Dr. Jim Stone from the University of Sheffield’s department of psychology has done research to explain why birds seem to attain or “learn” the skill of flight so quickly. The reason is that the ability in the individual does not constitute “original learning.” Such as, for example, me learning to play the guitar. A skill for which I have no genetic memory.
Birds are born with latent memories for flight formed by the actions of successive previous generations. This latent memory for flight is genetically inherited and stored in neural networks (in other words, distributed over many neuronal connections). Evolution, over time, adapts as necessary the information that the bird is born with. So, learning is involved in an evolutionary sense combined with the inherited ability for complex skills (such as nest building or flight) that are refined by the individual. The benefit to flying as an instinct refined is that the bird spends less time learning and therefore less time vulnerable to predators.
A little more on this:
From "Avian Growth and Development: Evolution Withing the Altricial Precocial Spectrum" edited by J. Matthias Starck, Robert E. Ricklefs,
"Early sensory stimulation and motor practice are not necessary for the development of locomotor patterns. This holds true both for motor patterns that are present in their final form soon after hatching and those, such as flying, that emerge in the course of ontogeny. With regard to the latter, Provine (1981) convincingly demonstrated that immobilization of young domestic chicks by elastic bandages from day one until just before testing did not hamper the normal devleopment of wing-flapping and flight. Furthermore, the bandaged individuals achieved flight distances equal to those of unbandaged contol birds. Similar experiments, with the same results, have been reported for the altricial domestic dove (Grohmann 1939)."
From “Aspects and prospects of the concept of instinct” by A. Kortlandt (Zoological Laboratory, University of Amsterdam),
"According to experiments made by Spalding (1873) and Grohmann (1938) the improvement in flying abilities observed in young birds after fledging should be ascribed to neural maturation. According to experiments and other data in Cormorants, however, this improvement in flying abilities must be ascribed largely to a special type of learning, called flash-learning, which may be considered the analogon on the motor side of behavior, as compared to what, on the sensory side of behavior, is usually called imprinting….”
From “Organic Foundations of Animal Behavior” by Joseph Altman,
“Is the delayed appearance of flying in birds attributable to delayed neuromuscular maturation, or is it explained by a functional organization largely dependent on learning? Spalding (1873 [1954]) investigated this question by hand-rearing swallows in small cages. He found that once the deprived swallows were released they flew as efficiently as normal animals."