When parents play favourites: Brood demand shapes parental preference for offspring UV colour
Data files
Apr 25, 2023 version files 89.37 KB
-
1._DRYAD_Nests.xlsx
25.50 KB
-
2._DRYAD_Nestlings.xlsx
61.27 KB
-
README.md
2.60 KB
Abstract
Parents might initially produce more offspring than they might be able to raise. However, when offspring demand exceeds their parents´ rearing capacity, parents might shift care towards the offspring that yield greater fitness returns to achieve their optimal brood size via brood reduction. Such favoritism could rely on offspring signaling traits if these inform parents about offspring quality and hence about the pay-offs of their investment. Here we investigated whether favoritism of blue tit (Cyanistes caeruleus) parents for an offspring signal (i.e., ultraviolet (UV) plumage coloration) varies with brood demand. To test this, we experimentally blocked the UV reflectance of yellow breast feathers in half of the nestlings of each brood, and then we sequentially performed two opposing brood size manipulations to vary nestling demand below or above parental rearing capacity. In reduced broods, nestlings begged overall less intensely and gained more body mass, supporting that parental rearing capacities sufficed to satisfy brood demand. Moreover, in reduced broods, UV-blocked nestlings (i.e., low-quality offspring) were fed and prey-tested more often. Yet, they begged more than control nestlings, suggesting that they were perhaps treated differently by other family members or that they may exploit parental preferences beyond actual need (at least in reduced nests). Parents flexibly shifted their feeding rate and favoritism in response to short-term changes in family size, as there was no parental preference in enlarged broods. Such flexible parental feeding rules may allow parents to gain the upper hand in parent-offspring conflict. However, we did not find evidence that parental favoritism facilitated brood reduction, at least in conditions were demand was temporally enhanced.
Methods
Statistical analyses (from the published article)
All analyses were performed with SAS 9.4. (SAS Inst., Cary, NC, USA), using the mean values per treatment in each nest. To analyze feeding rates, we performed a generalized linear mixed model with Poisson distribution, while the mean number of nestlings that were prey-tested and mean begging intensity were analyzed with mixed models with normal distribution. The mean number of nestlings that were prey-tested was “log10 (x+1)” transformed to fulfil the assumption of normality. For the three behavioral variables, we had four data per nest corresponding to non-UV-blocked and UV-blocked chicks in both the enlarged and the reduced brood size treatment. All models included dyad ID and nest ID (nested within dyad) as random factors, and two random slopes (nest ID*UV treatment and nest ID*brood-size treatment). As predictor variables, we included the UV treatment (non-UV-blocked/UV-blocked), brood-size manipulation (enlarged/reduced), the interaction between both treatments, the day of the video recording (day 12/13), the original brood size and the Julian hatching date. We also included the interactions between UV treatment and all other predictors, which were never significant (see Results). We did not include the interactions between brood size manipulation and the rest of variables to avoid over-parameterization, and because the brood size was manipulated in order to observe the effect of UV color on behavioral parameters.
To analyze the individual body mass change, we used a mixed model with normal distribution, which included dyad ID and nest ID (nested within dyad ID) as random factors, and the same two random slopes as above. Moreover, we included a new categorical variable (foster nestling yes/no) as a random factor to control for whether each nestling had been exchanged or not. We included the same predictor variables as in previous models, as well as nestling sex and all the interactions between UV treatment and predictor variables.