AP writer Randolph Schmid tells the story of the cowbird protection racket:
To see what would happen, Hoover and Robinson watched where the cowbirds left eggs in warbler nests, and then removed some of them.
They found that 56 percent of the nests where cowbird eggs were removed were later ransacked.
They also found evidence of what they called 'farming' behavior,' in which cowbirds destroyed a nest to force the host bird to build another. The cowbird then synchronized its egg laying with the hosts' 'renest' attempt.
It stands to reason that there would be more to this story -- I mean, warblers can be stupid, but it ought to be incredibly maladaptive to take care of the cowbird hatchlings. Cowbird eggs do tend to look similar to the host species' eggs, but the selection on the hosts to recognize and eject foreign eggs should be very powerful. The part of the story that is missing (and really necessary to make it comprehensible) is that cowbirds leave their eggs together with the host birds' eggs. Nest crashing reduces the payoff for warblers in detecting and ejecting cowbird eggs, because it imposes a high chance of nest and egg loss.
This is a crazy situation for working out optimal strategies. I wonder if there isn't more complexity here -- in the form of polymorphisms in the host species -- than has yet been recognized.