The project that most fires up University of Southern California psychologist Albert "Skip" Rizzo, PhD, is one he hopes will change the future of psychology training: creating virtual patients.

Drawing on artificial intelligence research, Rizzo's team is designing realistic-looking and -acting patients he hopes will be used to train budding clinicians. Using criteria from the Diagnostic and Statistical Manual of Mental Disorders (Fourth Edition), input from expert clinicians and user feedback, his team is programming the patients to respond to questions with realistic answers that help guide students to learn the right information. The "patients" can be viewed on a personal computer monitor or projected onto a screen in life-sized form.

So far, the team has created two such patients: "Justin," a treatment-resistant adolescent boy, and "Justina," an adolescent girl, programmed as the victim of a sexual assault. "I think at first people say, 'This is whacked,'" says Rizzo, "but all of a sudden when they get a legitimate answer to a question or two or three, they start acting like it's the real thing."

Rizzo's dream is to eventually create characters who represent all of the diagnoses in the Diagnostic and Statistical Manual of Mental Disorders and have students regularly use them as training tools.

"Based on a person's training needs, you could then drag and drop any model onto a specific patient," he says.

—T. DeAngelis