A critical period is a fixed window in early development. During it, specific experience must occur for a skill or brain system to wire up normally. Miss the window, and that skill can become far harder, or sometimes impossible, to learn later.
Key Takeaways
- Definition: A critical period is a fixed, biologically timed window early in development when an organism must receive specific experience for a skill to develop normally. It shapes hearing, vision, social bonding, and language.
- Imprinting: The concept comes from studies of imprinting, where young birds form an attachment to the first moving object they see, usually the mother, within hours of hatching.
- Neural Basis: High levels of brain plasticity mark a critical period. Maturing inhibitory GABA circuits help trigger its close, while structural changes like myelination stabilize the wiring that remains.
- Sensitive Periods: A “weak” critical period, called a sensitive period, is a stage when outside influences matter more than usual, though experience can still shape development outside it.
- Beyond the Window: Scholars still debate how much skills like a native-sounding accent can improve after the window closes. Large-scale modern research suggests the language window closes gradually through the late teens, not sharply at puberty.
Overview
A critical period is a biologically determined stage of development. During it, an organism is optimally ready to acquire a specific pattern of behavior, and this window will not recur later.
An organism may not receive the right stimulus to learn a skill during its critical period. If so, developing that skill later in life can become difficult or even impossible.
This happens because a range of functional and structural elements prevent passive experiences from eliciting significant changes in the brain (Cisneros-Franco et al., 2020).
The first strong proponent of critical periods was the biologist Charles Stockard (1921), who tested the effects of various chemicals on fish embryos. He credited Dareste with originating the idea 30 years earlier (Scott, 1962).
Stockard’s experiments showed that applying certain chemicals to fish embryos at a specific stage of development produced one-eyed fish. The malformation that resulted depended on exactly when in development the chemical was applied, because each organ system had its own narrow window of vulnerability (Stockard, 1921).
These experiments established that the fastest-growing tissues in an embryo are the most sensitive to changes in conditions, with effects that surface later in development (Scott, 1962).
Meanwhile, psychologist Sigmund Freud tried to explain the origins of neurosis in human patients as a result of early experiences. This implied that infants are particularly sensitive to influences at certain points in their lives.
Lorenz (1935) later showed that critical periods also govern primary social bonds in birds, a process he called imprinting. He saw a clear parallel with the critical periods that shape a developing embryo.
Aim: Lorenz (1935) wanted to know how young geese and ducks come to recognize and follow their own species, and whether this recognition is innate or learned.
Method: He hand-reared goslings so that the first moving object they saw after hatching was Lorenz himself rather than the mother goose, then compared them with normally reared birds.
Results: The goslings followed whichever moving object they met in the first day after hatching, and this attachment proved highly stable. When Lorenz mixed a Lorenz-reared group with a mother-reared group, each gosling still followed its own imprinting object.
Conclusion: A lasting social bond in these birds can form only within a brief, genetically timed window. Later research has softened Lorenz’s claim that the bond is completely irreversible, but the basic finding, that timing is everything, still stands.
Soon thereafter, McGraw (1946) pointed out the existence of critical periods for the optimal learning of motor skills in human infants (Scott, 1962).
Example: Infant-Parent Attachment
Critical and sensitive periods also apply to social development. One example is the infant-parent attachment relationship (Salkind, 2005).
Attachment describes the strong emotional ties between infant and caregiver. This reciprocal relationship develops over the child’s first year, particularly during months six to twelve.
During this attachment period, the infant’s social behavior becomes increasingly focused on the principal caregivers (Salkind, 2005).
The 20th-century English psychiatrist John Bowlby formulated and presented a comprehensive theory of attachment influenced by evolutionary theory.
Bowlby argued that the infant-parent bond exists because it helped infants survive. He placed the sensitive window for forming it between about six months and two-and-a-half years of age, with some risk continuing to age five.
Bowlby’s own later writing moved away from calling this a strict “critical” period. Instead, he used the softer term “sensitive period,” describing a window when a bond becomes strongly, but only relatively, fixed.
This coincides with an infant’s increasing tendency to approach familiar caregivers and to be wary of unfamiliar adults. After this critical period, it is still possible for a first attachment relationship to develop, albeit with greater difficulty (Salkind, 2005).
This has brought into question, in a similar vein to language development, whether there is actually a critical development period for infant-caregiver attachment.
Sources debating this issue typically include cases of infants who did not experience consistent caregiving due to being raised in institutions prior to adoption (Salkind, 2005).
Early research into the critical period of attachment, published in the 1940s, consistently found unusual and maladaptive social behavior in children raised in orphanages. These children showed difficulty forming close relationships and were indiscriminately friendly toward unfamiliar adults (Salkind, 2005).
Later research from the 1990s found that adoptees could still form attachment relationships after their first year of life. They also made developmental progress following adoption.
Nonetheless, these children had an overall increased risk of insecure or maladaptive attachment relationships with their adoptive parents. This evidence supports the notion of a sensitive period, but not a critical period, in the development of first attachment relationships (Salkind, 2005).
Mechanisms for Critical Periods
Both genetics and sensory experiences from outside the body shape the brain as it develops (Knudsen, 2004). However, the developmental stage that an organism is in significantly impacts how much the brain can change based on these experiences.
In scientific terms, the brain’s plasticity changes over the course of a lifespan. The brain is very plastic in the early stages of life before many key connections take root, but less so later.
Early experience is crucial for developing skills like language and music. Both are more challenging to take up in adulthood (Skoe and Kraus, 2013; White et al., 2013; Hartshorne et al., 2018).
As brains mature, the connections in them become more fixed. This shift from a plastic to a more fixed state lets the brain retain complex perceptual, motor, and cognitive processes (Piaget, 1962).
Children’s gestures, for example, help predict how they will acquire oral language skills (Colonnesi et al., 2010), which in turn are important for developing executive functions (Marcovitch and Zelazo, 2009).
However, this formation of stable connections can limit how the brain’s neural circuitry can be revised in the future.
If a young organism has abnormal sensory experiences during the critical period, such as auditory or visual deprivation, the brain may not wire itself to process future sensory inputs properly (Gallagher et al., 2020).
One illustration is the timing of cochlear implants, a prosthesis that restores hearing in some deaf people. Children who receive them before age two benefit more than those implanted later (Kral and Eggermont, 2007; Gallagher et al., 2020).
Cataracts in infants cause a comparable pattern of visual deprivation. Removing them during early infancy allows relatively normal vision to develop. Leaving them untreated until adulthood, however, results in substantially poorer vision (Martins Rosa et al., 2013).
After the critical period closes, abnormal sensory experiences have a less drastic effect on the brain. Barring direct damage to the nervous system, any changes that do occur tend to be reversible (Gallagher et al., 2020).
Much of what scientists know about critical periods derives from animal studies, since these give researchers far greater control over the variables being tested.
This research has found that different sensory systems, such as vision, auditory processing, and spatial hearing, have different critical periods (Gallagher et al., 2020).
The brain regulates when a critical period opens and closes largely through inhibitory neurotransmitters, chemical messengers that carry signals between neurons.
The window opens as inhibitory circuits that release GABA, particularly fast-spiking cells called parvalbumin-expressing interneurons, mature and start to sharpen the cortex’s response to patterned input.
Closure then comes from a set of structural “brakes.” Axons lose some capacity to grow, myelination stabilizes existing connections, and dense structures called perineuronal nets form around inhibitory neurons (Gallagher et al., 2020).
Because these brakes are actively built rather than simply lost, they can, in principle, be loosened again later in life, opening the door to future treatments.
Vision: The Ocular-Dominance Critical Period
The visual system provides the clearest evidence for a strict critical period in the brain. Two classic experiments on kittens, by neuroscientists David Hubel and Torsten Wiesel, mapped out exactly when and how this window operates.
Wiesel and Hubel (1963): Monocular Deprivation
Aim: Wiesel and Hubel (1963) wanted to see what happens to the brain’s visual cortex when one eye is deprived of normal input early in life.
Method: They sewed one eyelid shut in newborn kittens for several weeks while the other eye saw normally, then recorded which eye individual cells in the visual cortex responded to.
Results: Most cortical cells no longer responded to the deprived eye. The open eye had captured cortical territory that would normally serve both eyes.
Conclusion: Normal binocular vision is not simply hard-wired. It must be built and maintained through balanced visual experience during early development.
Hubel and Wiesel (1970): Mapping the Critical Window
Aim: Hubel and Wiesel (1970) then set out to define exactly when, and for how long, the visual cortex stays vulnerable to this kind of deprivation.
Method: They deprived kittens of vision in one eye for different lengths of time, starting at different ages, and compared the results with deprivation carried out in fully grown adult cats.
Results: Vulnerability began around three weeks of age, peaked around the fourth to fifth week, and was largely gone by three months.
During the peak, even a few days of deprivation produced a marked shift toward the open eye, while the same deprivation in an adult cat did almost nothing.
Conclusion: The visual cortex has a sharply timed critical period with a clear onset, peak, and end. Outside this window, the same manipulation that devastates a kitten’s vision barely affects an adult’s.
These two studies are among the most influential in developmental neuroscience, earning Hubel and Wiesel a share of a Nobel Prize. Together, they map directly onto human eye conditions.
A congenital cataract removed in infancy allows near-normal vision to develop, but the same cataract left untreated into adulthood leaves vision permanently impaired.
Untreated amblyopia (“lazy eye”) and childhood strabismus can likewise cause permanent loss of visual acuity if not corrected early. Because the effects compound with age, early eye exams remain the single most effective way to protect a child’s vision.
Critical Periods vs Sensitive Periods
Critical periods are similar to sensitive periods, and scholars have, at times, used them interchangeably. However, they describe distinct but overlapping developmental processes.
A sensitive period is a developmental stage where sensory experiences have a greater-than-usual impact on behavioral and brain development. This influence, however, is not exclusive to that time period (Knudsen, 2004; Gallagher, 2020). These sensitive periods are important for skills such as learning a language or instrument.
In contrast, a critical period is a special type of sensitive period, a window where sensory experience is necessary to shape the neural circuits involved in basic sensory processing. When this window opens and closes is well-defined (Gallagher, 2020).
Researchers also refer to sensitive periods as weak critical periods (Gallagher et al., 2020). Examples include:
- Strong critical periods: vision and hearing.
- Weak critical periods (sensitive periods): phoneme tuning (how children learn to organize a language’s sounds), grammar processing, vocabulary acquisition, musical training, and sports training.
Critical Period Hypothesis
One of the most notable applications of the concept of a critical period is in linguistics. Scholars usually trace the origins of the debate around age in language acquisition to Penfield and Roberts’s (2014) book Speech and Brain Mechanisms.
In the 1950s and 1960s, Penfield was a staunch advocate of early immersion education (Kroll and De Groot, 2009). It was Lenneberg who coined the term “critical period” in his 1967 book, Biological Foundations of Language.
Lenneberg (1967) described a critical period as “a period of automatic acquisition from mere exposure” that “seems to disappear after this age.” Scovel (1969) later summarized and narrowed Penfield’s and Lenneberg’s view on the critical period hypothesis into three main claims:
- Accent detection: Adult native speakers can identify non-natives by their accents immediately and accurately.
- Plasticity loss: The loss of brain plasticity at about the age of puberty accounts for the emergence of foreign accents.
- Speech-only scope: The critical period hypothesis only holds for speech (whether or not someone has a native accent) and does not affect other areas of linguistic competence.
Linguists have since asked whether scientific evidence actually supports the critical period hypothesis. Open questions include whether there is a critical period for accentless speech or for grammatical (“morphosyntactic”) competence, and how any age-related differences work at the neurological level (Scovel, 2000).
The critical period hypothesis is central to theories of language acquisition, applying to both first- and second-language learning. Early research on first-language acquisition drew heavily on cases of so-called “feral” children deprived of normal language input during childhood.
Aim: The case of “Genie” offered a real-world test of the critical period hypothesis: could she learn a first language if her exposure began only after the proposed window had closed?
Method: Genie had been deprived of nearly all social and linguistic contact from around 20 months of age until she was discovered at 13 years old.
Curtiss (1977) and colleagues studied her language development over several years of rehabilitation.
Results: Genie built a sizeable vocabulary and used language to communicate, but she never mastered grammar. Her speech stayed largely telegraphic, with lasting problems in word order and tense.
Conclusion: Genie’s case fits a closing window for grammar, though her extreme social isolation, malnutrition, and abuse mean age of exposure cannot be separated from these other harms. Her case is best read as suggestive, not conclusive, evidence.
More recent research has focused systematically on deaf children born to hearing parents, who are deprived of language input until at least elementary school.
These studies isolate the effect of missing language input without the extreme social deprivation seen in feral-child cases. The older a child is when first exposed to sign language, the worse their ultimate grammatical attainment (Emmorey, Bellugi, Friederici, and Horn, 1995; Kroll and De Groot, 2009).
Kroll and De Groot argue, however, that the critical period hypothesis does not apply to the rate of language learning. Adults and adolescents can learn a language at the same rate as children, or even faster, in the initial stage of acquisition (Slavoff and Johnson, 1995).
However, adults tend to have a more limited ultimate attainment of language ability (Kroll and De Groot, 2009).
There is a long lineage of empirical findings on age of acquisition. The most fundamental research comes from studies since the late 1970s. These consistently document a negative correlation between age of acquisition and ultimate language mastery (Kroll and De Groot, 2009).
Different periods of infancy bring sensitivity to different aspects of language. Even newborns can tell speech sounds apart categorically.
One- and four-month-olds notice a sound change that crosses an adult phoneme boundary, such as between “ba” and “pa,” but not an equally large change that stays within one category (Eimas et al., 1971).
Aim: Werker and Tees (1984) tested how the ability to tell apart foreign speech sounds changes across a baby’s first year.
Method: English-learning infants aged 6-8, 8-10, and 10-12 months were tested on consonant sounds that do not occur in English, using a head-turn procedure that rewarded them for noticing a change.
Results: At 6-8 months, infants told the foreign sounds apart easily. By 10-12 months, most no longer could, though babies raised on those languages still discriminated them easily.
Conclusion: Over the first year, perception narrows to fit the sounds of a baby’s own language. Because this narrowing can be partly reversed with training, it counts as a sensitive period rather than a hard-and-fast critical one.
Vocabulary learning experiences rapid growth at about 18 months of age (Kuhl, 2010).
Critical Evaluation
More than any other area of applied linguistics, the critical period hypothesis has impacted how teachers teach languages. Consequently, researchers have critiqued how important the critical period is to language learning.
Several early studies found that children were not necessarily superior to older learners at acquiring a second language. This held true even for pronunciation (Olson and Samuels, 1973; Snow and Hoefnagel-Hohle, 1978; Scovel, 2000).
In fact, most researchers at the time were skeptical that a critical period existed at all, and some denied it outright.
Several documented cases run counter to one of Scovel’s (1969) primary claims: adults who acquired a second language and still spoke with a native accent.
Moyer (1999) found that at least one highly proficient English-speaking learner of German was judged to have native-like pronunciation. Bongaerts (1999) similarly found that several highly proficient Dutch speakers of French spoke with accents judged to be native (Scovel, 2000).
Contemporary Research
Recent large-scale research has sharpened, rather than overturned, the classical picture of critical periods.
How Late Can the Grammar Window Stay Open?
Aim: Hartshorne, Tenenbaum, and Pinker (2018) measured, with far more statistical power than earlier studies, how age of first exposure affects the grammar someone eventually learns.
Method: They analyzed an online English grammar quiz completed by roughly 669,000 native and non-native speakers, using statistical modeling to separate the effects of age of first exposure, years of exposure, and current age.
Results: Grammar-learning ability stayed high through childhood and adolescence, then declined sharply around age 17 to 18. Reaching near-native grammar, however, required starting immersion by about age 10 to 12, well before that decline began.
Conclusion: There is a real, bounded window for learning grammar, but it closes gradually in late adolescence rather than snapping shut at puberty as Lenneberg proposed.
Can a Closed Critical Period Reopen?
Aim: Pizzorusso and colleagues (2002) tested whether structures called perineuronal nets, which condense around inhibitory neurons after the critical period, actively hold the visual window shut in adulthood.
Method: In adult rats, well past the normal visual critical period, they used an enzyme to dissolve perineuronal nets in the visual cortex, then deprived one eye and measured whether ocular dominance shifted.
Results: Once the nets were degraded, depriving one eye produced the same juvenile-like shift in ocular dominance normally seen only in young animals.
Conclusion: Closure of a critical period is not just plasticity fading away. It is an active molecular “brake,” and lifting that brake can partly reopen the window, a finding with real promise for treating conditions like adult amblyopia.
References
Bongaerts, T. (1999). Ultimate attainment in L2 pronunciation: The case of very advanced late L2 learners. Second language acquisition and the critical period hypothesis, 133-159.
Cisneros-Franco, J. M., Voss, P., Thomas, M. E., & de Villers-Sidani, E. (2020). Critical periods of brain development. In Handbook of Clinical Neurology (Vol. 173, pp. 75-88). Elsevier.
Colonnesi, C., Stams, G. J. J., Koster, I., & Noom, M. J. (2010). The relation between pointing and language development: A meta-analysis. Developmental Review, 30(4), 352-366.
Curtiss, S. (1977). Genie: A psycholinguistic study of a modern-day “wild child.” Academic Press.
Eimas, P. D., Siqueland, E. R., Jusczyk, P., & Vigorito, J. (1971). Speech perception in infants. Science, 171(3968), 303-306.
Emmorey, K., Bellugi, U., Friederici, A., & Horn, P. (1995). Effects of age of acquisition on grammatical sensitivity: Evidence from on-line and off-line tasks. Applied Psycholinguistics, 16(1), 1-23.
Knudsen, E. I. (2004). Sensitive periods in the development of the brain and behavior. Journal of cognitive neuroscience, 16(8), 1412-1425.
Hartshorne, J. K., Tenenbaum, J. B., & Pinker, S. (2018). A critical period for second language acquisition: Evidence from 2/3 million English speakers. Cognition, 177, 263-277.
Hubel, D. H., & Wiesel, T. N. (1970). The period of susceptibility to the physiological effects of unilateral eye closure in kittens. The Journal of Physiology, 206(2), 419-436.
Kral, A., & Eggermont, J. J. (2007). What’s to lose and what’s to learn: development under auditory deprivation, cochlear implants and limits of cortical plasticity. Brain Research Reviews, 56(1), 259-269.
Kroll, J. F., & De Groot, A. M. (Eds.). (2009). Handbook of bilingualism: Psycholinguistic approaches. Oxford University Press.
Kuhl, P. K. (2010). Brain mechanisms in early language acquisition. Neuron, 67(5), 713-727.
Lenneberg, E. H. (1967). The biological foundations of language. Hospital Practice, 2(12), 59-67.
Lorenz, K. (1935). Der kumpan in der umwelt des vogels. Journal für Ornithologie, 83(2), 137-213.
Marcovitch, S., & Zelazo, P. D. (2009). A hierarchical competing systems model of the emergence and early development of executive function. Developmental science, 12(1), 1-18.
McClelland, J. L., Thomas, A. G., McCandliss, B. D., & Fiez, J. A. (1999). Understanding failures of learning: Hebbian learning, competition for representational space, and some preliminary experimental data. Progress in brain research, 121, 75-80.
McGraw, M. B. (1946). Maturation of behavior. In Manual of child psychology. (pp. 332-369). John Wiley & Sons Inc.
Moyer, A. (1999). Ultimate attainment in L2 phonology: The critical factors of age, motivation, and instruction. Studies in second language acquisition, 21(1), 81-108.
Gallagher, A., Bulteau, C., Cohen, D., & Michaud, J. L. (2020). Neurocognitive Development: Normative Development. Elsevier.
Olson, L. L., & Jay Samuels, S. (1973). The relationship between age and accuracy of foreign language pronunciation. The Journal of Educational Research, 66(6), 263-268.
Penfield, W., & Roberts, L. (2014). Speech and brain mechanisms. Princeton University Press.
Piaget, J. (1962). The stages of the intellectual development of the child. Bulletin of the Menninger Clinic, 26(3), 120.
Pizzorusso, T., Medini, P., Berardi, N., Chierzi, S., Fawcett, J. W., & Maffei, L. (2002). Reactivation of ocular dominance plasticity in the adult visual cortex. Science, 298(5596), 1248-1251.
Rosa, A. M., Silva, M. F., Ferreira, S., Murta, J., & Castelo-Branco, M. (2013). Plasticity in the human visual cortex: an ophthalmology-based perspective. BioMed Research International, 2013.
Salkind, N. J. (Ed.). (2005). Encyclopedia of human development. Sage Publications.
Scott, J. P. (1962). Critical periods in behavioral development. Science, 138(3544), 949-958.
Scovel, T. (1969). Foreign accents, language acquisition, and cerebral dominance 1. Language learning, 19(3‐4), 245-253.
Scovel, T. (2000). A critical review of the critical period research. Annual review of applied linguistics, 20, 213-223.
Skoe, E., & Kraus, N. (2013). Musical training heightens auditory brainstem function during sensitive periods in development. Frontiers in Psychology, 4, 622.
Slavoff, G. R., & Johnson, J. S. (1995). The effects of age on the rate of learning a second language. Studies in Second Language Acquisition, 17(1), 1-16.
Snow, C. E., & Hoefnagel-Höhle, M. (1978). The critical period for language acquisition: Evidence from second language learning. Child development, 1114-1128.
Stockard, C. R. (1921). Developmental rate and structural expression: an experimental study of twins,‘double monsters’ and single deformities, and the interaction among embryonic organs during their origin and development. American Journal of Anatomy, 28(2), 115-277.
Werker, J. F., & Tees, R. C. (1984). Cross-language speech perception: Evidence for perceptual reorganization during the first year of life. Infant Behavior and Development, 7(1), 49-63.
White, E. J., Hutka, S. A., Williams, L. J., & Moreno, S. (2013). Learning, neural plasticity and sensitive periods: implications for language acquisition, music training and transfer across the lifespan. Frontiers in systems neuroscience, 7, 90.
Wiesel, T. N., & Hubel, D. H. (1963). Single-cell responses in striate cortex of kittens deprived of vision in one eye. Journal of Neurophysiology, 26(6), 1003-1017.
