Cognitive Connie
Understanding language acquisition & thought
Language acquisition is the most debated topic in cognitive science: how does a child go from non-verbal infant to fluent speaker of any human language in under five years, with no formal instruction? And once acquired, does language constrain or merely express thought? This theme covers the developmental trajectory from babbling to grammar, the theoretical positions (Chomsky vs Skinner vs Tomasello vs statistical learners), the critical period evidence, and what psycholinguistic research tells us about language's influence on cognition.
01. Overview
Theories of language acquisition
The acquisition debate is not just empirical — it is a contest between fundamentally different views of the human mind. Positions range from rich innateness (language knowledge is pre-specified) through interactionist (input and cognition cooperate) to emergentist (everything from domain-general learning).
Chomsky's Universal Grammar
Children are born with Universal Grammar — innate constraints on possible grammars. The Language Acquisition Device is triggered by linguistic input and sets parameters (e.g., head-final vs head-initial) to produce the target grammar.
Tomasello's Usage-Based Theory
Children acquire language through joint attention and intention reading in social interaction. Language is built bottom-up from item-specific constructions, abstracted into general patterns through analogy — no innate grammar required.
Statistical / Connectionist
General-purpose pattern-detection mechanisms extract linguistic structure from statistical regularities in the input. No innate grammar needed — only sufficient input and a powerful learning algorithm (PDP networks, transitional probability tracking).
Key figures
B. F. Skinner
1904–1990Proposed a behaviourist account of language in *Verbal Behavior* (1957) — language as operant behaviour shaped by reinforcement; largely refuted by Chomsky (1959).
Eric Lenneberg
1921–1975Proposed the critical period hypothesis for language acquisition in *Biological Foundations of Language* (1967); evidence from feral children, deaf individuals, and second-language learners.
Noam Chomsky
1928–Proposed Universal Grammar and the Language Acquisition Device; his 1959 review of Skinner demolished behaviourist language theory; remains the most influential figure in formal linguistics.
Michael Tomasello
1950–Developed the usage-based, social-pragmatic theory of language acquisition; emphasises joint attention and intention-reading over innate grammar; leading alternative to Chomskyan nativism.
Jenny Saffran
1969–Demonstrated statistical learning in 8-month-old infants (Saffran, Aslin & Newport, 1996) — infants use transitional probabilities to segment words from speech after just 2 minutes.
Lera Boroditsky
1975–Leading researcher on linguistic relativity; demonstrated effects of language on time cognition (2001), spatial reference, and number; advocates for real but moderate Whorfian effects.
Key concepts
Language Acquisition Device (LAD)
Chomsky's theoretical construct for the innate language-learning capacity that constrains possible grammars and enables rapid acquisition from impoverished input.
Universal Grammar (UG)
The set of innate constraints on possible human languages — proposed by Chomsky to explain why all languages share abstract structural properties (e.g., structure-dependence, recursion).
Poverty of the stimulus
Children acquire grammatical knowledge (e.g., structure-dependence) that is not consistently present in their input, suggesting innate grammatical constraints rather than purely data-driven learning.
Critical period
A developmental window during which language acquisition is maximally efficient; after puberty, native-like proficiency in a second language becomes very difficult (Johnson & Newport, 1989).
Fast mapping
Children's ability to form an initial word-to-concept link from minimal exposure — a single brief encounter with a new word (Carey & Bartlett, 1978); underlies the vocabulary spurt around 18 months.
Overregularisation
Applying a regular grammatical rule to irregular forms ("goed", "mouses") — evidence of rule extraction rather than rote learning; produces U-shaped development (correct → error → correct).
Statistical learning
Sensitivity to transitional probabilities between syllables in the speech stream, enabling word segmentation without explicit instruction (Saffran et al., 1996).
Infant-directed speech (IDS)
The speech register used by caregivers with infants — higher pitch, exaggerated intonation, simpler syntax, slower tempo — which attracts infant attention and facilitates language learning.
Linguistic relativity
The weak Sapir-Whorf hypothesis: language influences (but does not determine) cognitive processes such as colour discrimination (Winawer et al., 2007) and temporal reasoning (Boroditsky, 2001).
Sapir-Whorf hypothesis
The claim that language shapes thought; exists in a strong form (linguistic determinism — refuted) and a weak form (linguistic relativity — supported with caveats).
Inner speech
Vygotsky's term for internalized, condensed, predicative self-directed speech that develops from social external speech through the intermediate stage of private (audible) speech.
Home sign
Structured gestural communication systems spontaneously developed by deaf children who are not exposed to conventional sign language, showing an innate drive to impose linguistic structure on communication.
Test your knowledge
Frequently asked questions
What is the strongest evidence for the critical period in language acquisition?+
Several converging lines of evidence support the critical period. Johnson & Newport (1989) showed that Chinese and Korean immigrants to the US who arrived before age 7 achieved native-like English grammar scores; those arriving after puberty performed substantially worse with no further age-of-arrival effect. Feral children — notably Genie, discovered at 13 with no language exposure — never acquired normal syntax despite intensive teaching, though some vocabulary development occurred. Cochlear implant studies show that children implanted before 2 years achieve near-normal language outcomes, while those implanted after 5 show progressively worse outcomes for some language features. Finally, second-language acquisition data consistently show that phonological and morphological competence declines with age of first exposure, with puberty as a critical threshold. The critical period likely reflects the interaction of neural plasticity (which declines with myelination, synaptic pruning, and loss of cortical reorganisability) and the statistical learning advantages of the pre-pubescent brain.
Does language really shape thought, or does it merely describe it?+
The evidence supports a middle position: language influences some cognitive processes in real time without determining underlying conceptual capacities. The key evidence comes from studies with verbal interference controls. Winawer et al. (2007) showed Russian speakers have a colour discrimination advantage at the Russian blue boundary (goluboy/siniy) — but this advantage disappears when a verbal task occupies phonological working memory, showing it is produced online by language activation. Gilbert et al. (2006) showed categorical colour advantages are stronger in the right visual field (which projects to the left, language-dominant hemisphere), and are eliminated in the left visual field — again implicating active linguistic mediation. Against determinism: the Dani (Rosch, 1972) remember focal colours well despite lacking terms for them, showing that perceptual categories precede and constrain lexical encoding. The current consensus: language labels are fed back into perceptual and conceptual processing (Lupyan & Clark, 2015), making cognition more categorical and verbally organised when language is active — but not in a way that permanently determines or limits underlying thought.
Sources
Last reviewed July 2025- 1.
Chomsky, N. (1965). Aspects of the theory of syntax. MIT Press.
+About this source
The foundational statement of Universal Grammar and the innateness hypothesis for language.
- 2.
Tomasello, M. (2003). Constructing a language: A usage-based theory of language acquisition. Harvard University Press.
+About this source
The leading alternative to nativism — construction grammar grounded in joint attention and intention-reading.
- 3.
Saffran, J. R., Aslin, R. N., & Newport, E. L. (1996). Statistical learning by 8-month-old infants. Science, 274(5294), 1926–1928.
+About this source
Demonstrated that infants use transitional probabilities to segment words from fluent speech after just 2 minutes.
- 4.
Vygotsky, L. S. (1986). Thought and language (rev. ed., A. Kozulin, Trans.). MIT Press.
+About this source
Proposed that thought and language are initially independent but become intertwined, with inner speech emerging from social language.