De novo motor learning creates structure in neural activity that shapes adaptation

• Published in: Nature communications 2024-05, Vol.15 (1), p.4084-4084, Article 4084
• Main Authors: Chang, Joanna C., Perich, Matthew G., Miller, Lee E., Gallego, Juan A., Clopath, Claudia
• Format: Article
• Language: English
• Subjects: 631/378/116/1925; 631/378/2632; Activity patterns; Adaptation; Adaptation, Physiological - physiology; Animals; Humanities and Social Sciences; Learning; Learning - physiology; Models, Neurological; Motor Cortex - physiology; Motor skill learning; Movement - physiology; multidisciplinary; Nerve Net - physiology; Neural networks; Neural Networks, Computer; Neurons - physiology; Perturbation; Population dynamics; Recurrent neural networks; Science; Science (multidisciplinary)
• ISSN: 2041-1723; 2041-1723
• DOI: 10.1038/s41467-024-48008-7

Animals can quickly adapt learned movements to external perturbations, and their existing motor repertoire likely influences their ease of adaptation. Long-term learning causes lasting changes in neural connectivity, which shapes the activity patterns that can be produced during adaptation. Here, we examined how a neural population’s existing activity patterns, acquired through de novo learning, affect subsequent adaptation by modeling motor cortical neural population dynamics with recurrent neural networks. We trained networks on different motor repertoires comprising varying numbers of movements, which they acquired following various learning experiences. Networks with multiple movements had more constrained and robust dynamics, which were associated with more defined neural ‘structure’—organization in the available population activity patterns. This structure facilitated adaptation, but only when the changes imposed by the perturbation were congruent with the organization of the inputs and the structure in neural activity acquired during de novo learning. These results highlight trade-offs in skill acquisition and demonstrate how different learning experiences can shape the geometrical properties of neural population activity and subsequent adaptation. Using recurrent neural networks, here the authors show that learning the same task through different experiences can lead to important differences in how neural activity is structured. These differences can play a crucial role for subsequent adaptation, with networks that are equally good at the initial task showing opposing trends in adaptation.