Can a case of strep throat lead to a mental disorder? Some children seem to acquire behaviors associated with obsessive-compulsive disorder (OCD) after being infected with the Streptococcus bacterium, but for decades skeptics have claimed the connection is nothing but a coincidence. Now a new study in mice offers compelling evidence that strep can indeed affect the mind.

In the 1980s Susan Swedo, a pediatrician at the National Institute of Mental Health, came across several cases of children who seemed to have developed tics and behaviors resembling OCD, such as excessive hand washing, overnight. Swedo noticed that the children in all the cases had recently recovered from strep throat. The traditional strep symptoms were gone, but when she did laboratory tests, Swedo found the children’s blood still contained high levels of strep antibodies. Perhaps most compelling, the symptoms seemed to abate after renewed treatment with antibiotics. Swedo became convinced that the symptoms were the result of an overactive immune response to strep bacteria. She suggested a new diagnosis called “pediatric autoimmune neuropsychiatric disorders associated with strep,” or PANDAS.

Because strep throat is quite common in youngsters, many people claimed Swedo’s evidence was more coincidence than fact. Still, she has amassed a fair amount of clinical data over the years and has managed to win over many of her critics. The new study, conducted by researchers at Columbia University’s Center for Infection and Immunity, promises to sway many of the holdouts by providing the first conclusive evidence that strep antibodies can induce neurological and psychiatric symptoms in healthy animals.

The researchers started by injecting mice with strep bacteria. Then they injected a new set of mice with strep antibodies from the infected mice. The researchers found that not only did both sets of mice exhibit the same behavioral symptoms—including anxiety and compulsive rearing and flipping—but that the behaviors appeared to be linked to antibody deposits in brain areas that have been implicated in human studies. Other groups have attempted to induce PANDAS in animals, says James Leckman, a pediatric psychiatry researcher at Yale University, who was not involved in the study, but the results from those studies were inconsistent. “The design they used for this paper was much clearer,” he says.

Mady Hornig, the principal researcher behind the Columbia study, is now working with Swedo to apply the animal findings to a clinical setting. “We hope the mouse PANDAS model can help refine the diagnostics for the human disorder,” Hornig says. A more accurate method of diagnosing PANDAS could help get affected kids the right treatment—and Swedo estimates that these kids may make up as much as 25 percent of children diagnosed with OCD and tic disorders, such as Tourette syndrome. Farther down the road, the newly developed PANDAS mice could allow researchers to devise better or more specific treatments than the antibiotic regimens currently being used.