Descripción del proyecto
Neuroscience research tries to tease apart how brain circuits work together to give rise to behaviour, namely cognitive flexible processes, such as reinforcement learning (RL). RL requires neurons to be tuned to relevant inputs in order to maximize behavioural performance. Most of the excitatory synaptic inputs in pyramidal neurons are located in the thin dendrites where the majority of the spines are located. Thus, thin dendrites play a crucial role in synaptic integration and experience-dependent plasticity. They mediate local computations that endow single neurons with a variety of input–output transformations and change their information transfer. In agreement, individual dendrites have been proposed as the fundamental computational unit of neurons and the substrate for increasing the computational power of the brain. On the other hand, the neuromodulatory systems are known to be critical regulators of brain states and of adaptive responses relevant to RL. Different neuromodulators, acting through their receptors, can alter intrinsic properties of dendritic ion channels and receptors and synaptic efficacy, having a direct impact on dendritic computations. Thus, in order to understand how the brain allows to quickly improve behavioural performance during RL, it is critical that we 1) gain insight about the activity on thin dendrites, 2) to systematically understand the activity of different neuromodulatory inputs on those and 3) their impact on RL. This proposal aims at identifying how neurons change their properties during RL, by observing and manipulating the neurophysiology of behaving mice, using interdisciplinary approaches and state-of-the-art methods. The work will address several original questions: 1) what is the role of tuft dendritic activity to implement an adequate RL?; 2) what is the role of different neuromodulators in shaping the tuft dendritic activity and on RL?