That first “mama” or “dada” from a baby is one of the most moving moments in a parent’s life. The way children learn to communicate – from early babbling to full sentences – captivates not only parents but researchers as well. And while language development may look effortless from the outside, scientists view it as one of the most complex – and still not fully understood – cognitive processes.
One question has drawn particular scientific interest: how does vocabulary growth relate to the mastery of grammatical rules? Current research suggests that relationship is deeply complex – and far from one-directional. Dr. Karolina Muszyńska and her team at the Faculty of Psychology, University of Warsaw, are working to untangle it. Their ongoing project, “From Word to Sentence” takes a novel approach to studying how vocabulary and grammar interact in children between 18 and 36 months of age.
How does it work? Words and grammar
Language development is about far more than learning words – it is shaped by biological, environmental, and social factors alike. Some children speak in full sentences as early as age 2; others are still naming individual objects at that age. Yet researchers broadly agree on one thing: understanding how children “put together” the words they learn – the relationship between vocabulary and grammar – is fundamental to understanding language development as a whole.
For a long time, the assumption was simple: one area drives the other. Depending on the theory, researchers looked for evidence either that vocabulary shapes grammar, or that grammar shapes vocabulary. One group observed that individual words appear in children’s language before they are combined into simple utterances, concluding that children need to accumulate enough words before they can begin to form sentences.
Other theories argued that grammar is crucial for vocabulary growth: even very young children rely on grammatical cues to decode the meanings of new words. Many studies also showed that once children begin forming their first sentences, their vocabulary grows rapidly – they simply need more words to keep building longer utterances. Recent theories, however, propose that the relationship between vocabulary and grammar development is more complex than previously assumed: bidirectional, and likely shifting as the child grows older.
How to study language acquisition?
The researchers at the University of Warsaw started from a clear premise: to capture the interplay between vocabulary and grammar development – and track how children’s linguistic competence shifts over time – longitudinal studies are essential. These follow the same children over months or even years. In practice, this means asking parents to regularly report on their child’s progress. And that, it turns out, is easier said than done. Such measurements can be genuinely burdensome, demanding significant time and effort from already busy parents.
Adaptive questionnaires may offer a way out. A mobile app developed by Dr. Muszyńska’s team lets parents report on their child’s language development in a far less time-consuming way. The app uses Computerized Adaptive Testing (CAT) – a method in which an algorithm continuously tailors questions to the estimated level of the respondent – making the whole process significantly more efficient.
“In traditional questionnaires, a parent goes through every single test item – including those that are far too easy or far too difficult for their child. In the adaptive approach, items are selected on the fly: each response helps us estimate the child’s current level and choose the next item accordingly,” says Dr. Karolina Muszyńska.
In practice, traditional vocabulary checklists asked parents to work through a long list of words and tick those their child already knew. Such lists included words the toddler had mastered months ago along with words they had yet to encounter. Adaptive questionnaires cut through this by building an individualized path through the item bank – each word the parent is asked about is matched to the child’s current, estimated vocabulary level.
The algorithm, for instance, may skip “mama” or “grandma” the moment it determines the child already uses more advanced words like “dinosaur” or “bubbles.” Each subsequent question is recalibrated to the child’s level, making the whole process far more manageable – and allowing parents to track their child’s language regularly, without facing the same exhaustive list every time.
“From word to sentence”
The study, now getting underway, will follow children aged 18 to 36 months – a crucial window for language development, during which children progress from single words to simple and then increasingly complex sentences. Within just a few months, a toddler’s vocabulary can grow from several dozen to several hundred words, and the first multi-word utterances begin to appear. While observing this process, the team will also analyze the extent to which the child’s environment shapes its dynamics.
“The quantity and quality of the language children hear every day is crucial to their ability to acquire vocabulary and grammatical rules,” emphasizes Dr. Muszyńska.
To better understand these relationships, the researchers are working with both monolingual children and bilingual children acquiring Polish and Norwegian simultaneously.
Recent research by Dr. Muszyńska and the MultiLADA team suggests that monolingual and bilingual children do not differ in the rate at which they reach key language development milestones.* Yet bilingual children naturally hear less of each of their languages than monolingual children do. How such differences in language exposure influence the relationship between vocabulary and grammar development remains an open question – and one with real stakes. Understanding them could be crucial for the accurate assessment of language development and for avoiding misdiagnosis. The “From Word to Sentence” project is an important step toward better understanding and supporting bilingual children’s language development.
Results are not expected until 2027. The researchers hope their findings will advance our understanding of how children acquire language – and how that development can be supported early enough to prevent difficulties from becoming a serious problem once children start school.
The adaptive questionnaire developed by Dr. Muszyńska’s team is based on the Inwentarz Rozwoju Mowy i Komunikacji (IRMiK; Polish adaptation of the MacArthur-Bates Communicative Development Inventory), a tool widely used by clinicians – and familiar to many parents – in the early diagnosis of language development. Developing an adaptive version of IRMiK may ultimately allow the lengthy standard questionnaire to be replaced, at least in certain contexts, by a much shorter adaptive version.
The text was originally published in Polish on the Serwis Naukowy UW website on May 16, 2025.
*Read more: Bilingual children reach early language milestones at the same age as monolingual peers
