Short summary

This study investigates the relationship between host behavioural traits and parasitic infection using a phylogenetic multilevel meta-analytic approach.

Effect sizes were synthesized across observational and experimental studies, accounting for shared controls, species-level non-independence, and host phylogeny.

Analyses include overall meta-analytic models, parasite group–specific models,
tests for publication bias, and sensitivity analyses.

Repository contents

This repository contains information required to reproduce the analyses
reported in the manuscript. The repository is structured to allow users to run
the analyses from raw data to final model outputs using a single script.

Files and folders

Meta-analysis.R

Main R script that performs all steps of the analysis workflow, including data preparation, phylogenetic tree construction using the Open Tree of Life,
computation of variance-covariance matrices, fitting of phylogenetic multilevel meta-analytic models, publication bias analyses, sensitivity analyses, and plots.

data.csv

Dataset containing standardized effect sizes (Zr), sampling variances, study identifiers, host species information, host personality traits, parasite groups, and variables used in the meta-analyses.

Data description:

• MetaID: Identifier of the study, indicating whether it was an observational or experimental study.
• PaperID: Unique identifier assigned to each article.
• Authors: Authors of the article
• Year: Year of publication of the article
• Host: Subphylum of the host species (vertebrate/invertebrate)
• HostPhylum
• HostClass: Host species taxonomic class.
• HostOrder: Host species taxonomic order.
• species_OTL: Species name used for the phylogenetic analysis.
• species: latin name of the species
• behaviour_measured: Behaviour measured in the study, using the terminology reported by the original authors.
• operational_def: Behavioural test or assay used to measure the behaviour.
• behaviour: Standardized behavioural category assigned according to our inclusion criteria (see the main text).
• ParasiteGroup: ecological/functional groups of parasites.
• HostInfo: position of the host in the parasite's life cycle.
• Infection_stat: infection measure.
• n_1: Number of infected individuals.
• n_2: Number of uninfected individuals. If n_2 is missing, n_1 represents the total sample size.
• Mean_1: Mean behavioural value of infected individuals.
• Mean_2: Mean behavioural value of uninfected individuals.
• UpperCI_1: Upper confidence interval of infected individuals.
• UpperCI_2: Upper confidence interval of uninfected individuals.
• LowerCI_1: Lower confidence interval of infected individuals.
• LowerCI_2: Lower confidence interval of uninfected individuals.
• SE_1: Standard error of infected individuals.
• SE_2: Standard error of uninfected individuals.
• SD_1: Standard deviation of infected individuals.
• SD_2: Standard deviation of uninfected individuals.
• beta_coeff: beta coefficient
• z_value: z value
• t_value: t value
• Deeterm_coeff: determination coefficient
• Pearson_corr: Pearson correlation coefficient
• Spearman_corr: Spearman correlation coefficient
• Chi_square: Chi-square value
• PointBiseria_corre: point-biserial correlation coefficient
• df: degrees of freedom
• r: Pearson correlation coefficient calculated by us converting the previous values
• Zr: Fisher's Z-transformed effect size.
• Var_Zr: Variance of the Fisher's Z-transformed effect size.
• Effect_sizeID: Unique identifier assigned to each effect size.
• ControlMean: Mean behavioural value of the uninfected (control) group.
• SharedControls: Adjustment factor used when multiple infected groups were compared with the same uninfected (control) group.

Software and package versions

R version

Analyses were conducted in R (≥ 4.2.0).

R packages

The following R packages are required to run the analyses:

metafor

rotl

ape

geiger

picante

tidyverse

readxl

orchaRd

Rcpp

RRphylo

metaAidR

ggplot2

devtools

remotes

Some packages are installed from GitHub; installation instructions are
provided in commented lines within the analysis script.

Notes

• The code uses relative file paths and should run on any operating system.
• Internet access is required to retrieve phylogenetic information from the

Open Tree of Life.

• Some sections of the script are mutually exclusive (e.g., alternative data subsets)

Models:

General models

General model in observational studies

1. Load the dataset and subset it to include only observational (OBS) studies.
2. Remove Sociability
3. Run the script up to the model section.
4. Run the model "overall_model_OB"

General model in experimental studies

1. Load the dataset and subset it to include only experimental (EXP) studies.
2. Remove Sociability
3. Run the script up to the model section.
4. Run the model "overall_model_ED"

Intensity of infection in observational studies

1. Load the dataset and subset it to include only observational (OBS) studies.
2. Remove Sociability and Aggressiveness
3. Subset the dataset including only "intensity" data
4. Run the script up to the model section.
5. Run the model "overall_model_OB"

Presence of infection in observational studies

1. Load the dataset and subset it to include only observational (OBS) studies.
2. Remove Sociability
3. Subset the dataset including only "presence" data
4. Run the script up to the model section.
5. Run the model "overall_model_OB"

Intensity of infection in experimental studies

1. Load the dataset and subset it to include only observational (OBS) studies.
2. Remove Sociability, Aggressiveness, and Boldness
3. Subset the dataset including only "intensity" data
4. Run the script up to the model section.
5. Run the model "overall_model_ED"

Presence of infection in experimental studies

1. Load the dataset and subset it to include only observational (OBS) studies.
2. Remove Sociability
3. Subset the dataset including only "presence" data
4. Run the script up to the model section.
5. Run the model "overall_model_ED"

Parasite type models

Endoparasite model in observational studies

1. Load the dataset and subset it to include only observational (OBS) studies.
2. Remove Sociability
3. Subset the dataset including only "endoparasite" data
4. Run the script up to the model section
5. Run the model "endo_model_OB"

Endoparasite model in experimental studies

1. Load the dataset and subset it to include only experimental (EXP) studies.
2. Remove Sociability and Aggressiveness
3. Subset the dataset including only "endoparasite" data
4. Run the script up to the model section
5. Run the model "endo_model_ED"

Microparasite model in observational studies

1. Load the dataset and subset it to include only observational (OBS) studies.
2. Remove Sociability and Aggressiveness
3. Subset the dataset including only "microparasite" data
4. Run the script up to the model section
5. Run the model "micro_model_OB"

Microparasite model in experimental studies

1. Load the dataset and subset it to include only experimental (EXP) studies.
2. Remove Sociability and Aggressiveness
3. Subset the dataset including only "microparasite" data
4. Run the script up to the model section
5. Run the model "micro_model_ED"

Ectoparasite model in observational studies

1. Load the dataset and subset it to include only observational (OBS) studies.
2. Remove Sociability, Aggressiveness, and Boldness
3. Subset the dataset including only "ectoparasite" data
4. Run the script up to the model section
5. run the model "ecto_model_OB" or"ecto_model_OB"

Ectoparasite model in experimental studies

1. Load the dataset and subset it to include only experimental (EXP) studies.
2. Remove Sociability, Aggressiveness, and Boldness
3. Subset the dataset including only "microparasite" data
4. Run the script up to the model section
5. Run the model "ecto_model_ED"

Host type model

Host type model in observational and experimental studies

1. Load the dataset
2. Do not subset
3. Run the script up to the model section
4. Run the model "host_model"

Operational definition models

Operational definitions in observational studies

1. Load the dataset and subset it to include only observational (OBS) studies.
2. Subset the data according to what you are going to test (e.g., general model, ectoparasites, endoparasites, microparasites)
3. Run the script up to the model section
4. Run the model "measure_OB"

Operational definitions in experimental studies

1. Load the dataset and subset it to include only experimental (EXP) studies.
2. Subset the data according to what you are going to test (e.g., general model, ectoparasites, endoparasites, microparasites)
3. Run the script up to the model section
4. Run the model "measure_ED"

For publication bias, sensitivity analysis, and plots, follow the script.