little differential reinforcement.
How differential reinforcement works
Differential reinforcement has two basic elements: withholding reinforcement for undesired behaviors and applying reinforcement for desired behaviors.
Vicky’s venting about her weekly stresses was the undesired behavior I wanted to reduce. So whenever her venting behavior came up, I withheld engagement and excess attention, both of which reinforced the venting behavior and made it more likely to occur in the future. But I also made it a point to pile on the attention when she engaged in the desired behavior of bringing up her therapeutic goals. In this case, my application of a reinforcer — engagement and attention — increased the likelihood she would perform this behavior sooner in the future.
In short, the first part of differential reinforcement is to stop unwittingly feeding the undesired behavior, and the second part is to feed the desired behavior in the most appealing manner possible.
Five steps for implementing differential reinforcement
If you want to use this technique to positively influence the behavior of people around you, here’s the recipe:
1. Identify the problem behavior
It’s important from the outset to identify a specific and concrete behavior. You can’t reinforce a feeling or a thought in someone; it must be a physical behavior. In other words, you can work to reduce elbows on the table, showing up late for work, sarcastic comments, and nose picking, but you can’t reliably reduce someone’s sadness, fear, worry, or regret.
2. Decide on an alternative behavior
Once you’ve identified the negative behavior you’d like to reduce or eliminate, consider which positive behavior you would like to substitute for that one. For example: decrease shouting in the classroom and increase using “inside voices,” decrease leaving dirty clothes on the bathroom floor and increase putting them in the hamper, or decrease overly critical comments in weekly sales meetings and increase constructive feedback.
Genuine, heartfelt praise is often a surprisingly effective reinforcer.
3. Teach the alternative behavior
Now that you know the new desired behavior you’d like to increase, it’s important that the person has a clear picture of what that behavior is. Continuing an example from above, if you were the leader of an unproductive weekly meeting, you might take the first few minutes of the next meeting to emphasize to everyone in the room to work on providing constructive criticism going forward. You might even list out elements of what you’d like to have happen — in this case, respectful language and no sarcasm.
4. Pick a powerful reinforcer
While it’s crucial to clearly define what the alternative behavior you want to encourage looks like, you must also have a strong reinforcer to reward the person for adopting that new behavior. Ideally, the reinforcer will have two aspects: It should be something that can be administered immediately after the alternative behavior is implemented, and it should be something the subject finds personally valuable. Genuine, heartfelt praise is often a surprisingly effective reinforcer.
5. In the beginning, reinforce frequently and consistently, then gradually fade out
Many people shy away from the idea of using reinforcers to promote better behavior because it seems childish and they don’t want to have to do it indefinitely. Reinforcement is crucial early on and needs to be applied frequently and consistently to get the new behavior to stick. But once the new behavior has been established as a habit, the frequency of the reinforcer can — and should — lessen and even disappear entirely. You can experiment with decreasing both the frequency and intensity of reinforcement, and to really strengthen the new behavior, you can use a variable reinforcement schedule, which means reinforcing it in an unpredictable frequency.
Differential reinforcement in action
For a more concrete idea of how differential reinforcement looks in real life, let’s turn to some case studies of the strategy being applied in a work situation and a parenting one.
Changing a boss’ bad email habits
I worked with a client once who had significant anxiety problems that lead to major bouts of insomnia. The key driver of his insomnia, we discovered, was that he was often wound-up, tense, and anxious late at night before bed because his supervisor had a habit of sending “urgent” emails at all hours of the night — especially between 10 p.m. and midnight.
After asking some questions about my client’s boss and her late-night emails, I discovered the emails were almost never truly urgent despite almost always being framed as such. I also learned that my client always responded to them, no matter what time it was.
Understandably, my client would sigh longingly in therapy and say things like, “I just wish she would realize that people have lives outside of work and stop sending crazy emails in the middle of the night!” I told my client that while we probably couldn’t change his supervisor’s level of self-awareness, there was a good chance we could change her behavior of sending emails late at night and, along with it, his anxiety and insomnia problems. So we made a plan for it using differential reinforcement.
Step 1: Identify the problem behavior. Identifying the problem behavior, in this case, was straightforward: the boss sending late-night emails. My client and I went a step further and tried to understand that behavior on a deeper level, specifically, why it was occurring in the first place. One thing I suggested was that, unwittingly, my client had actually strengthened his supervisor’s behavior because he always replied immediately and thoroughly to each and every email no matter what time it was. He was “teaching” his supervisor to do the very thing he hated — email him at off hours — by rewarding and reinforcing the behavior.
Step 2: Decide on an alternative behavior. My client decided on cultivating a practice of sending work emails before 9 p.m. He said he actually didn’t mind responding to work emails outside of strict work hours so long as it wasn’t immediately before bed. Because he knew his supervisor liked to work in the evenings, he reasoned that just getting her to send him emails a bit earlier would be both effective and realistic.
There were no major negative effects from my client not responding to the late-night emails.
Step 3: Teach the alternative behavior. To make his supervisor aware of his new “rule,” we came up with a plan for him to set a meeting with her to discuss his work. He would explain that he was happy to respond to work emails in the evening when necessary but that because late-night email was interfering with his sleep — and, therefore, next-day work productivity — he would prefer a 9 p.m. cutoff. He was hesitant and worried she would be upset or flat-out refuse. But he summoned as much assertiveness as he could and reported back after the meeting that his boss said she understood his request but couldn’t guarantee she wouldn’t send late emails from time to time.
Understandably, my client was concerned what might happen when he didn’t respond to a late-night “urgent” email until the morning. We clarified that he wasn’t technically required to under company policy and, furthermore, that he didn’t really have any data on what would happen because he had always responded before. So I recommended that he treat it like an experiment and at least gather some initial data. If it turned out to be a complete disaster, we could always go back to the drawing board.
Step 4: Pick a powerful reinforcer. My client decided the best reinforcer would be to respond to emails that came at an appropriate time with even more promptness and thoroughness than usual. In fact, he set a strict series of notifications on his phone so he would know instantly if she emailed before 9 p.m. so he could respond right away.
Step 5: Reinforce frequently and consistently. Of course, my client didn’t want work email notifications going off in the evenings long term. We discussed how it was essential in the beginning to powerfully reinforce the alternative behavior. Then, once it had been established, he could ease up on those reinforcers and eventually remove them entirely.
As for the results, my client’s supervisor kept sending late night emails for the first week or two, and he described how difficult it was to not reply. In fact, his anxiety and insomnia actually got worse for a time because he worried about the effect not replying would have. But he held fast and stopped reinforcing the undesired late-night-email behavior. Occasionally, his supervisor would send an email earlier in the evening, to which he went above and beyond to send very high-quality, prompt responses back.
Sure enough, by week three, the frequency of the late-night emails was going down while the early-evening emails went way up. Importantly, there were no major negative effects from my client not responding to the late-night emails. He got a few slightly irritated follow-up emails for some of them, to which he politely replied the following morning. Within a month, my client’s anxiety and insomnia had decreased profoundly, and his supervisor had even commented several times about the improved quality of his work.
Discouraging rudeness and encouraging polite manners in kids
I am a parent, and one behavior my toddler has that grates on me is interrupting my wife and me while we’re in the middle of a conversation. We’ll be trying to catch up with each other over dinner after a long day, and all of a sudden, my daughter will start bombarding us with exclamations like “I want chocolate milk!” “It’s too hot!” “Down, down, down — I want to get down!”
Of course, this is normal toddler stuff. I’m sure every parent can relate. It’s also mostly a reflection of my daughter’s perfectly normal desire to want to be included and get some attention from Mom and Dad. Still, it would be nice if she could learn to be a bit more polite in the way she goes about trying to get what she wants. So one of the ways I’ve tried (mostly successfully) to reduce interrupting behavior and increase politeness is to use differential reinforcement.
Step 1: Identify the problem behavior. Again, this one’s pretty straightforward: verbal interruptions when my wife and I are talking to each other, especially during dinner.
Step 2: Decide on an alternative behavior. Importantly, the goal is not to have our daughter or any kid sit silently for half an hour while my wife and I eat and have fully uninterrupted adult conversations. Kids are kids. What’s important is that kids learn how to express themselves politely, not inhibit their wishes altogether. So the target alternative behavior we try to promote is for my daughter to say, “Excuse me, Mom and Dad.”
Step 3: Teach the alternative behavior. Periodically, we remind my daughter that yelling and interrupting are not very nice, and that if she wants to talk to us while we’re talking, we’d be happy to but that there’s a polite way to do it. In the beginning, we also had her repeat “Excuse me, Mom and Dad” back to us so she got practice saying it and hearing what it sounded like.
My daughter still interrupts from time to time. Differential reinforcement isn’t magic, after all. But she is getting better about politely asking a question.
Step 4: Pick a powerful reinforcer. As usual for kids, parental attention — especially enthusiastic, undivided attention — is a huge reward and reinforcer. Whenever my daughter does excuse herself first, we try hard to immediately pause talking to each other and reinforce the behavior by giving her our full attention and praising her for being polite.
Step 5: Reinforce frequently and consistently. At the start, it was important to be a little over the top in stopping our conversation and quickly reinforcing her polite behavior with lots of attention and praise. But as she’s gotten better at it, we don’t necessarily stop right away, or we’ll ask her to wait while we finish and then get back to her. Recall that in later stages, intermittent or variable reinforcement is actually more effective than consistent reinforcement.
As for the result, kids will still be kids, and my daughter still interrupts from time to time. Differential reinforcement isn’t magic, after all. But she is getting better about politely asking a question and saying “excuse me” when she wants something or would like to talk to us when we’re already talking. And, thankfully, it’s happened without too much stress or aggravation on our part.