How Do Birds Build Nests? Instinct, Learning and Some Very Odd Architecture

The variation is enormous, and each solution trades off effort, concealment and protection:

• •Scrapes, a simple depression in the ground, used by plovers, terns and ostriches, relying entirely on camouflage of the eggs and the sitting bird
• •Cups, the familiar design, built from twigs, grass, moss, mud and spider silk, which binds the structure elastically and lets it expand as chicks grow
• •Cavities, either excavated by woodpeckers or occupied secondhand, offering the best protection from weather and predators, which is why hole-nesting species have higher fledging success
• •Pendant and woven nests, hung from branch tips where climbing predators cannot reach, reaching their extreme in the weavers, whose males tie knots and hitches with strips of grass
• •Mud nests, built by swallows and martins from hundreds or thousands of individually carried pellets, and the edible nests of swiftlets made from saliva that hardens on contact with air
• •Mounds, built by the megapodes of Australia and Southeast Asia, which use decomposition or volcanic heat rather than body warmth to incubate and are the only birds that do not brood their eggs
• •Colonial structures, of which the sociable weaver's is the largest, a permanent thatched apartment block used by over a hundred pairs and occupied continuously for as much as a century

The traditional view was that nest building is innate, supported by the observation that a hand-raised bird with no example to copy still builds a recognisable nest of its species. The recent work complicates this considerably. Studies of captive zebra finches by Susan Healy's group have shown that first-time builders are slower and less efficient than experienced ones, that birds improve measurably with practice, that they show individual preferences for material colour and type that can be changed by experience, and that they watch and copy other birds. Weaver birds, filmed building successive nests, make fewer errors and work faster with each attempt, and individuals develop consistent idiosyncratic habits. The conclusion now accepted is that the general plan is innate and the execution is learned, in much the same way that the capacity for language is innate and a particular language is not.

Birds are selective in ways that have measurable function. Structural material provides rigidity and lining provides insulation, and species adjust the proportion according to climate, with populations of the same species building thicker-walled nests in colder areas. Spider silk is collected deliberately by hummingbirds and long-tailed tits because it is elastic and adhesive, allowing a nest to stretch. Several species incorporate fresh aromatic plants, including starlings, and experiments have shown that these reduce parasite loads in the nest, which raises chick survival. Human material is used opportunistically and with mixed results: cigarette butts have been shown to repel mites because of residual nicotine and also to damage the chicks' cells, and plastic string causes entanglement deaths. Some species decorate rather than build, most famously the bowerbirds, whose elaborate constructions of sticks, berries, shells and blue plastic are not nests at all but display structures built solely to attract a mate.

The obvious answer, holding eggs, is incomplete. A nest is primarily a thermal device, since eggs must be kept within a narrow temperature range and a parent cannot supply all the heat alone, which is why insulation and wall thickness correlate so strongly with climate. It is also a defence, and the choice between concealment and inaccessibility is the main design decision: a ground nest relies on being invisible, a pendant nest on being unreachable, a cavity on being defensible. Some nests are also signals, since studies of several species find that males build more elaborate structures where females can assess them, which introduces sexual selection into what looks like pure engineering. And a great many species do not build at all, including cuckoos and other brood parasites, which lay in other species' nests and let them do the work, and the emperor penguin, which balances its single egg on its feet and covers it with a fold of skin because there is no nesting material anywhere on the Antarctic ice.

Nest building has become a useful model system for questions well beyond ornithology. It is a case of complex structured behaviour produced without a plan, blueprint or teacher, which is directly relevant to robotics and to the study of how apparently designed outcomes emerge from simple rules, and researchers have built machines that assemble structures using the local rules a bird appears to follow. It offers a clean way to study the interaction of genetic predisposition and learning, since the outcome is physical, measurable and produced repeatedly by the same individual. And it functions as an environmental indicator, since timing of nest building has shifted earlier in many species as springs warm, and the materials found in nests document what is in a landscape, which is how the spread of plastic into remote environments has been tracked.

Nests range from bare scrapes to knotted woven pouches, saliva cups, mud pellet structures, compost mounds monitored by beak and hundred-year colonial thatches, and the design trades concealment against inaccessibility. The old view that building is pure instinct has been overturned: birds improve with practice, develop individual preferences and copy each other, so the plan is innate and the execution learned. A nest is primarily a thermal device, and some are also signals used by females to assess a male.