Despite mounting evidence that suggests dyslexia has biological roots, many studies show its prevalence differs from one country to the next. The number of dyslexic 10 year-olds in Italy, for example, was recently found to be half that in the U.S. So Eraldo Paulesu of the University of Milan Bicocca and colleagues devised a clever experiment to test whether different languagesor different neurological problemswere to blame. In today's issue of Science, they report that dyslexia in fact appears to have a common neurological cause across borders, but that complexities in certain languages can make the problem worse.

The researchers tested dyslexics who spoke English, French and Italian. Unlike Italian, the first two languages have what is called irregular orthography: the same combinations of letters don't always sound the same. (Consider the pronunciation of mint and pint, cough and bough, or clove and love.) The researchers took positron emission tomography (PET) scans of the subjects' brain activity while they took both reading tests and word sound tests. They discovered that whereas the Italian subjects did better on reading tests, they did as poorly as the English and French dyslexics on the word sound tests. All three groups showed the same reduced activation of the left temporal lobe while reading.

The results reinforce the theory that dyslexia arises from some sort of deficit in processing language sounds. "This research proves the existence of a universal neurological basis for dyslexia," says senior author Uta Firth from the Institute of Cognitive Neuroscience at University College London. "It also highlights the impact that the complexity of orthography can have on reading proficiency of dyslexics and therefore on the severity of the disease and the ease of diagnosis. This means that in the Italian population there may be hidden cases of dyslexia. On the other hand, otherwise mild cases of dyslexia may appear far worse in irregular orthographies like that of English or French."