Lab news

At the ninth annual Cognitive Computational Neuroscience conference (CCN 2026), held at New York University on August 3–6, graduate student Ahhyun Lucy Lee presented “NeuroBabyLM: A Benchmark for Brain Alignment in Language Models Trained on Developmentally-Plausible Datasets” as poster A25 in Poster Session A on August 4.

Polygenic risk-informed white matter integrity improves deep learning-based prediction of youth depression, led by first author Jungyoun Janice Min (MS, Brain & Cognitive Sciences) with lab members and collaborators, was published in Communications Medicine. Adding polygenic risk information to diffusion-MRI measures of white matter integrity improved deep-learning prediction of depression in adolescents.

The paper “Learning brain dynamics across distinct scaling regimes reveals psychiatric signatures,” led by Sangyoon Bae, was published in Communications Biology. The study presents Multi-Band Brain Net (MBBN), which separates resting-state fMRI dynamics into distinct scaling regimes and uses frequency-aware attention to examine predictive performance and network patterns across UK Biobank, ABCD, and ABIDE.

The lab has been selected for the National Research Foundation of Korea’s Mid-Career Researcher Program, a competitive individual research grant that provides a stable foundation for developing the lab’s research program. Announcing the news, PI Jiook Cha thanked the members — none of this would have been possible without them.

Decoding Dynamic Visual Experience from Calcium Imaging via Cell-Pattern-Aware Pretraining, a paper by PhD student Sangyoon Bae, has been accepted to the International Conference on Learning Representations (ICLR 2026), to be presented in Rio de Janeiro. The work decodes dynamic visual experience from calcium-imaging recordings using cell-pattern-aware pretraining.

Jubin Choi was selected for a Spotlight Presentation at the 2025 NeurIPS Workshop on Foundation Models for the Brain and Body for “NeuroMamba: A State-Space Foundation Model for Functional MRI”. This model addresses linear-time temporal context scaling on 4D neuroimaging datasets.

On November 8, 2025, Moon Seokjin, Wang Heehwan, and Oh Kyungjin presented “Reconstruction of Memory” at OB/Scene Focus, with research advice from Jiook Cha. The program introduced early experiments using EEG and AI to express emotion and memory through art.

“Polygenic architecture of brain structure and function, behaviors, and psychopathologies in children” — led by Yoonjung Yoonie Joo with lab members Bo-Gyeom Kim, Gakyung Kim, Eunji Lee, Jungwoo Seo, and Prof. Jiook Cha — was published in Nature Communications. The study analyzes patterns of covariation among polygenic scores, brain imaging, cognition, behavior, mental health and environmental measures in ABCD children. It connects observations across levels to investigate individual differences in development.

Eunji Lee and Bo-Gyeom Kim’s study titled “Polygenic scores for psychiatric traits mediate the impact of multigenerational history of depression on offspring psychopathology” was officially published in Molecular Psychiatry.

Dongyeop Han, Ahhyun Lee, Taeyang Lee, and Sebin Lee’s pioneering work “DIVER-0: A Fully Channel Equivariant EEG Foundation Model” was accepted for a Spotlight presentation at the ICML 2025 Workshop on Generative Biology & Healthcare.