Host alarm calls attract the unwanted attention of the brood parasitic common cuckoo

It is well known that avian brood parasites lay their eggs in the nests of other bird species, called hosts. It remains less clear, however, just how parasites are able to recognize their hosts and identify the exact location of the appropriate nests to lay their eggs in. While previous studies attributed high importance to visual signals in finding the hosts’ nests (e.g. nest building activity or the distance and direct sight of the nest from vantage points used by the brood parasites), the role of host acoustic signals during the nest searching stage has been largely neglected. We present experimental evidence that both female and male common cuckoos Cuculus canorus pay attention to their host’s, the great reed warbler’s Acrocephalus arundinaceus alarm calls, relative to the calls of an unparasitized species used as controls. Parallel to this, we found no difference between the visibility of parasitized and unparasitized nests during drone flights, but great reed warblers that alarmed more frequently experienced higher rates of parasitism. We conclude that alarm calls might be advantageous for the hosts when used against enemies or for alerting conspecifics, but can act in a detrimental manner by providing important nest location cues for eavesdropping brood parasites. Our results suggest that host alarm calls may constitute a suitable trait on which cuckoo nestlings can imprint on to recognize their primary host species later in life. Our study contributes to the growing body of knowledge regarding the context-dependency of animal signals, by providing a novel example of a beneficial acoustic trait intercepted by a heterospecific and used against the emitter.


Abstract
It is well known that avian brood parasites lay their eggs in the nests of other bird species, called hosts. It remains less clear, however, just how parasites are able to recognize their hosts and identify the exact location of the appropriate nests to lay their egg in. While previous studies attributed high importance to visual signals in finding the hosts' nests (e.g. nest building activity or the distance and direct sight of the nest from vantage points used by the brood parasites), the role of host acoustic signals during the nest searching stage has been largely neglected. We present experimental evidence that both female and 35 common cuckoos Cuculus canorus pay attention to their host's, the great reed warbler Acrocephalus arundinaceus alarm calls, relative to the calls of an unparasitized species used as controls. Parallel to this, we found no difference between the visibility of parasitized and unparasitized nests during drone flights, but great reed warblers that alarmed more frequently experienced higher rates of parasitism. We conclude that alarm calls might be advantageous for the hosts when used against enemies or for alerting conspecifics, but can act in a detrimental manner by providing important nest location cues for eavesdropping brood parasites. Our results suggest that host alarm calls may constitute a suitable trait on which cuckoo nestlings can imprint on to recognize their primary host species later in life. Our study contributes to the growing body of knowledge regarding the contextdependency of animal signals, by providing a novel example of a beneficial acoustic trait intercepted by a heterospecific and used against the emitter.

This supplementary material contains:
- Table S1: Results of the full model showing all of the predictors measured of the probability of brood parasitism of great reed warbler Acrocephalus arundinaceus nests by common cuckoos Cuculus canorus.
- Table S2: Contingency table with the responses of female and male common cuckoos Cuculus canorus in the 2 minutes prior and 2 minutes during the mobbing display and alarm calls of their host.
- Table S3: Contingency table with the number of individual female and male common cuckoos Cuculus canorus which approached the playback device playing their host' alarm call or the call of the collared dove Streptopelia decaocto, used as a control.
- Table S4: Measured nest site characteristics and their predicted effect on brood parasitism (as described in Moskát and Honza 2000).
-data regarding the correlative study, experiments and drone flights in csv formats -R markdown script of the data analyses in html format Table S1. Results of the full model showing all the potential predictors of the probability of brood parasitism of great reed warbler nests by common cuckoos. Values for the fixed factor, 'alarm call', indicate the difference of alarming hosts compared to non-alarming great reed warblers, 'laying stage' is a two-level factor of nests in their first 3 days of egg laying or second 3 days of egg laying, 'time of nest visit' is a scaled continuous variable which measures the number of elapsed minutes since midnight standardized with a Z transformation (mean = 0 and SD = 1).

439.700
Observer identity 0.000 Table S2. Responses of female and male common cuckoos in the 2 minutes prior and during the 2 minutes of alarm calls and mobbing display of their host, the great reed warbler. The response of the cuckoos was scored on a four-level scale, as follows: 0 -no reaction; 1-characteristic female bubbling call or male call on a perch closer than 100 m; 2 -flight towards the great reed warbler nest; 3 -flight towards the nest coupled with characteristic female or male call.

Sex
Group 0 1 2 3 females before playback 8 5 1 0 during playback 4 1 2 7 males before playback 17 9 3 1 during playback 5 8 4 13 Table S3. Number individual female and male common cuckoos which approached the playback device playing their host' alarm call (experimental trials; n = 16) and the sound of the collared dove (control trials; n = 16). The response of the cuckoos was considered positive if they approached the playback device compared to the distance they were initially observed, and neutral if the focal birds did not react or increased their distance (i.e. flew away) from the playback device compared to their initial position.  Table S4. Measured nest site characteristics and their predicted effect on brood parasitism (as described in Moskát and Honza 2000) of non-parasitized (n = 8) and parasitized (n = 8) great reed warbler nests. Nest volume was calculated as the volume of a cylinder with of the height of the nest (in cm) and base diameter equal to the width of the nest (in cm).