Egg recognition: The importance of quantifying multiple repeatable features as visual identity signals

Abstract

Brood parasitized and/or colonial birds use egg features as visual identity signals, which allow parents to recognize their own eggs and avoid paying fitness costs of misdirecting their care to others’ offspring. However, the mechanisms of egg recognition and discrimination are poorly understood. Most studies have put their focus on individual abilities to carry out these behavioural tasks, while less attention has been paid to the egg and how its signals may evolve to enhance its identification. We used 92 clutches (460 eggs) of the Eurasian coot Fulica atra to test whether eggs could be correctly classified into their corresponding clutches based only on their external appearance. Using SpotEgg, we characterized the eggs in 27 variables of colour, spottiness, shape and size from calibrated digital images. Then, we used these variables in a supervised machine learning algorithm for multi-class egg classification, where each egg was classified to the best matched clutch out of 92 studied clutches. The best model with all 27 explanatory variables assigned correctly 53.3% (CI = 42.6–63.7%) of eggs of the test-set, greatly exceeding the probability to classify the eggs by chance (1/92, 1.1%). This finding supports the hypothesis that eggs have visual identity signals in their phenotypes. Simplified models with fewer explanatory variables (10 or 15) showed lesser classification ability than full models, suggesting that birds may use multiple traits for egg recognition. Therefore, egg phenotypes should be assessed in their full complexity, including colour, patterning, shape and size. Most important variables for classification were those with the highest intraclutch correlation, demonstrating that individual recognition traits are repeatable. Algorithm classification performance improved by each extra training egg added to the model. Thus, repetition of egg design within a clutch would reinforce signals and would help females to create an internal template for true recognition of their own eggs. In conclusion, our novel approach based on machine learning provided important insights on how signallers broadcast their specific signature cues to enhance their recognisability.

The Role of Intra-Clutch Variation of Magpie Clutches in Foreign Egg Rejection Depends on the Egg Trait Considered

“…Our results confirm a key prediction of the co-evolutionary arms race hypothesis, namely that a low intra-clutch variation favors discrimination and rejection of foreign eggs in magpie for those traits that are more female-specific. Besides, these results are in accordance with a recent and innovative work using machine learning algorithms (as a proxy of a bird observer), which showed that birds may use primarily the most repeatable eggshell features (those with low intra-clutch variation) to recognize their own eggs and reject the foreign ones (Gómez et al, 2021). These findings stresses that the role of intra-clutch variation in rejection is contingent on the particular features of the egg phenotype considered, as some might show a higher degree of environmental determination whereas other are more female-specific.…”

“…In this work we have introduced a novel feature (in the context of foreign egg rejection experimental work) that we hypothesized can influence egg rejection in magpies, the fractal dimension of eggshell spottiness. The previous study that model host recognition used machine learning algorithms and characterized the patterning of spottiness using the FD of spottiness (Gómez et al, 2021), found that, among the several variables related with spottiness, the FD was the most important for recognition and almost the most repeatable within the clutch. Here, we have shown that this feature influences rejection in the same way in all the study years.…”

“…The model analyzing the influence of intra-clutch variation in the FD also included average spottiness as a covariate given that it is known that the FD depends on the degree of spottiness (Pérez-Rodríguez et al, 2013;Gómez et al, 2021). When dealing with color variation, we fitted separated models for the blunt pole, the sharp pole and the middle part to avoid collinearity issues, given that the degree of correlation between measures of intraclutch variation in egg color at the three egg areas were highly positively correlated to each other (Polačiková et al, 2011; see Supplementary Table 1).…”

“…The fractal dimension (FD hereafter) is a parameter that measures complex patterns in objects whose value is influenced by properties that include number, length, turgidity, and connectivity of elements within a given object (Mandelbrot, 1983). FD is scale invariant providing a single value that summarizes the way the pattern "behaves" across scales (Mandelbrot, 1983;Jovani et al, 2013), and it has been used to describe complex color patterns or designs in living organisms, such as the complex color patterns of sexual ornaments in birds (Pérez-Rodríguez et al, 2013;Cantarero et al, 2018), the porosity of avian eggs (Zhang and Wang, 2012) or the homogeneity of spottiness in bird eggs and the amount of protoporphyrin deposited on the eggshell (Gómez et al, 2019(Gómez et al, , 2021.…”

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

“…Yet, female-specific eggshell signatures may be retained in populations with a history of parasitism. Gomez et al [ 48 ] argued that a Polish population of the confamilial Eurasian Coot ( F . atra ) has retained maternally distinctive eggs though the level of CBP is now negligible due to widely dispersed nests.…”