Cognitive Control
Cognitive control, the ability to flexibly regulate thoughts and actions in accordance with internal goals, s increasingly understood as emerging from cortical–subcortical–cortical loops, rather than from cortex alone.
We conceptualize the cortical–subcortical–cortical loop as a dynamic control system and aim to identify the fundamental principles that shape its function in cognitive control. We tried to approach this question with both large-scale multi-region recordings and projection-specific labelling tools.
Continual Learning
Continual learning are pivotal in cognitive science and are also at the forefront of artificial intelligence research. While humans and animals can rapidly solve novel problems with limited data, AI relies heavily on extensive datasets and pre-training techniques. Investigating these learning processes will not only enhance our understanding of cognitive functions across species but also inform the development of brain-inspired intelligence algorithms
Problem solving
How the we plan in the mathematical and physical world? We are trying to understand the neural computations underlying multi-step planning and reasoning on both monkeys and human subjects, and also study human specific cognitive functions. We also tried to use AI method like LLMs or embodied AI to facilitate our research.
Exploring techs
We aim to achieve broader and deeper imaging and to enable large-scale electrophysiology with opto-tagging in nonhuman primates.