DS0407 -

Cerebellar circuits and function in vocal learning – VocaLearn

Submission summary

Motor skills, such as learning to ride a bicycle or to move our tongue and vocal cords during a conversation, naturally emerge from our experience and are executed without awareness. The acquisition of such fine motor skills (i.e. procedural learning) and sensorimotor learning in general involve both the basal ganglia (BG) and the cerebellum. Beyond their common cortical targets these subcortical structures also interact more directly through the thalamus. The respective contributions of the BG and the cerebellum to learning and how they cooperate during the acquisition of motor skills remain unclear, as each structure is usually studied separately.
The present project addresses this question in an original way by studying vocal learning in birds. Songbirds indeed provide an outstanding model to study the neural mechanisms of sensorimotor learning because a specialized part of their forebrain–BG circuitry is dedicated to the sole purpose of song learning. While several laboratories, including my own team, are investigating the mechanisms of vocal learning in the songbird model, and in particular the function of the avian BG-thalamo-cortical loop, the contribution of the cerebellum during song learning remains surprisingly neglected.
Our preliminary results however show that the song-related BG nucleus and cerebellar circuits interact via a pathway through the deep cerebellar nuclei and the thalamus, and that the cooperation between these two structures is essential for efficient juvenile song learning. The role attributed to the sole song-related BG-cortical circuit until now therefore likely relies, at least partly, on cerebellar signals mediated by this cerebello-BG pathway. This challenges the classical view (in songbirds and beyond) of dopamine reinforcement as the only teaching signal in the BG. In order to move toward an integrated theory of sensorimotor learning, we will use the songbird model to disentangle the specific roles of the BG, the cerebellum, and their interaction during learning.
We propose that the cerebellum provides a supervised teaching signal to the BG based on the comparison between real and predicted auditory feedback during singing. A tripartite learning rule at the cortico-striatal synapse, involving cerebellar inputs and consistent with current mammalian data, could serve as the basic algorithm to implement such conceptual framework.
We will combine up-to-date experimental tools (anatomical tracing, chronic and acute neural recordings, optogenetics, behavioral monitoring) to identify cerebellar circuits involved in song learning in zebra finches and to study the function of BG-cerebellar interactions in song learning and plasticity. As subcortical networks often display a closed loop structure, it is extremely difficult to understand the dynamics of neuronal activity within these circuits with classical word models or “box and arrow” models. For this reason, we will develop a theoretical framework to study the cooperation of the cerebellum and BG in song learning, relying on theoretical concepts and methods that draw on deterministic and stochastic dynamical system theory to make sense of electrophysiological and behavioral experimental results. Our project is interdisciplinary in its nature and lies at the interface of neurophysiology, applied mathematics, and theoretical physics.
Beyond vocal learning, our study will contribute to a general theory for the cooperation of BG and cerebellar networks in sensorimotor learning. As abnormal cerebello-BG interactions are associated with movement disorders, in particular dystonia, understanding their function in sensorimotor learning is an essential step to uncover physiopathology and towards developing strategies to prevent and treat disorders. Additionally, finding an avian song-related cerebellar circuit will open a new pathway to address cerebellar function.

Project coordination

Arthur Leblois (Centre de Neurophysique, Physiologie et Pathologies)

The author of this summary is the project coordinator, who is responsible for the content of this summary. The ANR declines any responsibility as for its contents.


CNRS - CNPP Centre de Neurophysique, Physiologie et Pathologies

Help of the ANR 217,740 euros
Beginning and duration of the scientific project: March 2017 - 48 Months

Useful links

Explorez notre base de projets financés



ANR makes available its datasets on funded projects, click here to find more.

Sign up for the latest news:
Subscribe to our newsletter