On the afternoon of December 6, 2025, the 60th Sui Yuan Xin Guang Young Scholars’ Forum was held in Room 312 of the School of Psychology at Nanjing Normal University. This lecture was delivered by Professor Cai Qing from the School of Psychology and Cognitive Science at East China Normal University, titled “Children's Vocabulary Acquisition and Lexical Network Growth.” The session was chaired by Professor Chen Qingrong, the Chair of the Academic Affairs Office of Nanjing Normal University and the full Professor at the School of Psychology, Nanjing Normal University. Many faculty members and students from the School of Psychology participated in the discussion.

Professor Cai Qing serves as a Professor and doctoral supervisor at the School of Psychology and Cognitive Science, East China Normal University. She is the vice dean of the Institute of Brain Science and Educational Innovation and the director of the Child Brain Imaging Center at East China Normal University, and a young scholar in the National Major Talent Program. Her research focuses on the neural mechanisms of language and higher cognitive abilities, child language acquisition, and the similarities and differences between humans and artificial intelligence in language and other higher cognitive abilities. Her currently serves as Vice Chair of the Language Psychology Committee of the Chinese Psychological Society and a committee member of the Cognitive Neurobiology Branch of the Chinese Neuroscience Society.

 

In her lecture, Professor Cai Qing systematically reviewed recent research on language representation based on large models. She pointed out that while current mainstream models have made progress in quantitative representation, their consistency with human neural representations remains controversial. By comparing text-trained large models with the “small-world network” model constructed from human behavioral data, the study found the latter outperforms in explaining semantic representations in the brain, particularly showing more significant activation in key language regions such as the temporal lobe.

 

Addressing children's vocabulary acquisition, Professor Cai introduced her team's multi-modal language resource repository, encompassing children's language, animated dialogue, and picture book texts. This initiative systematically quantified the impact of diverse input sources on child vocabulary acquisition within the Chinese linguistic context for the first time. The study revealed that during early language acquisition, the contextual diversity of children's language holds the strongest predictive power. As children age, the role of multimodal inputs like cartoons and picture books gradually increases, contributing significantly to the acquisition of complex vocabulary.

 

Regarding the growth mechanisms of lexical networks, the research tracked children's vocabulary acquisition from kindergarten to elementary school and proposed three network growth models: the Priority Connection Model, the Priority Acquisition Model, and the Connection “Bait” Model. Data analysis indicates that children in the early stage rely more on the “priority acquisition” mechanism, learning new words through extensive connections with high-frequency words and existing vocabulary. As they grow older, they gradually shift toward the “priority connection” mechanism, expanding their lexical networks by establishing links with central nodes (such as core concept words). This pattern aligns with the formation mechanism of adult lexical networks.

 

Concluding the lecture, Professor Cai Qing introduced several open-source language resources developed by his team, including a graded corpus for children and vocabulary diversity calculation tools. All resources support online querying and downloading, providing data and methodological support for subsequent related research.

 

 

During the interactive session, faculty and students engaged in discussions on topics such as age-related measurements of vocabulary acquisition, individual differences, and cross-linguistic comparisons. Professor Cai Qing noted that the research team will continue longitudinal tracking studies to further uncover the individualized pathways and cognitive mechanisms underlying vocabulary network development.

 

This lecture featured cutting-edge content and rigorous logic, demonstrating the potential of interdisciplinary research methods in language acquisition while providing scientific foundations for children's language education and practice. The forum concluded successfully amid a vibrant academic atmosphere.