The tones for “mother” and “horse” in Cantonese are very similar. The answer behind the question “why can some children pronounce it accurately while others cannot?” may be related to children’s language development.

Recently, the team led by Professor Patrick Wong, Stanley Ho Professor of Cognitive Neuroscience and Professor of Linguistics, along with Professor Ting Fan Leung and Professor Simon Lam of the Department of Paediatrics, Faculty of Medicine, CUHK has constructed a predictive algorithm which uses a simple electroencephalography (EEG) test to forecast language development in individual children. Combining with early intervention, it could reduce the severity of potential language impairment as well as optimise language learning for children at the earliest possible time.

Prediction with 92% precision

The team enrolled 118 infants up to 12 months of age from Chinese-speaking families and administered an EEG test. During the test, metal discs were attached to infants’ scalps and earphones were put into their ears to measure how their nervous system responded to different Chinese speech sounds. The 30 minutes EEG test is non-radioactive and therefore does not cause any damage to infants’ brains.

About 1.5 years later, when these children could use spoken language for communication, the team measured their language functions to see who had become better or worse language users. Based on the EEG data obtained during infancy and the later language data, the team has constructed a predictive algorithm forecasting language development in individual children, with 92% precision. In the future, the predictive algorithm can easily predict the language development of newborn infants who have taken the simple EEG test, and it also indicates to parents at the earliest possible time that intervention may be needed.

“An important feature of the technology is that it is precise enough to make predictions at the level of the individual child.” stated Dr. Nikolay Novitskiy, a co-author of the study at the Brain and Mind Institute, CUHK. The current hearing screening test can only measure infants’ hearing sensitivity. Also, infants cannot talk, thus behavioural methods cannot be used to reliably access their language development. Therefore, a simple EEG test, which helps to identify children with potential language impairment one to two years earlier than the traditional one, is preferable.

Linguistic and Paediatric researchers from CUHK have developed a test to predict infants’ future language ability by electroencephalography (EEG). From left: Professor Ting Fan Leung, Professor Patrick Wong, and Ms. Chow and her daughter who participated in the research.

Early detection is important when it comes to children’s development

Language ability has a long-term impact on many domains of children’s development including social interaction, behaviour, mental health, cognition, even academic achievement and employment. Therefore, it is extremely important to screen out children who may suffer from a developmental language disorder, so that parents can act and provide appropriate intervention and training as early as possible.

Professor Wong further explained the importance of early intervention, “The nervous system is at its most plastic in the early years. Early intervention is the most effective method to optimise language learning, as well as to reduce the severity of potential language impairment. The earlier the intervention, the higher returns delivered than those administered later in life. Children who are ranked low in language and communication ability among their peers at 2 years of age continue to rank low at age 15 years. That’s why having a prognostic tool which predicts infants’ future language ability is important. With early intervention and changes in learning environment, language can be improved.”

Develop database for English language and newborns

Supported by the Innovation and Technology Bureau, Professor Wong has established Foresight Language and Learning Solutions Limited as a result of knowledge transfer to promote the predictive algorithm. Speaking of its future development, Professor Wong said that the test can now be applied to Cantonese and Mandarin speaking communities, and they are working with overseas universities to study the accuracy of the test on English language development.

Also, the research team hopes that the EEG test and the predictive algorithm can be adopted in the routine universal newborn hearing screening test in the future. That would not only rule out infants’ hearing impairment, but also obtain data from new-borns for the algorithm.