IT TURNED OUT THAT there was more than one way to view Jonathan Diamond’s volatility. At least once a day, the 13-year-old would erupt during a confrontation with his parents or teachers, screaming and swearing. In one incident in 2001 he even started to become violent, forcing his parents to call 911.

Therapists, following the tenets of an approach known as parent training, told the Diamonds (not their real name) that they needed to set firmer limits for their son, show him who was boss and offer clear rewards and punishments. Bringing in the police seemed to fit that strategy, and his mother, Beth, hoped Jonathan might be scared into behaving. Yet while he certainly was frightened, the explosions didn’t stop.

“When he got back to his right mind, he was just crying, saying how sorry he was,” explains Beth. But the outbursts kept happening. “It breaks your heart,” she says. “You tell yourself, I know my child is in there, all his good qualities and all his potential, but they’re being destroyed by this thing.”

Ross Greene, director of the Collaborative Problem Solving Institute at the Massachusetts General Hospital (MGH) in Newton Corner and a Harvard clinical psychologist, takes a different view of children such as Jonathan. From his perspective, the problem isn’t permissive parents and ill-defined limits. These kids, Greene believes, have a kind of learning disability—difficulty being flexible, solving problems and dealing with frustration. His solution, collaborative problem-solving, or CPS, hinges on the idea that kids will do well if they can.

For the Diamonds, CPS meant getting to the root of the problem that routinely sparked Jonathan’s outbursts. In the incident that led his parents to call the police, they had pulled the plug on his computer after he refused to turn it off so his grandmother could get some sleep. After meeting with Greene’s partner, psychologist Stuart Ablon, the family decided to try the approach suggested by Greene, who has pioneered the CPS model. During a time of relative calm, Beth, her husband, Sam, and their son discussed why he had so much trouble getting off the computer. “Is it that you don’t want to go to sleep?” Beth asked. “Or you’re instant-messaging a friend and just can’t stop?”

After much thought and discussion, Jonathan figured out that there were some things on the computer he could easily stop doing—viewing favorite Websites, for example—and others, such as playing video games or instant-messaging, that he couldn’t pull himself away from. After he and his parents tried several solutions that didn’t work, such as setting a timer to limit computer time, Jonathan came up with one that did. He would do the things that were hard to stop only until dinnertime—because, for some reason, he found it easier to quit for dinner than at bedtime. After dinner, if he was on the computer, he’d stick to pursuits he could halt more easily.

As simple and intuitive as this approach may seem, CPS represents a departure from one of psychology’s most basic premises of parenting, a paradigm that has held sway for at least 40 years. Greene has been pushing for acceptance of his alternative approach, which he first laid out in a 1998 book, The Explosive Child, and it has been applied with seemingly remarkable results in many families, schools, juvenile detention facilities and inpatient child and adolescent psychiatric units. Even the Massachusetts Department of Social Services (DSS) is beginning to use the CPS model. Yet scientific evidence suggesting that it works has been slow in coming, and it has been questioned by practitioners who consider CPS either too new to be widely adopted or too obvious to create much of a stir.

SINCE LONG BEFORE PSYCHOLOGISTS got involved, the conventional wisdom of parenting held that children should be seen and not heard, and punishments awaited kids who broke the rules. Of course, not all parents took a hard line, and in the 1960s, in a landmark of child psychology, psychologist Diane Baumrind described three types of parents. The authoritarian parent values absolute obedience above all else. A child must stay in his or her place, order is paramount and there is no give and take—the parent is always right. A permissive parent is the opposite—the parent is accepting, doesn’t punish, affirms all the child’s desires, consults with the child about family rules and never exercises authority. (Baumrind later divided this category into two: permissive-indifferent and permissive-indulgent.) Finally, there is the authoritative parent, one who directs the child in a rational manner, encourages give and take, and values both autonomy and discipline but still displays firm control when parent and child differ.

It’s that third style most psychologists have long recommended, and it’s what parent training is supposed to cultivate, particularly in treating so-called defiant children—those who have oppositional defiant disorder (ODD), conduct disorder or other issues, including attention deficit/hyperactivity disorder, which Greene lumps into the “explosive child” category. The treatment concept is largely based on Defiant Children: A Clinician’s Manual for Parent Training, by Russell Barkley, first published in 1987. Parents are advised to learn to issue commands effectively, set up systems of rewards with points or tokens and use time-outs when a child seriously violates rules. For the parents of troubled children, social workers at group homes, teachers and juvenile detention workers, this method has been the recommended approach, and sometimes it’s effective. Studies have shown that parents seem to feel better about their child after attending a parent-training program, and they rely less on critical or violent discipline. Even indicators of child behavior improve with parent training, depending on which of the many versions is used.

Still, many parents don’t fully comply with parent training, and studies showing the effectiveness of the method obviously include only those families that stuck with the program. Moreover, although most parent-training studies record statistically significant improvements, there are few clinically significant changes—concrete alterations the family can recognize and enjoy apart from what may register on a clinician’s lab tests. Finally, there’s not much information on what happens to the child and parent after treatment is complete, because studies tend to end after an initial 12 weeks of therapy.

In devising his alternative to parent training, Greene was influenced by developmental psychology and neuropsychology, both of which emphasize the development of cognitive skills crucial for handling life’s social, emotional and behavioral challenges. Kids who explode really do have a sort of learning disability, Greene believes: He suggests that just as no one would expect a kid with dyslexia to start reading if he were given a sticker every time he got an answer right and a time-out when he got it wrong, it’s unrealistic to think a child lacking the skills to deal with his frustration is going to behave perfectly given those same rewards and consequences.

With CPS, parents start by empathizing with the child as they gather information about why he isn’t meeting expectations. And even though a more authoritarian approach might seem easier and faster, says Greene, over the long haul it sucks up more energy and time than figuring out the real issues and dealing with them beforehand. “When adults solve problems collaboratively with kids, both learn a lot of skills the kids—and sometimes the adults—had been lacking,” Greene says. “Many children don’t have the language to articulate their concerns and can’t take another person’s concerns into account. That’s something CPS teaches them to do.”

In a 2004 paper in the Journal of Consulting and Clinical Psychology, Greene and Ablon described the results of a study involving 47 children who met the criteria for oppositional defiant disorder—they were all extremely disobedient and hostile toward authority figures—and who also had signs of juvenile bipolar disorder or major depression. The researchers randomly assigned half to a treatment consisting of traditional parent training—setting limits, rewards and consequences—and half to CPS for 10 weeks. Both sets of children improved, showing fewer signs of ODD. But those on CPS also had improved relationships with their parents post-treatment (relationships in the other families deteriorated) and were significantly better behaved. And these benefits seemed to persist well beyond the end of the initial study period. “At four months post-treatment, the kids who had received CPS continued to improve,” says Greene. “The ones who’d had parent training started to lose their gains.”

Now other groups are attempting to confirm those findings. The department of psychology at Virginia Tech in Blacksburg is comparing CPS with traditional parent training in a study of 150 kids with ODD and will also consider whether CPS leads to improvement in specific skills the children may lack. But Ablon and Greene think the most important work on CPS is being done in the real world, at large institutions treating the country’s most troubled children.

THE ITALIAN HOME FOR CHILDREN in Jamaica Plain, Mass., provides shelter and counseling to children who have been taken from their families because of abuse or neglect. The children are often explosive, and for years they were treated with the rewards and punishments typical of parent training. But when Simona Arcidiaco took over as chief operating officer a few years ago, she began to question whether the system was working. “Unfortunately, a lot of kids were coming back to us,” says Arcidiaco. “Even when behavioral interventions worked here, they didn’t last when the child was discharged into a new environment.”

Arcidiaco had already begun to move away from the standard system of rewards and consequences when she heard about CPS, which was being used successfully at another Boston-area institution, Cambridge Hospital. She went to a CPS workshop given by Greene and began implementing the model in the summer of 2006. Though it’s early yet, Arcidiaco has seen encouraging signs, including the case of a young girl with selective mutism who improved significantly. “When we admitted her, we didn’t demand that she talk,” says Arcidiaco. “That took away some anxiety, and gradually she began doing things such as raising her hand to go to the bathroom, which she had been afraid to do before. Our approach was to acknowledge what she could and couldn’t do at this time and to adjust our interventions accordingly.” Eventually, the girl began speaking to her roommate and one staff member

Other institutions that have longer track records with CPS have also seen progress. At Cambridge Hospital’s locked, 13-bed inpatient child psychiatry unit, 95% of the kids are admitted for out-of-control behavior and 80% have a history of trauma. In the past, a patient could erupt at any time, punching a nurse or attacking a counselor, and children were being forcibly restrained, often compelled to take psychotropic medication or strapped to a bed. In the nine months before the team at Cambridge Hospital began implementing CPS, there were 281 episodes of restraint and an average of almost 11 staff and patient injuries each month. During the 15 months since implementing CPS, there has been just one episode of restraint and monthly injuries have dropped to 3.3.

The Maine Department of Corrections’ Long Creek Youth Development Center, in South Portland, for 14- to 21-year-olds, many of whom have committed multiple offenses, has also made gains. In 2000 the state began making widespread changes to the juvenile detention system. They started working with Greene and later implemented the CPS model. Long Creek, which in 2000 had more than 100 incidents in which kids had to be restrained, saw that number fall to 38 by 2006, according to Rod Bouffard, superintendent of the center. And as recently as 2005, there were 87 times when a child was put in “observation,” a room for calming down. In 2006 there were only 28 such incidents, and the rate of recidivism in the state’s youth correctional system has fallen to 15%, one of the lowest rates in the country.

NOT EVERYONE IS ENTHUSIASTIC about the collaborative problem-solving model, however. Some critics, even those who agree with its underlying principles, complain that CPS has yet to be proven more effective than traditional parent training. Child psychologist and psychology department chair Stephen Hinshaw of the University of California, Berkeley, notes that only one rigorous study on children with only one type of disorder, ODD, has been published, whereas at least two good trials by two different research teams are typically required before the American Psychological Association considers a treatment “empirically supported.” “The proof is going to be in repeated demonstrations in trials with three groups: children with bipolar disorder, with conduct disorder and with other forms of aggressive or explosive behavior,” says Hinshaw. “And CPS needs to work when other scientists investigate it too. What we have so far are interesting, even fascinating, first-round studies.”

And while some people worry that CPS is too new, others point out that its basic tenets are not new at all. For example, a book published in 1970 and revised in 2000, Parent Effectiveness Training, by Thomas Gordon, advocated parent-child communication, problem-solving and empathy. The idea that some parents and children don’t “fit” well isn’t novel either; it originated more than 30 years ago and suggests that explosive behavior may be one outcome of incompatibility between the personalities of parent and child. “At a conference last year, CPS was discussed, and one comment from the floor was, thank you for reminding us of stuff we should have been doing all along,” says Jacob Venter, a child and adolescent psychiatrist at the MGH.

CPS can also be difficult to implement, primarily because it requires parents and workers to adapt to a new way of looking at behavior problems. In the Italian Home, Maine’s correctional facilities and the Massachusetts DSS, it took months to convince staff members to change—and some never did. “Our staff initially thought CPS was just a way to give in to the kids, and a lot of people outside the juvenile system feel they should be punished—that’ll fix them,” says Maine’s Bouffard.

Inevitably, even people sold on the model start worrying that children are being indulged. But Greene discounts that concern: “Nothing about CPS says you let a child do whatever he wants.” And while he admits that the model builds on previous research, he says that’s how science works. What’s more, studies by other investigators are planned in inpatient, outpatient and institutional settings, and also in schools, Greene says. Meanwhile, he and Ablon each spend at least one day a week speaking about CPS at schools, hospitals and residential facilities around the country, as well as some 12 hours a week supervising institutions that are implementing the model. They have also established a nonprofit organization, the CPS Institute, and as part of a public-awareness campaign called Think:Kids, have launched a Website for parents, pediatricians, teachers and mental health clinicians.

“What Ross and Stuart have come up with is not exactly brand new under the sun, but it’s a really creative synthesis of ideas that may change the way parents view their child’s difficult behavior,” says Hinshaw. “Not as willful, defiant or the product of poor parenting, but as the result of a deficit in some basic skills.”

For Jonathan Diamond, now 20 and newly admitted to the theater department of a Boston college, gaining those skills through CPS has enabled him to live a life that seemed inconceivable back in the days of his daily eruptions. Though still fighting anxiety and other difficulties, he hasn’t had an explosion in about six years, and he says he even uses the method to help other students. “CPS takes a lot of adapting to, and you don’t see instantaneous results, but it produces permanent changes,” he says. “Real solutions just take more work.