Figure 4

Schematic presentation of the experimental design. In step I, we created plot- and species-specific apparent symbol preferences during breeding period in year t. In step II, we randomised symbols on boxes after the breeding seasons in year t and year t + 1. In step III, the symbol preferences of resident tits and flycatchers in the experimental patches were monitored during the nest-site selection period. After each choice the symbol was removed from the box to exclude social information from the current spring for the subsequent birds. The proportions of the three symbol types during the spring were kept equal by switching symbols of empty nest boxes if needed.