One Rhythm Task Did Not Care Which Language You Spoke

What the research found

Spotting dyslexia in a bilingual child is genuinely hard. The standard warning signs, such as struggling to repeat made-up words or to hear the sounds inside a word, also look like the ordinary marks of a child who simply has not finished learning the language of instruction. A child could be flagged as at risk of dyslexia when they are, in fact, just early in their second language. Or a real difficulty could be waved away as “only bilingualism.” This study set out to find tasks that could tell those two situations apart.

The researchers tested 72 children aged nine to twelve, all schooled in Italian. Thirty-seven were monolingual and thirty-five were bilingual, with home languages ranging from Spanish and Arabic to Chinese, Filipino and Ukrainian. Within each group, children were sorted into good and poor readers using standard Italian reading tests. Poor readers scored at least 1.5 standard deviations below average on two or more measures of reading speed and accuracy. The four resulting groups were small, between 16 and 21 children each.

Three language-based tasks separated poor readers from good readers, regardless of whether a child was monolingual or bilingual. On nonword repetition, where a child hears an invented word and repeats it, poor readers scored well below good readers (a large effect, roughly a third of the variation explained). On phoneme blending, a test of hearing and combining sounds, poor readers averaged 67 per cent correct against 80 per cent for good readers. On sentence repetition, poor readers again trailed, though the gap was smaller: 90 per cent of sentences repeated correctly versus 94 per cent.

The crucial question was whether bilingualism, on its own, dragged these scores down. For the two most promising tasks, it did not. Bilingual and monolingual children performed alike on nonword repetition and on phonological awareness once reading ability was accounted for. In other words, a low score on these tasks pointed to a reading difficulty rather than merely to a second-language learner finding their feet. Sentence repetition was the exception: here bilingual children did score lower than monolingual children, so this task carries a genuine language penalty and must be read with more care.

Then there was rhythm. Children tapped along to a beat, and only one measure survived: tapping to fast rhythms. Poor readers were more erratic than good readers when keeping time with a quick pulse. This was the single task in the whole study that responded to reading ability without depending on language at all. It is an appealing result, because a drumming task asks nothing of a child’s vocabulary or grammar. But the authors are careful. Every correlation between rhythm and reading was weak, no stronger than r = 0.3, and these links did not hold up once the statistics were corrected for testing many things at once. The effect was real but modest.

What it means for you and your child

If your child is growing up with more than one language and is struggling to read, the honest headline is this: the confusion is well recognised, and researchers are actively building better tools. This is not a settled science with a ready-made test you can request. It is a careful, early-stage piece of work pointing in a useful direction.

The practical reassurance is that a genuine reading difficulty tends to show up on certain tasks whether or not a child is bilingual. Difficulty repeating nonsense words, or hearing and blending the sounds inside words, is not simply what bilingual children do. In this study, once you accounted for reading, the bilingual and monolingual children looked the same on those two tasks. So a professional who leans on these measures is less likely to mistake a second-language learner for a struggling reader, or to dismiss a real problem as “just the second language.” That cuts both ways, and both matter.

The rhythm finding is the intriguing one, and it deserves calm framing rather than excitement. A tapping task is attractive precisely because it sidesteps language entirely, which would be valuable for a child tested in a language they are still acquiring. But this single study cannot carry that weight yet. The rhythm links were weak and did not survive the more demanding statistical checks. Treat it as a promising lead, not as a test to seek out.

Some limits are worth holding in mind. The groups were small, as few as sixteen children. The study was a snapshot in time, not a study that followed children forward, so it cannot yet tell us whether these tasks predict who will go on to have lasting reading difficulties. And all the children read Italian, a language with regular spelling and simple sound patterns. English, with its far messier spelling, may not behave the same way, so these exact results should not be transplanted wholesale to an English-speaking setting.

Most importantly, the authors themselves are explicit that these are screening tasks for flagging risk, not tools for diagnosing dyslexia. Their own words are that the tasks show “potential,” and that further research is needed to “establish the tasks’ diagnostic accuracy.” A diagnosis still rests on a full assessment by a qualified specialist who can weigh a child’s whole history. If you are worried, the sensible step is unchanged: seek a proper evaluation from a professional experienced with bilingual learners, and mention your child’s full language background so it can be taken into account rather than mistaken for the problem.