2Critical & sensitive periods

FThe discovery of the window

The idea began with a goose. The naturalist Konrad Lorenz showed that newly hatched goslings would attach themselves to the first moving object they encountered — their mother, ordinarily, but equally a person if that was who was there — and would then follow it as if it were their parent. This imprinting was quick to form, hard to reverse, and, decisively, possible only during a brief window in the first hours and days of life. Lorenz called that window the critical period, and the term has organised the study of development ever since.[2009]

The deep point is not about geese. It is that the nervous system has scheduled readiness: there are moments when a particular experience must coincide with a particular stage of brain maturation, or the opportunity is lost. The physiology is primed and waiting; if the expected input does not arrive on time, the system does not pause — it moves on, and moves on changed.

FTA song learned on schedule

Birdsong gives the clearest worked example, because a bird's song is both innate and learned. A white-crowned sparrow raised in silence grows up to sing only a crude, abnormal song; one that is deafened sings something more distorted still. But a young isolated bird that is played recordingsof its species' song during the right early weeks will learn to sing normally — and will reproduce the particular dialect it heard. The bird is born with a rough template and must hear the real thing, and itself, at the right time to complete it.[2009]

Two lessons carry straight over to humans. First, heredity sets the boundaries — a bird learns its own species' song far more readily than any other — but experience fills them in. Second, the learning is time-locked: the same recordings outside the window do nothing. Hearing oneself matters too, which is why a bird deafened after learning still sings, while one deafened before never does.

TWindows for human language

Human language obeys the same logic. The rare and tragic cases of children raised without language — through extreme isolation or neglect — show that those who reach adolescence without it rarely acquire normal spoken language thereafter, however intensive the teaching. Even brief early exposure to language seems to be a prerequisite, and the longer and earlier the exposure, the better the eventual outcome.[2009]

Normal development shows the window directly. A newborn can discriminate the speech sounds of every language; over the first year, that universal sensitivity narrows to the sounds of the language actually heard, as experience selects which distinctions are worth keeping. The infant brain is, in effect, tuning itself to its environment on a schedule — and a child deprived of sound during that tuning misses it.[2004]

TCritical versus sensitive periods

Not all windows are equally rigid. A true critical period is a hard window: miss it and the capacity is lost more or less permanently (binocular vision is the classic example). A sensitive period is a softer version — a time when the brain learns most easily and the costs of deprivation are greatest, but outside which some learning is still possible, just slower and less complete. Most human abilities, including auditory and language development, behave like sensitive periods rather than absolute critical ones.[2010]

Underneath the abstraction, a window closing has a cellular meaning. The opening of a sensitive period coincides with the maturation of inhibitory (GABAergic) circuitry, and its progress can be read in the developing balance between excitation and inhibition. In the auditory cortex, inhibition begins life untuned and mismatched to excitation; over roughly the first postnatal month it becomes tuned and calibrated to the local excitatory input, and that emerging excitatory–inhibitory balance depends on experience — patterned sensory input, not the calendar alone, drives it. A window, in other words, is held open by an immature, still-plastic circuit and closed by the inhibitory machinery that eventually stabilises it.[2010]

TDifferent functions, different windows

Crucially, there is not one window but many, each on its own schedule. Binocular vision closes early; the phonology of a first language tunes in the first year; skills like absolute pitch favour early musical exposure; and the central auditory pathway has its own sensitive period. Select a function below to compare them — and notice that every one opens at birth.[2004, 2002]

FTThe window the implant races

For cochlear implantation, the practical message is stark and simple. The central auditory pathway has a sensitive period of roughly the first few years of life; an implant provided within it lets the pathway mature, and one provided long after it closes cannot fully recover the lost ground. This is why universal newborn screening and early implantation are not refinements but the core of paediatric practice — they exist to get the signal to the brain while the window is still open.[2002, 2010]

The next two modules look at why the window closes — what the brain actually does to a pathway that is deprived of its input — starting with the experiments that first revealed it, in the visual system: return to the overview or read on.