From whole-brain neuronal dynamics to behavior
The ultimate goal of neuroscience research is to understand how the brain performs computations and to construct or repair an intelligent entity equivalent to the brain. The main obstacle to this goal is that the brain contains a vast number of basic computing units—neurons; moreover, these units can communicate through complex long-distance connections, forming an enormous potential for combinations—so much so that the number of possible combinations of neurons in a small fish’s brain far exceeds the total number of atoms likely present in the universe! This complexity is the source of the brain's extraordinary capabilities, surpassing any other type of biological organ. As a result, brain science inevitably becomes a cutting-edge interdisciplinary field that must integrate techniques and methods from multiple disciplines to unravel its mysteries.
Our laboratory is committed to developing this interdisciplinary research paradigm and exploring neuroscience questions to discover neural principles at the whole-brain scale.
Research Strategy
Our strategy adopts a complex systems science perspective to analyze multidimensional information about structure, function, and molecules at the whole-brain scale, revealing the patterns of interaction between different brain regions and between cells. Complex systems science has shown its significant value in fields such as climate, electrical networks, and social networks, explaining the principles of self-organization and emergence through nonlinear processes. This field's research contributions were recognized with the 2021 Nobel Prize in Physics. Such a research perspective is crucial for explaining how the brain operates, where the key lies in the acquisition of high-quality biological data and its effective integration with theoretical methods.
Tools we use
To gain a coherent understanding of the brain from local to global, it’s essential to measure neural activities from multiple brain regions simultaneously. We chose larval zebrafish as model animal, because it is the only genetically accessible vertebrate model animal whose brain is small and transparent enough, thus we can image the whole-brain at single cell resolution. Our research paradigm is:
Whole-brain imaging
Performing whole-brain calcium imaging with high spatial and temporal resolution
Electrophysiology
Recording the cells identified by imaging, with in vivo patch clamp, collecting information at synaptic level
Virtual reality
Using virtual reality to induce behaviors during imaging, and capturing the whole-brain activity during sensorimotor transformation
Data digestion
Analyzing the big data (hundreds of TB), identifying the cells whose activities correlate with sensory or motor features
Modeling
Generating whole-brain models from complex system perspectives
Perturbation
Model validation with optogenetic or chemogenetic perturbations
Bada Shanren
(八大山人)
Double Fish
Joy of fish — research projects
In the past, leveraging a 'whole-brain all-neurons' research paradigm, we have made several discoveries, some of which were quite unexpected. For instance, we demonstrated that astrocytes, in conjunction with the neuromodulatory system and inhibitory neurons, play a crucial role in mediating the cessation of futile behaviors (2019 Cell). We explored how various sensory inputs converge and are categorized within the brain (2018 Neuron). Further, we unveiled an oscillatory circuit that guides animal behavior in the absence of sensory cues (2016 eLife). Moreover, we have developed innovative tools for brain-wide imaging in zebrafish, including fluorescent sensors that detect changes in metabolites (2023 Cell Discovery), and techniques for real-time analysis of whole-brain activity using FPGA-GPU configurations (2024 Nature Neuroscience).
We are continuing our research and focusing on the process of sensory-motor transformation, we analyze the activity of excitatory and inhibitory neurons across the whole brain, while also monitoring neurotransmitter systems such as dopamine, serotonin, and norepinephrine. With this comprehensive neural data, we focus on:
(1) What are the fundamental principles of how the brain encodes sensory information across multiple modalities and types?
(2) What is the whole-brain model that maintains sensory information and generates decision-making?
(3) What are the whole-brain mechanisms that modify cognition and states through behavior feedback during active exploration?
Furthermore, we also apply our methods to broader and more complex systems, such as interactions between the nervous system and the enteric nervous system.
Final goal
We closely cooperate with our center's zebrafish whole-brain neural connectivity atlas platform to integrate the brain's structure and function. By analyzing complete data, we decipher the underlying architecture and information-processing mechanisms of brain networks. Our ultimate goal is to reveal the neural mechanisms of flexibility and robustness in visuo-motor transduction based on whole-brain neural connectivity structures and dynamics, driving a paradigm shift in brain science research from localized to holistic studies.
When these research goals are truly realized, a clear indicator will be our ability to integrate all our discoveries about structure and function into a virtual intelligent agent. This agent will be able to simulate the behavior of real fish and possess the ability to survive in unknown environments. Moreover, given sufficient resources and virtual life conditions, this agent will also be able to evolve, developing more complex and sophisticated strategies for intelligence. Further, when the body of this virtual fish is replaced with that of another species, it will be able to adapt and learn, thriving in new environments and different bodily conditions, demonstrating remarkable adaptability