Like a Rat
Animal research and your child's behavior.
Posted Thursday, Nov. 12, 2009, at 1:23 PM
Psychologists who work with children and families tend to avoid mentioning to parents that the treatments they use are often based on research done on animals. It's no secret that the widely used technique of the timeout was developed in studies on rats or that important early research leading to treatments for anxiety in humans was done on dogs, cats, and other species—but the subject doesn't come up a lot in conversation. We will confess to doing our bit to perpetuate this professional shyness about animal research by tiptoeing around it in our Slate pieces and in the book we wrote together. Why?
Talking about the underpinnings of psychology in animal research tends to make parents uneasy, even upset—not because they fear that the animals were mistreated but because of what they think it implies about their children. "You're saying my kid's like a rat? You're saying my kid's not complex and unique? What about this picture he drew of Spider-Man sobbing in a rainstorm?" When they're outraged, insulted, and indignant, people tend to be less receptive to whatever it is that those who upset them have to say, so there's a sound practical reason for psychologists not to provoke parents needlessly.
And it's understandable that parents do get upset. Some human-animal comparisons are less acceptable than others. It's not considered offensive for a wife to compare improving her husband's domestic chops to training killer whales at Sea World, and it's almost always OK to compare men to dogs—a characterization that many men seem to find flattering. It's also OK for parents to joke about their children being wild beasts. But when an accredited professional proposes such a comparison, defenses go up and tempers shorten.
We don't like the idea that some of the same forces that guide the behavior of rats or monkeys also guide that of our children, part of a larger unwillingness to see our children as mere dumb animals herded this way or that by various influences—from advertising to hormones—other than their own parents' authority and values. Many parents are also committed to a certain model of moral complexity, be it Freudian or otherwise, that figures prominently in their definition of what it means to be human, and so they want to see their child's behavior as expressing suitably human complexity. When they're told that they can, say, control their child's tantrums by employing a simple program of attention and praise that works pretty much the same way that food rewards work with mice, it seems to demean the deep-seated trauma, sorrows, sinfulness, or angst that they believe "caused" the tantrum.
But, whatever the understandable reasons for popular uneasiness about the role of animal research in our understanding of the psychology of parenting, maybe it's a mistake to shy away from talking about it—or, more properly, from talking about nonhuman animal research, since engaging it head-on moves us toward recognizing that people are more like animals, and animals more like people, than we might naturally want to assume. And, more to the point, maybe talking forthrightly about animal research would help psychologists do a better job of explaining their field to the families they're trying to help—where treatments come from, how they work, and why they do or don't succeed.
What happens before (antecedents), during, and after (consequences) a behavior changes how we act. We know a lot about this aspect of human nature thanks to extensive research on rats, pigeons, and monkeys, as well as humans. The area of research known as operant conditioning, which goes back to the beginning of the 20th century, focused in its early stages on the basic processes of learning and performance. One of the most common ways to study how behaviors were influenced by consequences such as rewards was to evaluate what influenced animals to press levers in a laboratory cage. Increasingly elaborate studies were done to see what rewards—food pellets, drinking water—influenced performance. When lawful (scientifically reliable) relations between cause and effect were identified, they tended to be similar across the species that were studied. Different species learned faster, better, or differently from others, but the basic processes of reward and punishment operated in the same ways. If one were to graph a pigeon's, a rat's, and a monkey's responses to reinforcement (the application of a consequence for a behavior that increases the likelihood that the behavior will be performed in the future), the lines would look almost identical.
In the 1950s and '60s, this basic research was extended to humans, who were invited into the laboratory to press levers under various conditions, just as the animals had done. Humans displayed systematic stark and subtle responses to reward very similar to those of the other animals. So add a fourth near-identical line to the graph and label it "Human."
The animal and human laboratory research showed the strong effect of antecedents and consequences on behavior. The next step was to extend the research beyond the laboratory. Could consequences be provided to humans in everyday life that would alter behavior more meaningful than pressing a lever? The first extensions were to adult psychiatric patients who had irrational speech or other symptoms and to children who had difficulty in learning. A new round of studies showed the power of rewards (e.g., attention, praise, tokens) to change self-care behaviors, speech, and the ability to work on tasks. By the 1970s, applications of antecedents and consequences were being extended to a variety of populations in many different settings, including schools from elementary grades through college, in regular and special education; military training camps; nursing homes; hospitals for psychiatric patients, people who abuse drugs and alcohol, and delinquents; and facilities treating individuals diagnosed with autism, mental retardation, or learning disabilities.
Over the years, as animal research on operant conditioning has continued, human research has expanded, and the ways of applying operant conditioning techniques have expanded, the field of psychology has developed what amounts to a science and technology for changing behavior. That body of knowledge and techniques has deep roots in animal research.
Take the technique of timeout, for instance, which has widely replaced spanking as parents', teachers', and baby sitters' first recourse in dealing with a child's misbehavior. In addressing the question of how to suppress or eliminate a particular behavior, like pressing a lever or pausing between two tasks, researchers working with rats and other animals developed the tactic of time out from reinforcement, which is the full and more accurate name of what most people know as the timeout. The procedure consists of a brief period in which the animal cannot receive reinforcers for the usual reward-earning behaviors, such as getting food by pressing the correct lever. From timeout in a laboratory cage came timeout in a corner of the playground, classroom, or living room.
It's useful to remember, when you employ it with children, what the animal researchers found: Only a brief period of timeout is needed in order to change behavior, and reinforcing other behaviors during time-in (that is, the vast bulk of the day when the child is not in timeout) makes a big difference. Only the first minute or so of timeout is really essential for changing a child's behavior, after 10 minutes even a well-conceived timeout becomes counterproductive, and if you give timeouts for hitting then you should attend to and reward a child's success in keeping her hands to herself during time-in. And it can be valuable to remember, when you're tempted to declare, "Now you sit there for the rest of your life and think about why we don't hit," that the effectiveness of time out from reinforcement never depends on the perp's being given a chance to reflect on her sinfulness. The subject's ability to ponder what led to his fate, whether the subject is a pigeon or a child (or, for that matter, a soldier, or a psychologist), is irrelevant.
The treatment of anxiety disorders offers another example of the movement from animal research to human application. There are now effective psychotherapies for specific fears, traumatic reactions, and uncontrollable thoughts and actions associated with anxiety. These are based primarily on graduated exposure to anxiety-provoking situations under conditions overseen by a mental health professional. The profession's path to these techniques began in research on dogs—and then on cats and other species—that studied how fear is acquired.
Once researchers had begun to understand how to develop fear among animals in the laboratory (using puffs of air to induce eye blinks, for instance, or teaching them to respond one way to circles and another to ellipses and then changing the ellipses to look more like circles), the next step was to eliminate it. After the efficacy of controlled exposure to the feared stimulus had been well demonstrated, the next step was to ask whether it could be applied to humans. It is not always clear how fears actually develop in humans, but, nonetheless, it was found that children and adults suffering from a great range of anxiety disorders—panic attacks, fear of open spaces or heights or social interactions, and so on—benefit from the extension of methods developed in the laboratory to eliminate fear in animals.
More recently, some of the key brain receptors that change in response to fear and elimination of fear have been identified in animals. Activating these receptors in animals can improve elimination of the fear, reducing its level sooner and more dramatically. For humans, exposure therapy works well to treat a variety of anxiety disorders, but if they take medication that activates the critical brain receptors before their therapy sessions, the treatment is even more effective.
The lessons go deeper. For decades, the dominant approach to therapy for anxiety assumed that something in a person's early development—usually the mother's fault, natch—led to psychological problems, and the best way to solve these problems was to talk about the past. The animal work has provided a different understanding of environmental influences, such as learning, and how these can affect the biochemical and brain underpinnings of behavior as well as the biological predispositions that make people and animals prone to fear and anxiety. The graduated-exposure procedures based on this understanding, therapies that are demonstrably more effective than any other, do not involve talking through experiences that probably had little or no direct influence on the problem.
Some of the research involved only human subjects, of course. How to get children to comply when parents ask them to do something, for instance, has been studied only with humans. However, the principles on which the techniques of such research are based were developed in the study of animals. This is a key feature of animal research—namely, to develop an understanding of behavioral mechanisms and then use that understanding to apply the knowledge.
In physics and engineering the process is similar. Take the study of flight. Basic science elaborates the features of aerodynamics using mathematics, physics, and wind tunnel tests. This basic research determines the principles and lawful relations of flight, which engineers then translate into the design of aircraft. Because the move from armchair understanding and wind tunnels to actual flight includes important new variables, one still needs to test these translated principles. As in the move from animal research to human application, the move from physics to engineering looks for principles and then applies them, testing constantly every step of the way. No one objects by saying, "I'm terribly insulted that you would ask me to fly to Acapulco in a plane designed according to findings from wind tunnel experiments under highly contrived simulated conditions."
Over the past few decades, there has been a great expansion in methods of studying humans and nonhuman animals. We can look far more deeply at brains, genes, and communication between individual cells, and processes within the cells of animals. Also, we have increased our ability to focus on understanding thought (cognition) and emotions. The net effect of this is to understand many processes that elaborate not only human development, functioning, and health but also emotions, thought processes, worry, and other features in nonhuman animals. (For more on this convergence of humans and animals, read on.
Now, there are many topics in psychology—including leadership, group decision-making, the effects of law and policy on human behavior, witness behavior in the court room, the effects of day care, racism and its causes, teaching reading and math, marital harmony and discord, and crowd behavior—in which there is a solid base of science that rarely or never draws on animal research. It would be wrong to leave you with the impression that psychology always comes back to animal research and to omit mention of the broad scientific swath of work on the unique features of humans. But psychologists tend to do the public a similar disservice when they underplay the importance of animal research.
There are larger points in play here, including a growing sense that the lower animals are not so low and we are not so high. Many psychological characteristics and their biological underpinnings appear to be conserved—that is, many characteristics of human psychology evolved in other animals and have been retained in humans. This is why we can study fish, flies, and frogs in order to learn about ourselves. But there's also a smaller, practical point to make about the psychological technology for shaping children's behavior. Acknowledging the roots of this body of knowledge in animal research is a good way to remind ourselves how to use it properly. Next time you are tempted to give your kid a six-hour timeout so that she has ample opportunity to reflect on the moral consequences of storing used chewing gum in the couch, remember that a rat will not come to regret its trespasses in usefully behavior-changing ways no matter how long you put it in timeout—and, the research clearly shows, neither will your lovely, unique, highly advanced human child.
Alan E. Kazdin, who was president of the American Psychological Association in 2008, is John M. Musser professor of psychology and child psychiatry at Yale University and director of Yale's Parenting Center and Child Conduct Clinic.
Carlo Rotella is director of American studies at Boston College.
Illustration by Robert Neubecker.