Robust Network-level Inference from Neuronal Data Underlying Behavior

单独的神经元是高度不可靠的计算单位，因为它们的试验到试验的反应有很大的可变性。然而，当它们作为一个网络一起行动时，它们会产生强大的大脑功能和精确的行为结果。大规模神经记录技术的出现，如双光子钙成像，使科学家和工程师能够研究大量神经元的活动，以破译它们如何共同编码来自外部世界的信息，并提取它们以产生强大的行为，从而创造了一种范式转变。为了充分利用这些数据，需要在计算上高效和在数学上有原则的技术来进行健壮的网络级推理。这项建议的研究目标是开发这样的方法来从双光子成像数据中推断集合神经元活动的网络级特征，并将这些方法应用于大规模记录，以揭示感觉处理和行为背后的计算原理。这些研究方法包括：建立一个稳健的框架，用于联合推断神经元活动的内在和刺激驱动的相关性，设计一个功能分类来表征神经元活动与感觉加工和行为结果的相关性，以及构建一个估计框架来捕捉高阶同步神经元活动的动力学和功能相关性。该项目解决了现有方法面临的几个突出挑战，包括两阶段分析管道引起的偏颇网络表征，混合外源和内源过程对集体神经元活动的贡献，以及将感觉处理和行为启发作为分离问题进行研究。通过使用小鼠和斑马鱼的双光子钙成像数据，所提出的建模和估计框架将被用于研究系统神经科学中的几个基本问题，例如听觉皮质的立位多样性，感觉处理和决策的相互作用，以及视觉-运动协调。该项目预计将通过提供用于神经控制和神经形态系统的信号处理解决方案来影响技术。这项研究还与教育和推广活动相结合，包括高中水平的研讨会、本科生参与研究和课程开发。该奖项反映了NSF的法定使命，并通过使用基金会的智力优势和更广泛的影响审查标准进行评估，被认为值得支持。

Individual neurons are highly unreliable computational units in isolation, due to their drastic trial-to-trial response variability. Yet, when they act together as a network, they result in robust brain function and precise behavioral outcomes. The advent of large-scale neural recording technologies, such as two-photon calcium imaging, has created a paradigm shift by enabling scientists and engineers to study the activity of large populations of neurons in order to decipher how they collectively encode information from the external world and distill them to elicit robust behavior. In order to fully utilize these data, computationally efficient and mathematically principled techniques for robust network-level inference are required. The research objective of this proposal is to develop such methodologies to infer network-level characteristics of ensemble neuronal activity from two-photon imaging data, and to apply these methods to large-scale recordings in order to reveal the computational principles that underlie sensory processing and behavior. The research approaches include: developing a robust framework for joint inference of the intrinsic and stimulus-driven correlations of neuronal activity, designing a functional taxonomy to characterize the relevance of neuronal activity to sensory processing and behavioral outcomes, and constructing an estimation framework for capturing the dynamics and functional relevance of higher-order synchronous neuronal activity. This project addresses several outstanding challenges faced by existing methodologies, including biased network characterization incurred by two-stage analysis pipelines, intermixing the contributions of exogenous and endogenous processes to collective neuronal activity, and studying sensory processing and behavioral elicitation as disjoint problems. By employing two-photon calcium imaging data from mice and zebrafish, the proposed modeling and estimation framework will be used to investigate several fundamental problems in systems neuroscience such as tonotopic diversity in the auditory cortex, interaction of sensory processing and decision-making, and visuo-motor coordination. The project is expected to impact technology by providing signal processing solutions to be used in neural control and neuromorphic systems. The research is also integrated with educational and outreach activities including high school level workshops, undergraduate involvement in research, and course development.This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.