Automatic identification of bird females using egg phenotype

Abstract

Identification of individuals greatly contributes to understanding animal ecology and evolution, and in many cases can only be achieved using expensive and invasive techniques. Advances in computing technology offer alternative cost-effective techniques which are less invasive and can discriminate between individuals based on visual and/or acoustic cues. Here, we employ human assessment and an automatic analytical approach to predict the identity of common cuckoo (Cuculus canorus) females based on the appearance of their eggs. The cuckoo’s secretive brood parasitic strategy makes studying its life history very challenging. Eggs were analysed using calibrated digital photography for quantifying spotting pattern, size and shape, and spectrometry for measuring colour. Cuckoo females were identified from genetic sampling of their nestlings, allowing us to determine the accuracy of human and automatic female assignment. Finally, we used a novel ‘same-different’ approach that uses both genetic and phenotypic information to assign eggs that were not genetically analysed.Our results supported the ‘constant egg-type hypothesis’, showing that individual cuckoo females lay eggs with a relatively constant appearance and that eggs laid by different females differ more than eggs laid by the same female. The accuracy of unsupervised hierarchical clustering was comparable to assessments of experienced human observers. Supervised random forest analysis showed better results, with higher cluster accuracy. Same-different analysis was able to assign 22 of 87 unidentified cuckoo eggs to seven already known females.Our study showed that egg appearance on its own is not sufficient for identification of individual cuckoo females. We therefore advocate genetic analysis to be used for this purpose. However, supervised analytical methods reliably assigned a relatively high number of eggs without genetic data to their mothers which can be used in conjunction with genetic testing as a cost-effective method for increasing sample sizes for eggs where genetic samples could not be obtained.

Interclutch variability in egg characteristics in two species of rail: Is maternal identity encoded in eggshell patterns?

“…In both Common Moorhens and King Rails, a high degree of phenotypic overlap in egg pattern among clutches laid by different females made discrimination of clutches difficult. Eggs laid by different female Common Cuckoos similarly showed a high degree of phenotypic overlap and were also difficult to classify using automatic clustering methods, such as random forest analysis, or human assessment [ 29 ]. Better success in discriminating eggs has been demonstrated in species under strong selection for egg recognition.…”

“…They found that when fewer variables were included correct egg classification decreased, suggesting that multiple eggshell characteristics are used by birds for egg recognition. Due to how the vertebrate visual cortex processes all aspects of egg variation simultaneously, humans may be better at visually discriminating eggs than any program currently available (though opinions on this differ [ 29 , 48 , 51 ]). Nevertheless, we were able to use matching output from NPM in combination with spatial and temporal data collected in the field to identify probable cases of parasitism and returning breeders that were previously undetected.…”

“…NPM and other egg matching programs have been shown to match eggs correctly without considering genetic information [ 10 , 29 ]. While it will be important in the future to validate studies using pattern matching programs to characterize eggs of different species with genetic parentage and pedigree data, image matching techniques offer an alternative minimally invasive sampling method.…”

“…Images are converted to grayscale, and the program characterizes major pigment features on the eggshell surface to match it with the egg it most closely resembles [ 10 , 28 ]. Statistical tests such as principal component analyses, non-metric multidimensional scaling and cluster analysis can then be used to identify differences among differently patterned groups [ 10 , 15 , 29 ].…”

Multicomponent shell traits are consistent with an individual recognition function of the appearance of common murre (Uria aalge) eggs: A biological replication study

“…Spot density was determined for each egg by counting the number of distinct (i.e., human‐visibly disconnected) maculations along the long axis of the egg and dividing the resulting number by the length of that egg. We did not use digital data or other software‐based approaches to analyze eggshell maculation patterns in our study because none of the published spot‐based metrics are based, sensu stricto, on measurement‐driven avian visual‐physiology derived perceptual modeling (unlike coloration analyses, see below), despite their increasing use for such purposes (e.g., Stoddard et al., 2014, Gómez and Liñán‐Cembrano, 2016, Šulc et al, 2020).…”