Consider a related situation that produces a similar effect. When a parent encounters stress at work or elsewhere outside the home, her tone of voice and commands to her children are affected. The voice is slightly more pressed, slightly more impatient. The child, reacting, engages in slightly more defiant behavior and is slightly more oppositional. The two responses reinforce each other, and conflict escalates from there. When the parent's stress is reduced, the child's behavior returns to the less defiant and oppositional mode.
There are two well-studied topics in psychology that help explain oppositional behavior and what to do about it. Reactance refers to a reaction that is directly opposite to some rule or request. It occurs when someone feels he is being pressured and there is some added limit being placed on his freedom or choice. This kind of opposition is not unique to children; in fact, most of the research on it has been done with adults. Reactance explains why people are eager to reject what they think is forced on them and seek out something they cannot or should not have. When you crank up the pressure on a child, you're more likely to see the cognitive component of reactance ("No! I won't do it!") intensified by its emotional component (folded arms, raised voice, increased stubbornness: "Leave me alone!"). The pressure on the child does not have to be as direct as "Do it, or else"; it can take the form of a cloud of eager expectation in the household.
The second topic, ironic processes, has to do with oppositional actions within ourselves. Occasionally, the pressure we put on ourselves to do or not to do something backfires. A familiar case is telling yourself not to think of an elephant, which can lead to thinking about nothing but elephants. If you tell insomniacs to try really hard to go to sleep, they'll find it even more difficult to get to sleep than a control group of insomniacs who received no such instruction. In fact, it can be better to tell insomniacs to try to stay awake, because that command does not trigger all the reactions that pop up when they are told to go to sleep.
We all usually succeed in controlling what we do. A body of research studies has illuminated how we consciously and subconsciously monitor ourselves to exclude things that could interfere with what we're doing. However, when we are stressed or overloaded—when we're trying to multitask, for instance—the monitoring can break down and those other things we're trying to exclude are much more likely to come up and be expressed. Say you're a writer and you have two imminent deadlines to meet, so you need to sit down and focus on the work at hand. But you find that on this occasion, for some reason, your concentration is shot. You spend hours fooling around online trying to find out if your great-great-grandmother really was in the first row when Lincoln delivered the Gettysburg Address, and then you suddenly develop a compelling interest in rereading articles you wrote and published years ago. Distractions seep in because the stress and pressure of the deadlines break down the process that keeps you focused.
So, at times, we all do the opposite of what we intended and what we know needs to be done. When you mispronounce a name when you meet someone, even though you knew better, that's an accidental seeping out of the opposite behavior, an ironic process that occurs just when—or just because—you are putting more pressure on yourself by trying extra-hard to control what you're saying. How does this bear upon your kid's mulelike resistance to your crystal-clear instructions or requests? Well, when you do something that's the opposite of what you want, you do not conclude that you are being manipulative or enacting the pernicious genetic influence of an in-law whose obnoxious traits somehow made it through to the present generation (no doubt by bullying other DNA along the way). The failure to do what's expected is part of normal thought processes, which under certain special conditions lead us to do things we were exactly not trying to do.
Especially if you've hit a sticking point in trying to get your child to do something you really want her to do, it will help to see your child's behavior the same way. Stop thinking about it as a condition, like a tumor, that's located inside the child, and think of it rather as an effect produced by the context of expectation surrounding the desired behavior. This is not a blanket excuse for all misbehavior, criminality, and horrible judgment. Rather, while we are almost always pretty effective when trying to control mental processes and actions of our own, the opposite behavior can come out under conditions of stress, including the stress of heightened pressure and expectation. The research on thought processes and behavior has helped us to understand how this happens and how to keep it from happening.
So how can a parent put this knowledge to work?
The first step, when you've hit the kind of wall we're describing here, is to try to eliminate the cloud of desperation hovering around the behavior. At the clinic, the therapists encourage parents to tell the child it is OK if she does not do the desired behavior, or, if it's essential (bathing, for instance), if she does it superficially and minimally. Parents also practice nonchalance in talking about the behavior—a shoulder-shrugging, laissez-faire attitude of staged indifference. In addition, the therapists ask the parents to find opportunities to explicitly tell their child something like, "Don't worry about this now; you will be able to do this when you get older," a pressure-reducing antecedent that can actually speed up compliance.
If this relaxation of the pressure leads the child to do the behavior on his own, as often happens, the parents should continue to be matter-of-fact. In most contexts, the therapists advocate effusive and enthusiastic praise, following the findings of the research on the strong role of special praise in changing behavior. But this situation is different. Here effusiveness has become associated with stress and implicit pressure, so instead we need some low-key acknowledgement—a simple "That was good" augmented by a nonverbal adjunct like a pat on the arm or the back of the head, all low-key and in passing before moving on to some unrelated subject.