CRI: A Model-Driven Investigation of Hippocampal Functionality in the Rat

ID:伊本- 9634424。海马体是大脑中一个极其重要的区域。众所周知，海马体对大多数哺乳动物（包括人类）的记忆相关表现至关重要。记忆的最佳候选机制——长期增强（LTP）——已知发生在海马体的许多部位。最后，它也是一个重要的癫痫病灶区域，也是人类阿尔茨海默病的主要靶点。因此，了解这个大脑区域的功能是至关重要的。到目前为止，大多数关于海马体的研究都是由生理和行为实验驱动的。这产生了令人印象深刻的数据体，但对海马体功能的系统理解仍然缺乏。要使海马体的实验研究在理论与实验的相互反馈中合乎逻辑地进行下去，这样的理解是必要的。提出的研究的广泛目标是开发和验证海马体作为一个系统的具体定量模型。这个基于现有数据的模型将为提出精确的问题和设计系统的实验系列提供一个理论框架。此外，由于其计算性质，该模型将允许在模拟的基础上做出具体的预测，从而为不断提高模型本身的质量提供了一种手段。在短期内，该项目将侧重于空间学习（例如，记住地点）的问题，这方面有大量组织良好的实验数据。解决的具体问题将是：1)海马体的多重输入的作用；2)相似地点和环境的歧视；3)时间在信息存储和回忆中的作用。该模型将使用在神经网络和动力系统领域发展起来的方法，从现象学层面进入更大的生理细节层面。该模型的假设和预测将通过行为和生理实验进行严格验证。因此，该项目代表了计算和实验技术的重要整合。未来的工作将把结果推广到记忆功能的其他方面。

ID: IBN-9634424. PI: Ali A. Minai The hippocampus is an extremely important region of the brain. It is well-established that the hippocampus is crucial to memory-related performance in most mammals, including humans. the best candidate mechanism for memory --- long-term potentiation (LTP)--- is known to occur in many parts of the hippocampus. Finally, It is also an important epileptic focus region and a prime target of Alzheimer's disease in humans. It is thus critically important to understand the functioning of this brain region. So far, most research on the hippocampus has been driven by physiological and behavioral experiments. This has produced an impressive body of data, but a systematic understanding of hippocampal function is still lacking. Such understanding is necessary if experimental research on the hippocampus is to proceed logically with mutual feedback between theory and experiment. The broad objective of the proposed research is to develop and validate a concrete quantitative model of the hippocampus as a system. This model, based on existing data, would then provide a theoretical framework for posing precise questions and designing systematic series of experiments. Further, because of its computational nature, the model will allow concrete predictions to be made on the basis of simulations, thus providing a means for continuously improving the quality of the model itself. In the short term, the project will focus on the issue of spatial learning (e.g., remembering places) for which a large amount of well-organized experimental data is available. The specific issues addressed will be: 1) The role of multiple inputs to the hippocampus; 2) Discrimination between similar places and environments; and 3) The role of timing in the storage and recall of information. The model will use methods developed in the area of neural networks and dynamical systems, proceeding from a phenomenological level to one of greater physiological detail. The hypotheses and predictions of the model will be rigorously verified using behavioral and physiological experiments. The project thus represents a significant integration of computational and experimental techniques. Future work will generalize the results to other aspects of memory function.