Data for: Personality, plasticity and predictability in sticklebacks: bold fish are less plastic and more predictable than shy fish

SUPPLEMENTARY DATA FOR: Jolles, Briggs, Araya-Ajoy, & Boogert (2019) Personality, plasticity and predictability in sticklebacks: bold fish are less plastic and more predictable than shy fish. SUMMARY: In this study we tested 80 individual wild-caught three-spined sticklebacks (Gasterosteus aculeatus) repeatedly over a 10-week period in a classic boldness assay to investigate the links between the personality, plasticity and predictability. Fish were tested in random order in one of eight replicate testing boxes under controlled laboratory conditions. Using a mixed-modelling approach we demonstrate that individual variation in boldness, i.e. the tendency to trade off risks versus potential rewards, is fundamentally linked to temporal plasticity and predictability: bold fish were less plastic in their behaviour and more predictable than shy fish. Statistical simulations using a novel statistical framework (‘SQUiD’) show that our experimental dataset has sufficient statistical power to provide accurate and precise statistical parameter estimates. DATA: All raw data of the boldness trials is provided in the file boldnessdat.csv and the code for running the final mixed model of the experimental dataset before the testing break and subsequently setting up and running statistical simulations using the SQuID framework cna be found in the file squidsimulations.r. id = identity of the fish. testwk = week of testing relative to start of experimental period. session = sequence number of testing trial for that individual over the testing period. trial = sequence number of testing trial during the day. box = testing box in which the fish was tested. prop.out = the proportion of time a fish spent out of cover in the boldness assay. id.group = identity of the group a fish was housed in over the testing break between weeks 6 and 10.

配套补充数据：Jolles、Briggs、Araya-Ajoy与Boogert（2019）《棘鱼的个性、可塑性与可预测性：大胆个体比羞怯个体可塑性更低且可预测性更强》
摘要：本研究于10周周期内，针对80尾野外捕获的三棘刺鱼（Gasterosteus aculeatus）反复开展经典大胆性行为测验，旨在探究个性、可塑性与可预测性之间的关联。实验在受控实验室条件下，于8个重复测试箱中按随机顺序对鱼类进行测试。本研究采用混合效应建模方法，证实大胆性的个体差异（即权衡风险与潜在收益的倾向）与时间可塑性及可预测性存在本质关联：大胆个体的行为可塑性更低，且相较于羞怯个体具有更强的可预测性。通过新型统计框架‘SQUiD’开展的统计模拟显示，本实验数据集具备足够的统计效力，可获得准确且精准的统计参数估计值。
数据集：所有大胆性行为实验的原始数据均存储于文件boldnessdat.csv中；用于运行实验数据集在测试中断前的最终混合效应模型、后续基于SQuID框架搭建并运行统计模拟的代码，可在文件squidsimulations.r中获取。字段说明：id为鱼类个体标识；testwk为相对于实验周期起始的测试周数；session为该个体在整个测试周期内的测试试次序列编号；trial为当日测试试次的序列编号；box为鱼类接受测试的测试箱编号；prop.out为大胆性实验中鱼类离开遮蔽物的时间占比；id.group为第6周至第10周测试中断期间鱼类所属饲养组的标识。