Flexible gating between subspaces in a neural network model of internally guided task switching

• Published in: Nature communications Vol. 15; no. 1; pp. 6497 - 20
• Main Authors: Liu, Yue, Wang, Xiao-Jing
• Format: Journal Article
• Language: English
• Published: London Nature Publishing Group UK 01.08.2024; Nature Publishing Group; Nature Portfolio
• Subjects: 631/378/116/1925; 631/378/2649/2150; Analog circuits; Animals; Brain - physiology; Cell culture; Humanities and Social Sciences; Humans; Interneurons; Interneurons - physiology; Mapping; Models, Neurological; Modules; multidisciplinary; Nerve Net - physiology; Neural networks; Neural Networks, Computer; Neurons; Neurons - physiology; Recurrent neural networks; Representations; Science; Science (multidisciplinary); Sensorimotor system; Somatostatin; Subspaces; Switching
• ISSN: 2041-1723; 2041-1723
• DOI: 10.1038/s41467-024-50501-y

Behavioral flexibility relies on the brain’s ability to switch rapidly between multiple tasks, even when the task rule is not explicitly cued but must be inferred through trial and error. The underlying neural circuit mechanism remains poorly understood. We investigated recurrent neural networks (RNNs) trained to perform an analog of the classic Wisconsin Card Sorting Test. The networks consist of two modules responsible for rule representation and sensorimotor mapping, respectively, where each module is comprised of a circuit with excitatory neurons and three major types of inhibitory neurons. We found that rule representation by self-sustained persistent activity across trials, error monitoring and gated sensorimotor mapping emerged from training. Systematic dissection of trained RNNs revealed a detailed circuit mechanism that is consistent across networks trained with different hyperparameters. The networks’ dynamical trajectories for different rules resided in separate subspaces of population activity; the subspaces collapsed and performance was reduced to chance level when dendrite-targeting somatostatin-expressing interneurons were silenced, illustrating how a phenomenological description of representational subspaces is explained by a specific circuit mechanism. The neural mechanism underlying flexible switching between different task rules is unclear. Here the authors analyzed trained modular recurrent neural network models with different cell types to reveal a circuit mechanism for uncued task switching.