Spontaneous nonlinear neural dynamics in information processing and brain cognition

Chang-Song ZHOU (Associate Head and Professor, Physics)

Current ICTS team members: Chang-Song ZHOU (Leader), Yiu-Ming CHEUNG, Liang TIAN

Complex neural/brain activities have been largely treated as noise in neurocognitive studies, but we believe they are self-organized dynamical patterns emerging from nonlinear interaction in neural circuits and large-scale brain networks. This gap offers tremendous opportunities to study brain structural network and dynamics patterns underlying neural information processing and brain cognition by integrating the perspectives and approaches of nonlinear dynamical systems and statistical physics with big brain data analytics and machine learning. The Centre for Nonlinear Studies (CNS) within ICTS has been persistently focusing on this theme in the last decade and has built ourselves a unique profile in such interdisciplinary brain study, achieved with active research collaborations with neuroscience and cognitive neuroscience groups from several international institutions, such as Humboldt University, University of Hamburg, University of Oldenburg, Imperial College London, University of Sydney and Beijing Normal University. Collaborators came to ICTS for short or sabbatical visits and jointly organized focused international workshops and public lectures.

Our research covers different levels of the brain network structural and dynamical organization, from cost-efficiency trade-off in network wiring and firing (PLoS Compt Biol, 2013; 2017; PNAS 2021a), mean field theory of local circuit to explain multilevel critical brain states in experiment (PLoS Compt Biol 2017; Front Syst Neurosci 2020), to cortex- wide propagating waves (J Neurosci 2021) and the characterization of whole-brain functional diversity, segregation and integration in resting state with eigenmode analysis and predicting domain specific and general cognitive abilities across human individuals (PRL2019; PNAS 2021b).

Within ICTS, we are developing collaboration on interpretable machine learning for brain- behavior data. We also used machine-learning to study the neural computational principle by training artificial neural networks to learn tasks like animals (e.g., working memory, PNAS2020), a very promising direction at the interface of neuroscience and AI. These topics can have close interaction with the other themes of ICTS.

Currently, ICTS is establishing an EEG lab (FSC1015). Internally within HKBU, this hardware and the software (concept and methodology, e.g., or EEG data analysis toolbox RIDE) capacity at ICTS has started to make contributions to interdisciplinary collaborations on cognitive studies across faculties (e.g., education (autistic trait), language (hearing loss), social stress, aging, physical education) through internal projects. We anticipate to play an active role in the new HKBU cognitive facility (fMRI Center) under construction to promote collaboration and the opportunities of securing major external grants by HKBU teams.

In the coming years, we plan to further consolidate and expand the collaborations basing on the established capacity at ICTS, through group-building and major collaborative grants covering different levels of spontaneous brain network activity. These collaborative initiative, if successful, will further enhance the international recognition of ICTS for the complex brain study.

Organization of workshop_2017

Germany-Hong Kong Joint Workshop on Brain Signal Variability: Genetic Determinants and Relevance for Cognition and Creativity, July 31-Aug. 2, 2017, Hong Kong Baptist University, organized by Changsong Zhou, Andrea Hildebrandt (Greifswald Uni) and Werner Sommer (Humboldt U).


Public Lectures on “Promoting Creativity Measurement at the Level of Brain and Behavior”, July 31, 2017, Hong Kong Baptist University, organized by Changsong Zhou, Andrea Hildebrandt and Werner Sommer.

In our research group seminar_2019.JPG

Research group seminar on Complex brain network and activity


Meeting on interdisciplinary project: “Machine learning in Neuroscience” (left Aug 22, 2019, right: Sep 12, 2019)