Part 1 • Short-Term Change Strategy: You can’t control any behavior but your own

(For confidentiality, this is a combination of more than one incident. Details have been altered so that it does not describe a real person.)

In a town in South Dakota, I was working in emergency services, which included consultation to the local jail. The manager in charge of the jail guards called my boss and explained that they had a prisoner who was driving the guards and other prisoners crazy. My boss told the jail manager she had someone who likes to work with that sort of thing. Lucky me.

The prisoner, a woman in her forties, was wreaking havoc. The crime that put her there was minor, but she was causing more problems for the jail staff than all the other prisoners combined. Her jail transgressions included putting a match to her mattress and banging on her cell walls for hours on end; the noise, of course, drove everyone nuts. The jail staff disciplined her the only way they knew how; they took away all her privileges until there were none to take away. Despite the punishment, her defiant behavior continued. The baffled and frustrated staff had blown through their repertoire with no payoff. Far from being subdued, this woman continued her incessant demand for cigarettes, but the guards were in no mood to reward her with the time to supervise her smoking, which was mandatory after she ignited a fire in her cell.

My boss met with me and suggested that we come up with a strict behavioral modification plan for the guards to implement. Instinctively, I shook my head side to side. The guards were already using an informal behavioral modification plan used by parents everywhere: eat your broccoli, and you’ll get dessert. Or, in this case, behave, and you’ll get smoke breaks when we have the time. If that worked, it would have already.

After meeting with the jail manager, I determined that the prisoner was not mentally ill, that her response to an intervention would be predictable. So I laid it out: The prisoner was to be given cigarettes, with the guard supervising her smoking, four times per shift at arbitrary intervals. By arbitrary, I explained that the cigarette breaks should in no way be connected to the prisoner’s behavior. I strongly cautioned against the guards referring to the cigarettes as a reward.

To get cooperation, I guaranteed success, but only if my instructions were followed to the letter. Of course (why else would I be telling this story?), the prisoner stopped her disruptive behavior within the next shift. The jail staff was happy, the prisoner was happy, her lawyer was happy, and I was happy. No other problems with her arose during her two-month stay.

Why did the intervention work? The answer comes in two parts: the reason the prisoner stopped her early disruptive behavior, and the reason the prisoner did not cause any new disruptions during the rest of her stay.

The first part is simple. The guards and the prisoner got caught in a power struggle early on, but once the guards knew how they would behave—regardless of how the prisoner would behave—the power struggle was over. The guards knew that they were in control of their own behavior and stopped worrying about being in control of the prisoner’s behavior—a good idea since the only behavior you can control is your own.

Part 2 • Long-Term Change Strategy: Unearned Fish

If all there was to the above intervention was that the guards no longer responded to the prisoner’s disruption, I believe that the guards would have been in for an unwelcome encore of mischief. But there was a subtle alteration in the guards' behavior that created a continuing change in the relationship between the prisoner and the guards: they gave the prisoner cigarette breaks, not as a reward, but as a kindness. While the guards did not initially think in terms of her cigarette break being kindness, people’s emotions nearly always eventually align with their behavior. It’s like the expression, if you want to feel better, begin by smiling; that is, smile before you feel better to lead your emotions to feel better. Another example: stand straight with your shoulders square, and your confidence will improve.

Once your basic physical needs are met, relationships become more important than anything else. And by relationships, I don’t mean just family, friends, and co-workers, I mean how you get along or don’t get along with everyone you deal with. Every person in your life, for better or worse, can have an effect on your sense of self, which is why you say thank you (or not) to someone you’ll never see again. Or why you might get into a stupid power struggle with prison guards, even when it makes your own life miserable.

People often respond to reward and punishment as you’d expect. If I tell my daughter that I will give her a candy bar if she cleans her room—and she does, and I do—I will conclude that she was motivated by the promise of candy. But what if she went along with my requests most of the time, but not always? The most likely reasons for her occasional refusals are, that she is tired, that she’d rather do something more amusing, or that she’s mad at me.

If my daughter prefers doing something else over the reward of candy, then my control over what constitutes a reward at any particular time is limited. Or, if my daughter is angry with me, then she’s not likely to either give me the satisfaction of a clean room or of accepting candy from me. My daughter may prefer to suffer the consequences rather than submit to my authority.

So what does this say about reward and punishment? That it cannot be separated from the relationship between the involved parties. By what’s most important to her at the time, my daughter gives me permission to reward her or not.

John Lilly was the scientist whose study of dolphins gave them Lassie-like popularity during the 70s. When she was training them to perform (as we humans would characterize it), she noticed that if the dolphin took too long to learn a trick, and was subsequently not rewarded with fish for a long time, the dolphin would become demoralized. That is, it appeared that the dolphin was having a more difficult time learning after a threshold of repetitive failures was reached. Lilly found that the training of the dolphin could be revitalized by giving “unearned fish.” The unearned fish was a reward despite the absence of success with the trainer’s trick. Lilly speculated that the unearned fish demonstrated (to the dolphins) that the relationship between trainer and dolphin went beyond rewards for learning tricks.

The strategy of unearned fish was stumbled upon after much frustration, failure, and false assumptions about the root of the dolphin’s behavior. The trainers had to break out of the mindset they were dealing with only the dolphin’s modest desire to be fed. They had to allow for more complexity than simplistic views of reward and punishment. What was missing was the possibility that dolphin and trainer were in a relationship that went beyond a pure commerce exchange.

Back to my task at the jail:

I explained why the guards' change in behavior had the desired effect in the short run: the guards knew that they were in control of their own behavior and stopped worrying about being in control of the prisoner’s behavior, which ended the power struggle.

Why did this seemingly nonsensical change in the guards' behavior have the desired effect? Because both the jail staff and the prisoner were allowed to break out of their one-dimensional relationship with each other. Both had regarded all their interactions as part of a power struggle. The guards' lone weapon was reward and punishment, leaving the prisoner with two lousy choices: surrender or defiance—neither with which she could feel comfortable. Introducing a conciliatory gesture on the staff’s part—at the height of animosity between staff and prisoner—allowed both an exit from their rigid, predictable stance.

The prerequisite for this working was that there be no catch. Fine print would have cast the gesture in the same old dingy light. So unearned fish is not simply an undeserved reward. It’s an assertion that the relationship between the giver and receiver of a reward is not solely dependent on what have you done for me lately? Or to put it another way, it implies the understanding that all relationships have a history and future (in the mind if not concrete terms). And most people (and some other mammals, apparently) prefer to be on good terms rather than bad. To change a relationship, sometimes all that is necessary is for it to be allowed to be different, and let nature take its course.

To influence behavior, expand your repertoire

There’s a saying attributed to Albert Einstein: “Doing the same thing over and over again and expecting a different outcome is the definition of insanity.” This is wrong for two reasons. First, Einstein never said it. Second and more to the point, doing the same thing over and over again and expecting a different outcome is not the definition of insanity, it’s the definition of ineffectiveness.

The jail personnel were not trained to envision novel solutions, but they demonstrated an ability to change their behavior. They were sufficiently motivated to try something different rather than double down on failure. They had that flexibility in their behavioral repertoire.

The dolphin trainers demonstrated something more interesting, something about how science is done. Something about the myth of objectivity. You can find only what you’re looking for.