Animal Behaviour

Data on sniffing, proximity and relatedness of Cape ground squirrels kin discrimination experiments.

The data archived are occupancy data and capture-recapture data collected in a population of Bombina variegata over 9-years period. The file "multistate_data.txt" includes occupancy data. To investigate the influence of patch size and disturbance level on breeding occurrence, we used multistate occupancy models (Nichols et al. 2007, Gimenez et al. 2014). The data from the relevant two years of the breeding survey (2007 and 2008) was compiled in the same one dataset; one site therefore has two annual replicates with two detection occasions (July and August) each. The year was included in the model as a group effect. In the multistate model, we considered three states: a site could be unoccupied for breeding (U), occupied with non-effective reproduction (i.e. presence of eggs and larvae without metamorphosis success: L) or occupied with successful reproduction (i.e. presence of newly metamorphosed individuals: M). The observations (encoded in the txt file) were coded as undetected (0), eggs and/or larvae detected (1) and newly metamorphosed individuals detected (2). The file "data_2007-2008" includes capture-recapture data. The capture histoiries are encoded as following: for an individual captured at t and t–1, a code of 1 or 4 was attributed if it did not change patch and was in patch category s (small) or l (large) respectively, and a code of 2 or 5 was attributed if it did change patch and was in patch category s or l respectively. For an individual not captured at t–1 and captured at t in patch category s or l, a code of 3 or 6 was attributed respectively. An individual not captured at t was given a code of 0. The file "Croix_global_2000-2008_surface" includes capture-recapture data. The capture histoiries are encoded as following: for an individual captured at t and t–1, a code of 1 or 4 was attributed if it did not change patch and was in patch category l (low) or h (high) respectively, and a code of 2 or 5 was attributed if it did change patch and was in patch category l or h respectively. For an individual not captured at t–1 and captured at t in patch category l or h, a code of 3 or 6 was attributed respectively. An individual not captured at t was given a code of 0.

This data set reports nest search times and search accuracy across treatments manipulating visual and olfactory cues associated with nest entrances of the carpenter bee, Xylocopa varipuncta.

DATA: Results from 3 behavioural trials and 2 analsyes of similarities (of preen oil chemical composition). Experiment 1: Conspecific preen oil versus absence of preen oil odour cues We presented subjects with a two-choice test involving same-sex preen oil in one maze arm, and residual solvent only in the other arm. Experiment 2: Opposite-sex versus same-sex conspecific preen oil We presented subjects with a two-choice test involving opposite-sex preen oil in one maze arm and same-sex preen oil in the other arm. Experiment 3: Cowbird preen oil versus absence of preen oil odour cues We presented subjects with a two-choice test involving preen oil from a female cowbird in one maze arm and residual solvent only in the other arm. Preen Oil Analysis: We quantified the relative size of each chromatogram peak identified by GC-FID, retaining for analysis only peaks that comprised ≥ 0.1% of the total chromatogram area. To prevent large peaks from disproportionately influencing distance measures, we normalized the data using the ‘range’ method in the decostand function in the ‘vegan’ package in R. We log x + 1 transformed the normalized dataset then constructed pairwise matrices of Bray-Curtis dissimilarity, which we interpret as chemical distances between samples. To assess the statistical significance of chemical differences between the sexes and between species, we used nonparametric analysis of similarities (ANOSIM) with 10,000 iterations.

An Ipython notebook that illustrates a sequential-sampling model of decision making, which works in one-, two- and multiple-choice contexts. While the model predicts rational behaviors in one- and in two-choice tests, it predicts irrational behavior (decoy effect) in three-choice tests. The notebook is presented as a supplemetary material of the submitted paper "A distributed consensus mechanism of decision making explains economically irrational behaviors"

Two data sets are uploaded to allow replication of analyses in this manuscript. First, a traits file containing date, time, and location of sample collection as well as phenotype data for individual marmots (N=78). Second, we include a file containing regressed and normalized gene expression counts for individual marmots.

Our aim was to evaluate among-individual differences in habitat use of 64 GPS collared female roe deer (Capreolus capreolus). To investigate whether deer differed in their habitat use tactics in relation to their personality type and their annual reproductive success, we assessed the link between individual habitat use patterns (the use of different habitat types, namely woodland, hedgerow, meadow, spring crop and autumn crop) with boldness (measured as the strength of behavioural responsiveness to handling) and annual reproductive success (measured by the presence/absence of fawns at heel during autumn). In the dataset, the Female.Status column corresponds to the reproductive status of a given female (successful vs unsuccessful) and its boldness (bold vs shy).

Data accompanying manuscript "Visual threat signals influence social interactions in a cooperatively breeding fish".

Data sets to accompany Bose et al. Cannibalism of young is related to low paternity and nest takeovers in an intertidal fish. The larger file comprises field data collected from multiple years and sites. The smaller file comprises the nests that we selected for genetic paternity testing. See main text for additional details.

These three data files contain raw data of (1) juvenile female behaviours, (2) adult female behaviours, and (3) components of female (asexual) reproductive fitness. Datasets are organised in long format (i.e., one row per subject).