Why do birds of a feather flock together?

This blog post is provided by Shannon Buckley Luepold and tells the #StoryBehindthePaper for the paper “Habitat detection, habitat choice copying, or mating benefits: what drives conspecific attraction in a nomadic songbird?“, which was recently published in Journal of Animal Ecology. The authors explored why wood warblers settle near each other, comparing hypotheses about other birds acting as a signal of good habitat, simply copying other birds, or gaining mating benefits from settling near others.
Conspecific attraction and why it matters:

The tendency for animals to settle in locations where other members of their species are already present (“conspecific attraction”) has long intrigued behavioural ecologists. This behaviour has also drawn the attention of conservation biologists, who have explored how conspecific attraction might facilitate settlement in desired locations. Despite longstanding interest in this topic, however, there is still much uncertainty about why attraction occurs. Importantly, the more we can learn about why animals choose to settle near other individuals, the better equipped we are to understand why they might not make this choice when we desire or expect them to.

Study species:

In our study, we investigated conspecific attraction during breeding habitat selection in the wood warbler Phylloscopus sibilatrix. Wood warblers are migratory songbirds that breed in the temperate forests of Europe and spend much of the non-breeding season in sub-Saharan Africa. Weighing in at only 10 grams, these lemon-colored birds sing a silvery trill song that has been likened to the sound of a spinning coin on a marble slab. Natural history accounts often describe wood warblers forming “territory clusters” during the breeding season, and birds have been attracted to settle near speakers broadcasting conspecific song in both Switzerland and Poland. We studied a population of wood warblers breeding in the Swiss Jura Mountains.

Singing male wood warbler (Photo credit: Sandro Carlotti).
Our hypotheses:

We tested three hypotheses for why conspecific attraction occurs in male wood warblers. Two of these were based on social information (i.e. information that is gained from observing or interacting with other individuals). According to the habitat detection hypothesis, the presence of others at location X informs individuals looking for somewhere to settle that the habitat at location X might be good, and is certainly worth checking out. Whether attracted individuals choose to remain at this exact location or not, however, depends on their own assessment of the habitat. Thus, this hypothesis predicts that within a forest, spatial variation in habitat will be more important than conspecific locations as a predictor of male settlement patterns. Alternatively, the habitat choice copying hypothesis suggests that settling individuals are simply copying (i.e., imitating) the choices of others, which, by definition, means they do not make (or heed) their own assessment of the habitat. Thus, this hypothesis predicts that conspecific locations will be more important than spatial variation in habitat as a predictor of male settlement patterns.

The third hypothesis that we tested was the female preference hypothesis, which suggests that males derive mating benefits from being in a group. According to the female preference hypothesis, females prefer males that are aggregated because this facilitates comparison of potential mates. Thus, this hypothesis predicts a positive relationship between male pairing success and male aggregation.

Reduced nest predation is another common hypothesis for territory aggregation that is based on group benefits, but we considered it less relevant for wood warblers because predation often happens in contexts when group defense behaviours do not typically occur (e.g., at night, by relatively large mammals).

Our methods:

To determine the relative importance of spatial variation in conspecific presence and habitat features, we first manipulated the locations where social information was available by broadcasting wood warbler songs from an array of speakers in several forests. We did this for two breeding seasons (unluckily for our field team carrying car batteries for the playback stations, wood warblers have a habit of choosing the steepest hills for breeding sites…).

Intrepid field assistants carrying song playback sets on a snowy April day (Photo credit: Florian Moser).

We then used point pattern analyses to assess whether speaker locations or spatial variation in habitat features was the stronger predictor of male settlement patterns.

We used an unconventional approach to represent spatial variation in habitat features. Given that female wood warblers prospect for nest sites and choose them with little regard for male territory boundaries, we assume their choices reflect the presence of habitat features they find attractive. Therefore, we used kernel density estimation to create maps of spatial variation in “habitat attractiveness,” which were based on all known wood warbler nest locations within a forest (including up to 10 years of data). Of course, when making the habitat attractiveness maps for predicting male settlement in year t, we did not include nests from year t because male and female locations within the same year are not independent. With this approach we avoided making assumptions about which specific habitat features birds were cueing in on, and instead based our definition of attractive habitat on patterns in the choices of the birds themselves.

To address the female preference hypothesis, we determined the spatial arrangement of males that were available to each settling female. Then we determined the “connectivity” of each male in a female’s set, a metric from metapopulation theory which we used to represent the number and proximity of neighboring males. With this information, we could model whether males that were more aggregated (i.e., with higher connectivity) were more successful in gaining a mate.

Female wood warbler in nest (Photo credit: Michael Gerber).
What we found:

We found that spatial variation in habitat was more important than conspecific (speaker) locations as a predictor of male settlement patterns, and we found that being aggregated did not increase a male’s success in attracting a mate. Thus, our data were most consistent with the habitat detection hypothesis, suggesting that singing conspecifics simply alert other individuals to the presence of potential habitat. The ultimate benefit of using social information in this way could be reduced search time, which may be especially relevant for species like the wood warbler that move breeding sites both within and between years.

Wood warbler habitat in the Swiss Jura Mountains (Photo credit: Lukas Linder).
Going forward:

The results of our study highlight the hierarchical process behind settlement decisions and underscore the contingent nature of conspecific attraction. We encourage future research on these topics to carefully distinguish making the same choice as others from copying the choice of others. When studying clustered distributions, it may be useful to consider the processes by which animals detect and assess habitat in addition to explanations based on benefits that accrue from aggregation per se.

Author bio

Shannon Luepold recently received her PhD from the University of Zürich and is currently a postdoctoral researcher at the Swiss Ornithological Institute. She is broadly interested in the behavioural ecology and natural history of birds, but is particularly fascinated by social interactions and vocal behaviour.

Read the paper

Read the full paper here: Luepold, S. B., Kokko, H., Grendelmeier, A., & Pasinelli, G. (2022). Habitat detection, habitat choice copying or mating benefits: What drives conspecific attraction in a nomadic songbird? Journal of Animal Ecology, 00, 1– 12. https://doi.org/10.1111/1365-2656.13844

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s