A topic in the Open Knowledge Graph — a free, open map of 15,290 topics and the order to learn them in.
A critical period is a time-limited window during which specific experiences are necessary for normal development; without them, the relevant capacity fails to develop properly and cannot be fully recovered later. A sensitive period is a broader window when experience has especially strong effects but is not strictly necessary. Classic examples include Lorenz's imprinting in birds, the visual cortex's dependence on patterned input in early infancy, and first-language acquisition. Understanding these windows reveals both the importance of early enrichment and the limits of 'catch-up' development.
Compare case studies: Genie (language deprivation), children with congenital cataracts restored after infancy, and cross-cultural adoption studies. Distinguish which abilities show hard critical periods vs. softer sensitive periods.
To understand critical and sensitive periods, start with what you already know about neuroplasticity — the brain's capacity to reorganize itself in response to experience. Plasticity is not constant across the lifespan. It peaks in early development and then narrows, and the narrowing is not uniform: different systems have different windows. A critical period is the extreme end of this narrowing — a time window so important that, without the right input during it, the system simply fails to wire up correctly and the deficit becomes largely permanent. A sensitive period is a softer version: experience during it has outsized developmental impact, but the window is not a hard cutoff.
The clearest illustration comes from the visual system. Kittens raised with one eye sutured shut during the first weeks of life never develop normal binocular vision — even if the eye is opened years later, the cortex has already allocated that input territory to the open eye. This is not about the eye itself but about how experience guides neural circuit formation. Hubel and Wiesel's work established that the visual cortex requires patterned visual input during the critical period to develop normally. Children born with cataracts face the same risk: if optical correction is delayed past the visual critical period, normal acuity in that eye cannot be fully recovered. The same logic applies to Lorenz's imprinting — newly hatched ducks will follow and bond to whatever moving object they see during the first hours of life. Miss that window, and the normal attachment template never forms.
Language acquisition sits in a different category. Children acquire first languages effortlessly before puberty and then with increasing difficulty afterward — this is the sensitive period for language acquisition. The case of Genie, a child kept in near-total isolation until age 13, illustrates the limits: though she learned many words, she never achieved normal grammar. However, language is not a single critical period for all components — phonological distinctions (hearing the difference between sounds like /r/ and /l/ in a non-native language) have an earlier and steeper cutoff than vocabulary. Adults can still acquire language, but they work harder for less native-like results. This gradient is the hallmark of a sensitive period rather than a strict critical period.
The conceptual distinction that trips students up is this: critical periods are about the necessity of input, not just its usefulness. During a critical period, the relevant neural circuits are actively pruned and consolidated based on incoming experience — this is the "use it or lose it" logic. If the input doesn't arrive, the circuits responsible for processing it may be repurposed or pruned away. The nature-nurture framing you've encountered before is sharpened here: what matters is not nature *versus* nurture but the precise *timing* of nurture. The same experience can have dramatically different effects depending on whether it arrives during the sensitive window or after. This reframes early childhood enrichment not as a vague good but as a period of particular biological leverage.