Data and code for: Boldness shapes skill refinement in a novel foraging task in wild squirrels

Boldness shapes skill refinement in a novel foraging task in wild squirrels Folder: Code contains R script to fully reproduce analysis, create tables and figures Folder: Data Note that for each file there are two versions - one with an inclusion criterion of 10 visits and/or 10 solves, respectively, which was used for the main analysis. The second set has an inclusion criterion of 5 visits and/or 5 solves (all files have the addition _min5) in the file name. Since colummn names in both are the same, we will list them here only once: Visits data: each row represents one distinct to a puzzle box Subject: Individual identifier for each squirrel Cumulative_visits: # of cumulative visits to any puzzle box Solves_per_visit: How many solves did the squirrel do during the given visit? Latency_to_first_solve: Latency between arrival and the first solve during a vist (given a solve occurred). In s Consumption: Did the squirrel visibly consume seed during the visit Total_time_present: visit duration in s age: A for adult, P for juveniles sex: F for female, M for male con_pres: 0 for no other squirrel present upon arrival, 1 for conspecific(s) present at the box upon arrival log_cumulative_visits: log # of cumulative visits Combined data: each row represents one solve Subject: Individual identifier for each squirrel lever_side: which lever was pushed for solving (right or left) reaction: behavioral reaction after solving (exit means leaving the antenna area; flinch, no reaction or alert mean they stayed in the antenna area). Is changed into exit or stay for modeling purposes cumulative_count = # of cumulative solves per individual colony = which population (C or P). Note that they are pooled for this analysis due to limited sample size body_part = which body part was used for solving (left_paw, right_paw, both_paws). Is modified into contralateral paw, ipsilateral paw and both paws for modeling purposes (last column) log_cuulative_count = log # of cumulative individual solves box_between_solves = how many times did the squirrel (unsuccessfully) interact with the puzzle box (sniffing, touching) in between solves paw_comb = which paw did they use in relation to the lever (ipsi = left paw on left lever, or right paw on right lever; contra = righ paw on left lever or left paw on right lever; both = both paws on same side of lever) con_pres: 0 for no other squirrel present during a solve, 1 for conspecific(s) present at the box during a solve Subject_data: Individual co-variates Subject: Unqiue identifier for each squirrel n_solves: number of total solves produced during the trial phase n_locations: how many puzzle box locations did they visit (out of 6 possible) age: A for adult, P for juvenile sex: F for female, M for male trainig_solves: how many solves did they produce during the training phase (as a proxy for task exposure during training) logtrainig_solve: log number of the above num_trapped: total number of times trapped during summer trap_rate_per_day: average number of times an individual was trapped on a trapping day (used to calculate boldness) any_beh_prop: proportion of trapping occasions during which a squirrel showed a fear response to an approaching human observer when trapped (struggle, chatter, alarm call). Used to compute boldness score Folder: Tables output Model summaries (as presented in SI) Folder: Figures output Figures as presented in the MS and SI (fully reproducible with the provided R code) Folder: model output contains R Data objects of all five models for each of the two inclusion criteria (10 and 5)