FAYETTEVILLE, Ark. - Psychologists face a challenge when trying to treat convicted batterers, but a new study by University of Arkansas researchers shows that professionals can sort batterers into categories that may be of therapeutic benefit during treatment. This is the first study to show that clinicians can successfully identify different battering personalities, and may lead eventually to improved treatment of batterers.

Jeff Lohr, professor of psychology, graduate student Tricia Witte presented their findings at a recent meeting of the Association for the Advancement of Behavior Therapy. They co-authored the study with professor emeritus Dennis Bonge of the University of Arkansas, Kevin Hamberger of the Medical College of Wisconsin, and Jennifer Langhinrichsen-Rohling of the University of South Alabama.

"We’re seeing that batterers are not all the same," said Witte. Researchers sought patterns using statistical methods and found three distinct subgroups of batterers.

The first subtype, called "antisocial," has no regard for authority, ends up in trouble with the law, and gets into frequent fights. The second subtype, sometimes called passive-aggressive/dependent, tends to be reactive and emotional. Lohr calls this group "angry, chronically depressed guys."

"They’re the kind of guy you don’t want to take out to lunch," he said.

The third type, called "non-pathological," shows none of the antisocial or anger traits of the other two.

"The only thing they do wrong is abuse their spouses," Witte said.

Lohr, Witte and their colleagues looked at profiles from the Millon Clinic Multiaxial Inventory (MCMI) personality measure that showed these three subtypes. They wanted to see if clinicians could evaluate the profiles and identify the subtypes, so they gave 36 MCMI profiles to doctoral level psychologists, along with samples of prototype profiles, and asked them to sort the profiles into the three subtypes.

"They were very accurate with their sorting," Witte said. The clinicians correctly identified 93 percent of the passive-aggressive/dependent profiles and 99 percent of the antisocial profiles. They had more difficulty with the non-pathological profiles, correctly identifying 54 percent.

Being able to recognize different battering profiles may be the first step towards improving treatment of batterers, Witte said.

"We can be more sophisticated and effective with treatment if we know what kind of batterer we’re treating," she said.