[预告]06.13系统理论进展系列讲座:Quantitative analysis and computational modeling of individual-level interactions in fish with a burst-and-coast swimming mode

　　题目： Quantitative analysis and computational modeling of individual-level interactions in fish with a burst-and-coast swimming mode
 
　　时间： 2016年6月13日周一，14:00
 
　 　地点： 英东教育楼217会议室

　　主讲人： Guy Theraulaz 教授

　　 主讲人简介：

　         Guy Theraulaz ，法国国家科学研究院及图卢兹三大教授。长期从事动物群体和社会中的集群行为背后的行为和认知机制研究，包括蚁群、鱼群、机器人群体等,是该领域顶尖学者之一。目前共发表包括1篇Nature、2篇PNAS等高水平期刊共160多篇，谷歌学术引用次数高达19897次，其中最高单篇引用6658次，H指数为50.

　 　摘要：

　         We use a bottom-up approach to investigate the swimming behavior and individual-level interactions in the Rummy-nose tetra (Hemigrammus rhodostomus). We first conduct experiments to characterize and model the spontaneous behavior of an isolated fish and its reactions to obstacles. We then analyze the behavior of pairs of fish to measure the effects of social interactions on the behavioral responses of individuals. H. rhodostomus has a burst-and-coast swimming mode which combines an acceleration phase with heading changes. The identification of these short events allows us to segment fish trajectories as series of decision points (called “kicks”) in time and space. We use a kick as a proxy to precisely identify the potential neighboring stimuli (e.g. the distance and orientation to an obstacle or to a neighboring fish) that influence the heading changes of fish.We build a model of spontaneous burst-and-coast swimming and interactions with the wall that takes as input the experimental distributions of length and duration between successive kicks coupled with discrete heading changes. All the parameters used in the model are estimated or measured from the experiments with a high degree of confidence. We then measure the consequences of interactions between pairs of fish on their respective orientations.