标题： 走向人类和非人类灵长类大脑-环境相互作用研究的综合方法 显示英文标题

标题： Towards an integrative approach to the study of brain-environment interactions in human and non-human primate

摘要： 通过回顾我20年前开始的科学旅程，我在本论文中强调了考虑大脑组织的必要性，这显然是全局层次化的，但其不同区域的神经元响应也高度分布且混合（通过关注感觉运动皮层-基底节网络）。 我还强调了采用尽可能反映动物真实生活中遇到的情景特征的行为范式的重要性。 最后，我提到了“分解”传统数据处理方式的重要性，并考虑到行为的动态性质，例如通过侧重行为变量和所谓的“潜在”神经活动的研究。 我通过提出我理想实验室的愿景来结束本论文，从现在起5到10年的视角来看。 这个实验室将有两个实验设备，第一个是经典的，用于在受限和因此不自然的环境中进行测试，其数据将与第二个称为“自然主义”的设备的数据进行比较，动物在其中可以表达其完整的的行为谱系。 测试的任务将具有强烈的生态学原则，如模拟觅食的特性，数据将有助于基于在这些任务中逐步增加生态成分来测试假设，所有这些旨在保持这些本质上复杂的数据的可解释性。 显示英文摘要

摘要： By retracing my scientific journey that began 20 years ago, I highlight in this thesis the need to consider the organization of the brain, which is certainly globally hierarchical, but also highly distributed and mixed in the neuronal response of its different areas (by focusing on the sensorimotor cortex-basal ganglia network). I also emphasize the importance of adopting behavioral paradigms that reflect as much as possible the characteristics of the scenarios encountered in the real life of animals. And finally, I mention the importance of "disintegrating" the way data analysis is traditionally carried out, and of taking into account the dynamic nature of behavior, for example by favoring the study of behavioral variables and so-called "latent" neuronal activities. I conclude this thesis by presenting the vision of my ideal laboratory in a perspective of 5 to 10 years from today. This laboratory would have two experimental devices, a first, classic, for carrying out tests in a constrained and therefore unnatural environment, the data of which would be compared to those from a second device called "naturalistic" in which the animals could potentially express their entire behavioral repertoire. The tasks tested would be characterized by strong ecological principles, such as in those simulating the properties of foraging, and the data would make it possible to test hypotheses based on the progressive addition of ecological components in these tasks, all this in order to maintain control over the interpretability of these data which are complex by nature.