Data for Identifying influential neighbors in animal flocking

Schooling fish are an impressive group-level coordination in which multiple individualsmove together in a seamless way. This is possible because each individual in the groupresponds to the movement of other group members. But how many individuals doeseach fish pay attention to? Which are the influential neighbors? We need to answerthese questions if we want to understand how directional information propagates acrossa group. Our research shows that in the rummy-nose tetras species there is a limitednumber of influential neighbors which are not necessarily the closest ones.<br>The data set here contains all the data used in our research. There are three compressed zip files for experimental data of group of 2 fish, experimental data of group of 5 fish and null model data of group of 5 fish, respectively.<br>To use the data so as to interpret our research, let's take the group of 2 fish for an example.<br>First, unzip the file "UturnPositionData_2fish.zip" into a folder, then you'll find 1133 RData files. Which can group into 3 categories: position_uturns_fishNo_2_uturn_i.RData with i=1,2,...,1135; uturn_info_fishNo=2.Rdata and uturn_bad_fishNo=2.Rdata.<br>Second, now you can load the data with software R.You'll find that all the data are in dataframe. The position_uturns_fishNo_2_uturn_i.RData has the trajectory information around the U-turn events, while the uturn_info_fishNo=2.Rdata and uturn_bad_fishNo=2.Rdata contain the beginning and ending time information for the good U-turn events and the U-turn event IDs for the bad ones, respectively. The bad U-turn events are those with too much errors from the video tracking procedure thus we omitted in our analysis.<br>Third, then you can follow the methods in our manuscript and reproduce our results.