Imagine two students with similar difficulties. One receives a diagnosis placing him on the autism spectrum. The other gets a label of an intellectual disability. The consequences of both diagnoses can be significant: different expectations, a different trajectory, different opportunities. But what if that difference does not lie only in the student, but also in how we deal with multilingualism, especially when considering multilingual learners special education diagnosis?

A recent systematic review by Cuba and colleagues looks at how multilingual learners are represented in specific categories of special education: autism spectrum disorder, intellectual disability and developmental delay. One thing immediately stands out: out of nearly 300 studies, only ten remain. We are therefore building conclusions on a surprisingly narrow base. Despite the large number of studies, the empirical foundation is rather thin.

At first glance, the results seem contradictory. Multilingual learners are underrepresented in autism but overrepresented in intellectual disabilities. For developmental delay, the picture is less clear. So this is not a simple story.

It becomes more understandable if you do not see multilingualism as a side issue, but as a core variable. Multilingual learners develop across multiple languages. That process is complex, uneven and highly context-dependent. What a student shows in the language of schooling is therefore not necessarily a direct reflection of what they can do. And yet, we often treat it as such.

The problem starts when we treat performance in the language of schooling as a proxy for cognitive functioning. That move quickly turns multilingualism into a risk factor—not because it is one, but because we misinterpret it. We blur language acquisition with learning difficulties, and in that grey area, we assign labels.

This also helps explain the pattern. Autism is often identified through specific communicative and behavioural characteristics, but these can be culturally and linguistically shaped. In multilingual learners, such signals may be less visible or interpreted differently. The result is underdiagnosis. The irony being that bilingualism can actually help children with autism.

With intellectual disabilities, the opposite tends to happen. Language-dependent tests and expectations play a major role. When a student performs less well on these, it is more readily interpreted as a cognitive issue. The result is overdiagnosis.

What we are seeing, then, is not a neutral measurement of differences between students. It is a system that struggles to deal with multilingualism. And that has consequences. Students labelled with an intellectual disability are more likely to end up in more segregated settings, with less access to a rich curriculum. The question of interpretation thus becomes a question of opportunity and equity.

As if that were not complex enough, the authors also show that studies measure these issues in different ways, using different indicators and comparison groups. This makes it difficult to draw clear conclusions and can even amplify or obscure differences.

What this review mainly does is shift the question. Less: What is wrong with multilingual learners? More: how do we understand and assess their development, and what might we be overlooking in the process?