- A red squirrel in Kluane, Yukon, one of the individuals in the study population.
For my Master’s project, I was interested in red squirrel territorial behavior and the vocalizations, known as rattles, used to defend their territories. Red squirrel rattles are individually unique and have been shown to be used to discriminate kin, though the mechanism underlying this ability is unknown. In a recently published paper in Behavioral Ecology, I sought to distinguish between the mechanisms of ‘prior association’, where animals learn the phenotypes of kin they associate with early in life, and ‘phenotype matching’, where animals use a template to match phenotypes, thereby allowing them to recognize kin without an association early in life. I recorded rattles from squirrels in the field, and used those recordings in playback trials to measure the behavioural responses of squirrels to rattles from familiar kin, unfamiliar kin, and non-kin. One of the major benefits of the Kluane Red Squirrel project is that there is pedigree information for each squirrel, which means that we know who their mother and father is and who their siblings are. Without this information, this project would not have been possible, and full pedigree information is difficult to obtain for wild populations of animals.
- Recording rattles from squirrels in the field to use in the playback trials
For red squirrels, familiar kin consisted of pair of squirrels that shared a natal nest (e.g. mother-offspring pairs and siblings from the same litter), and unfamiliar kin consisted of pairs of squirrels that did not share a natal nest (e.g. father-offspring pairs, siblings from different litters). Initial analyses revealed that red squirrels did not discriminate between familiar and unfamiliar kin, but also did not discriminate between kin and non-kin, despite previous evidence indicating this capability. Post-hoc analyses showed that a squirrel’s propensity to rattle in response to playback depended on an interaction between relatedness and how the playback stimuli had been recorded. Rattles used as the playback stimuli were either recorded from squirrels as they moved freely around their territories (unsolicited), or from squirrels as they were released from a trap or in response to a broadcast rattle (provoked). Red squirrels discriminated between rattles from close kin (relatedness coefficient of at least 0.5) and rattles from less related kin or non-kin (relatedness coefficient of less than 0.5) when the rattles were recorded from provoked squirrels. Squirrels did not exhibit kin discrimination in response to rattles that had been recorded from unprovoked squirrels.
This figure show the probability of a rattle response from the subject squirrel during the playback period by relatedness coefficient calculated from the pedigree and the collection method of obtaining the rattle stimulus. Unsolicited rattles were recorded from squirrels moving freely around their territories (n = 67 trials), and provoked rattles (n = 38 trials) were recorded from squirrels as they emerged from a live-trap or from squirrels responding to a rattle playback
This is potentially quite interesting, but it is important to note that this relationship was identified through exploratory post hoc analyses and needs to be tested more rigorously. If these results are robust, however, they would suggest that a squirrel’s physiological state might influence the structure of its rattles, including those individually distinctive structural features that are presumably used in discrimination. This raises interesting questions about what kind of information may be contained in the rattles and suggests that rattles may reflect the current state of stress or aggressiveness of the squirrel.
Photos and post by Julia Shonfield
Julia Shonfield, Jamieson C. Gorrell, David W. Coltman, Stan Boutin, Murray M. Humphries, David R. Wilson, Andrew G. McAdam. 2016. Using playback of territorial calls to investigate mechanisms of kin discrimination in red squirrels. Behavioral Ecology arw165. doi: 10.1093/beheco/arw165.
The abstract and a link to the full text can be found here:
If you are unable to access the article but are interested in reading it, you can email me at julia.shonfield@gmail.com and I can provide you with a copy.