Intimate Iguanas: social opportunities and mate preference improve breeding success in Caribbean iguanas
Description
Effects of socialization on reproductive success and breeding behavior All analyses were performed using R studio (2022) using R version 4.0.1 GUI 1.72 Catalina. To evaluate whether socialization affected reproductive performance, we used the lme4 package in (Bates et al. 2015). We ran GLMMs to examine the relationship between our socialization explanatory variable (Howdy Door 1,0) and the reproductive performance variables Intromission Success, Average Number of Copulations, Average Copulation Time, Egg Production, and Average Number of Eggs. Socialization information for the focal pairing was matched to the reproductive success and breeding season mate introductions behavior data collected for that same year. We included male ID and female ID as random factors to account for repeated animals across years and variation in reproductive success across animals. Guassian regression was used for the Number of Copulations, Average Number of Copulations, and Average Copulation Time. A poisson distribution was used for Average Number of Eggs. Logistic regression with a logit-link function was used for intromission success and egg production. Effects of mate preference on reproductive success and breeding behavior To evaluate whether mate preference affected reproductive performance, we used the lme4 package in R (Bates et al. 2015). We ran GLMMs to examine the relationship between both the male and female preference explanatory variable (Preference 1,0) and the mutual mate preference explanatory variable (Preferred, Mixed, Nonpreferred) versus the reproductive performance variables Intromission Success, Average Number of Copulations, Average Copulation Time, Egg Production, and Average Number of Eggs. Preference information for the focal pairing was matched to the reproductive success and breeding season mate introductions behavior data collected for that same year. We included male ID and female ID as random factors to account for repeated animals across years and variation in reproductive success across animals. Guassian regression was used for the Number of Copulations, Average Number of Copulations, and Average Copulation Time. A poisson distribution was used for Average Number of Eggs. Logistic regression with a logit-link function was used for intromission success and egg production. Similar to our analyses on socialization, we used the summed frequency behavioral categories: Social Display, Olfactory Communication, Feed, Basking, and Resting. We also evaluated the individual behavior Proximity. Behavioral data was subsetted to the breeding season during mate introductions as we were most interested in determining how socialization affected breeding behaviors during mate introductions. We ran GLMMs assuming a negative distribution using the glm.nb function from the MASS package (Venables & Ripley 2002) to examine the relationship between mate preference designations and our summed behaviors from the breeding season.
Files
Steps to reproduce
Behavioral observations and study protocol We used a newly defined behavioral ethogram to score observational periods defined below. We grouped behaviors based on a combination of weightings from a Principal Component Analysis and the experience of expert’s familiar with iguana breeding and behavior (gray rows in Supplemental Table 1). We summed frequencies across each observation of the following behavioral categories (gray rows in Supplemental Table 1): Conspecific Investigation, Breeding Behavior, Nesting Behavior, Olfactory Communication Behavior, Contact Aggression, Social Display Behavior, Feed, Rest, Basking Behavior, Keeper Interactions, Not Visible, and Other. Iguana behavior was scored manually for frequency through visual inspection of the taped recordings. We performed behavioral observations across four distinct trial periods as follows: 1. Pre-socialization mate choice: One female housed between two males and scored for a solid day with all howdy doors open on both sides of the enclosure during the pre-breeding season (September - November). 2. Howdy Socialization: Females were allowed access to one male through the howdy door prior to the breeding season (December - April) while the other neighboring male had a blocked howdy door. Behavioral observations were recorded from 1300-1400 hours daily in 2018 and 1300-1400 hours on Mondays, Wednesdays, and Fridays in 2019. Behaviors were averaged across all days for analysis. 3. Post-socialization mate choice: One female between two males scored for a solid day starting at 0700 hours with both howdy doors open in the last week of April. 4. Breeding Season Pairing (May - June): One female and one male allowed continuous access for breeding (see detailed description below). Behavioral observations were recorded from 1300-1400 hours daily in 2018 and 1300-1400 on Mondays, Wednesdays, and Fridays in 2019. Behaviors were averaged across all days for analysis. We designated preference in both the pre-socialization and post-socialization dichotomous choice tests by obtaining the percentage of behaviors females performed toward each neighboring male out of total behaviors performed during the trial for the following behavioral categories: Conspecific Investigation, Breeding Behavior, Nesting Behavior, Olfactory Communication Behavior, Social Display Behavior, and Basking Behavior. Males and females were designated as preferred if the focal animal performed more than 50% of their behaviors toward that male/female and the other male/female was designated as non-preferred. Pairs were additionally scored as mutually preferred if both the male and female had chosen each other as preferred and as mixed preference if either the male or female did not prefer the opposite sex conspecific and mutually nonpreferred if neither preferred the opposite sex