Data from: Colourful network: Pair-bonding temporal dynamics involve sexual signals and impact reproduction
Data files
Jan 08, 2025 version files 113.45 KB
-
MSBE_Indcharacteristic.csv
90.26 KB
-
MSBE_Reproduction.csv
1.90 KB
-
MSBE_Socialinteraction.csv
15.55 KB
-
README.md
5.73 KB
Abstract
Mate choice and pair-bonding are the products of complex decisions involving repeated social interactions over time. They generally rely on multiple sexual signals and behaviours, depend on the phenotype and experiences of the chooser and are affected by environmental conditions and competition for mate access. However, studies investigating the mechanisms of mate selection often ignore the dynamic aspects of the pair formation. In the current study, we used social network analyses to (1) describe sexual interactions over time in multiple groups of captive zebra finches (Taeniopygia guttata), (2) ask how individual sexual signals and body condition relate to pair-bonding temporal dynamics and (3) investigate whether sexual networks influence assortative mating and reproductive performance. We followed sexual interactions of 8 sex-balanced groups of 8 individuals to extract social network metrics over 8 sessions of observation. We assessed individual body condition, sexual signals and couples’ pairing latency, laying date, clutch size and embryo viability. Pair-bonding dynamics were first characterized by song interactions between most individuals, then by numerous allopreening and clumping behaviours, targeting more specific partners. More colourful individuals became engaged in more sexual interactions more rapidly, and such network dynamics explained assortative mating for beak colour. Interestingly, being involved in song interactions was negatively associated with reproductive performances while the reverse was true for being involved in allopreening and clumping interactions. Our study highlights the need to study pair-bonding dynamics to better understand how variations in individual phenotype within sexual networks explain assortative mating and reproductive performance in monogamous species.
README: Data from: Colourful network: Pair-bonding temporal dynamics involve sexual signals and impact reproduction
https://doi.org/10.5061/dryad.0rxwdbsb5
Description of the data and file structure
The data were collected on 64 adult zebra finches (Taeniopygia guttata) in 2022 (between March and October) at the Centre d’Etudes Biologiques de Chizé (CNRS) in France.
Data from the file MSBE_Socialinteraction.csv was collected using a group-level scan sampling method to record the sexual behaviour of all individuals within each group (8 groups of 8 individuals in total). We conducted two 20 min observation sessions per group each day, from the release of the birds until pair formations were observed, or after eight sessions of observation (i.e. 4 days). We noted the ID of birds expressing one of the three types of sexual interactions: male song (i.e. singing directed song toward the partner), allopreening (grooming/preening the feather of the partner) and clumping (i.e. perching in physical contact next to each other) behaviours as well as the absence of interaction.
Data from the file MSBE_Indcharacteristic.csv includes the network metrics we extracted from social network analysis as well as the individual characteristics that were measured before the pair formation experiment. We measured body mass, the tarsus, beak and wing length and transformed the tarsus, the beak and the wing length into a single index of body size (Principal component analysis). As a measure of body condition, we extracted the residuals from the correlation between mass and the body size index. We assessed the beak colour of all males and females and the cheek patch colour of males using 3 photographs of each individual against a neutral white board and extracting the redness, brightness, and chromaticity using the Munsell® Book of Color and the eyedropper tool in Photoshop CC2.
Data from the file MSBE_Reproduction.csv was collected after each pair was formed (4 clumping interactions observed across successive sessions) and placed in a reproductive cage. We monitored nest building daily and noted the laying latency and the clutch size. After 10 days of incubation, we assessed the embryo viability using a Buddy digital egg monitor.
Files and variables
File: MSBE_Indcharacteristic.csv
Description: This file includes all individual characteristics we measured before the experiment and the social network metrics extracted from social network analysis.
Variables
- id: Identity of each individual
- strength: network metric (the sum of the weight of interactions) at the pair level
- degree: network metric (the number of partners involved with) at the individual level
- group: the identity of the group observed
- session: the number of the session of observation
- behaviour: the type of sexual behaviour (song, allopreening, clumping)
- sex: sex of individuals
- cage: identity of the reproductive cage
- PC1beakcol: beak colour score
- cheekhue: male cheek patch redness
- cheekchroma: male cheek patch chromaticity
- cheekbrightness: male cheek patch brightness
- bodysize: body size of individuals
- bodycondindex: body condition index of individuals
File: MSBE_Reproduction.csv
Description: This file includes the data collected during the reproduction.
Variables
- IDres: Identity of each individual
- Sex: Sex of individuals
- Cage: Identity of each reproductive cage
- Aviary: The group identity
- Success: Whether the pair successfully reach the pairing criteria (4 clumping)
- PairingLat: The number of sessions needed to reach this criterion
- LayingLat: The number of days between couple isolation into a reproductive cage and the day of the first egg laid in the nest
- Clutchsize: The number of eggs laid in the nest by the female
- EmbryoSurvival: The number of viable eggs divided by the clutch size
File: MSBE_Socialinteraction.csv
Description: This file included the row data of all sexual interactions we observed between males and females within each group. This table is a row matrix. Each line represents female data and the columns represent male data. Cl: clumping, S:song and Al: allopreening
Variables
- IDFem: Ring number of each female
- IDpointF: Dot number used to identify individual during the session of observation
- DateObs: Date of observations
- HourObs: Hour of observations
- NbAviary: The identity of the group observed
- NbSession: The session number
- Male1Cl: The number of clumping interactions with male 1
- Male1S: The number of song interactions with male 1
- Male1A: The number of allopreening interactions with male 1
- Male1Tot: The total number of interactions with male 1
- Male2Cl: The number of clumping interactions with male 2
- Male2S: The number of song interactions with male 2
- Male2A: The number of allopreening interactions with male 2
- Male2Tot:The total number of interactions with male 2
- Male3Cl: The number of clumping interactions with male 3
- Male3S:The number of song interactions with male 3
- Male3A:The number of allopreening interactions with male 3
- Male3Tot:The total number of interactions with male 3
- Male4Cl: The number of clumping interactions with male 4
- Male4S:The number of song interactions with male 4
- Male4A:The number of allopreening interactions with male 4
- Male4Tot:The total number of interactions with male 4
- Tot_tot:The total number of all interactions
Missing data : NA
Code/software
All analyses were conducted on R Studio (R version 4.0.2; R Development Core Team, 2021). For the social network analysis we used the ANTs package (Sosa et al., 2020).