Simultaneous viewing of own and parasitic eggs is not required for egg rejection by a cuckoo host

Many hosts have evolved diverse cognitive mechanisms to recognize and reduce the cost of social parasitism. For example, great reed warblers Acrocephalus arundinaceus can accurately reject closely mimetic eggs of brood parasitic common cuckoos Cuculus canorus. Yet, these same hosts are less effective at identifying and rejecting parasitism when the clutch is parasitized by multiple cuckoo eggs, suggesting a role for discordancy (the rejection of the egg type in the minority of the clutch) and/or online self-referent phenotype matching (the simultaneous viewing of cuckoo and own eggs in the nest) to reject foreign eggs. We tested whether the presence of host's own eggs is required for the discrimination of foreign eggs by dyeing hosts' own eggs with one of several colors so that clutches contained (a) 1 dyed and 4 unmanipulated eggs, (b) 3 dyed and 2 unmanipulated eggs, or 5 eggs dyed either (c1) differently or (c2) similarly. Rejection rates of dyed eggs varied widely between different colors and were highest in treatment (a), with 1 dyed egg, compared with treatments with the majority (b) or all (c1 and c2) dyed eggs. However, relative rejection rates of dyed eggs were also consistent among specific colors across treatments, including (c1) and (c2), where no unmanipulated own eggs were available for viewing and irrespective of whether eggs were dyed all different colors (c1) or the same colors (c2). We conclude that these hosts can rely on comparisons of foreign egg colors against an internal recognition template of acceptable (own) egg phenotypes.