Mechanisms for Language Learning

PSYC 216 Child Psychology | Spring 2026

Mechanisms for Language Learning

How Do Children Create Infinite Sentences?

  • All theories recognize contributions from both nature and nurture but emphasize one more than the other:

    • NATURE:

    • Nativist perspective emphasizing innate language abilities and mechanisms.

    • NURTURE:

    • Empiricist perspective focusing on properties of the environment, including language experience, input, and usage.

Part 1: Does Timing Matter?

Age of Arrival Predicts English Grammar Skills: Johnson & Newport (1989)

  • Participants: Korean & Chinese adults who learned English as a second language.

  • Key Variable: Age of arrival in the U.S.

  • Task: Grammaticality judgment test.

  • Result: Earlier arrival = higher grammar scores.

    • Grammaticality Scores by Age of Arrival:

    • Native: 280

    • Ages 3–7: 270

    • Ages 8–10: 260

    • Ages 11–15: 250

    • Ages 17–39: 240

    • (Graphical representation of scores from ages 3-39).

More Recent Evidence: Hartshorne, Tenenbaum, & Pinker (2018)

  • Participants: Approximately 669,000 English speakers tested online.

  • Groups: Monolinguals, immersion learners, & non-immersion learners.

  • Key Findings:

    • Earlier first exposure = higher grammar accuracy.

    • Immersion learners who started before age 10 performed near-natively.

    • Grammar-learning ability holds steady through childhood and begins to decline around age 17.

    • Learning Rate Change with Age:

    • Graph depicting learning rates for monolingual & immersion learners versus non-immersion learners.

Is It Age That Matters — Or Prior Language Experience?

  • Possible Explanations for Age Effects in Second-Language Learning:

    • To isolate age, we need subjects who had no early language access.

    • Interference & Entrenchment: Older learners' first language may be more established, leading to difficulties in adapting to a new grammar.

    • Maturational Effects: Brain plasticity for language declines with age, making language acquisition more challenging for older learners.

A Natural Experiment in First-Language Acquisition

  • Studying Variation in Age of First Language Exposure:

    • Subjects: Deaf children of hearing parents, many born to parents with little or no sign language knowledge.

    • Education debates (oral vs. manual) historically influenced early exposure, creating variation.

Age Effects on Acquiring First Language: Newport & Supalla (1990)

  • Participants: Deaf adults with over 30 years of daily ASL use.

    • Learner categories:

    • Native learners: exposed from birth

    • Early learners: first exposed at ages 4-6

    • Late learners: first exposed after age 12

  • Task: Video description in ASL.

  • Results:

    • Earlier first exposure = higher accuracy on complex grammar.

    • No age differences for simple grammar; all groups performed near perfectly.

    • Test Score on Complex Grammar (ASL Morphology):

    • Native learners: Highest score

    • Early learners: Slightly lower

    • Late learners: Lowest score.

Summary of Timing Influence

  • Conclusion: Timing matters; there is a sensitive period — or possibly several — for language learning.Part 2: Does Input Quality Matter?

Parental Speech Predicts Children's Language Development

  • Key Findings:

    • Parental speech strongly influences children's vocabulary growth and word recognition (Hoff, 2003, 2013; Rowe & Goldin-Meadow, 2009; Schneidman & Goldin-Meadow, 2012; Weisleder & Fernald, 2013).

    • More child-directed talk leads to:

    • Longer, more complex sentences.

    • Richer vocabulary.

    • Positive Outcomes:

    • Larger vocabularies → Faster word recognition → Greater school readiness → Stronger literacy skills.

Huttenlocher et al. (2002)

  • Participants: 4–5-year-olds.

  • Results: Children produce more complex sentences when:

    • Their mothers use complex sentences frequently.

    • Their preschool teachers also use complex sentences frequently.

Romeo et al. (2018)

  • Participants: 36 children ages 4–6, with at-home audio recorders capturing naturalistic speech.

  • Key Variable: Number of conversational turns per hour between child and adult.

  • Key Findings:

    • More conversational turns lead to higher verbal scores.

    • Children with more conversational turns showed greater activation in Broca's area during language processing.

Part 3: What If There's No Input?

Children as Language Inventors
Children Invent Their Own Language: Home Sign

  • Research by:

    • Susan Goldin-Meadow & colleagues (Goldin-Meadow & Feldman, 1977; Brentari & Goldin-Meadow, 2017).

  • Observations: Deaf children of hearing parents often face linguistic isolation with limited access to shared language.

  • Outcome: Children spontaneously create gestural communication systems known as Home Sign, reflecting early stages of spoken language development.

How Researchers Studied Home Sign

  • Participants: Children who are profoundly deaf with no prior exposure to conventional sign language.

  • Methods:

    • Longitudinal videotaped interactions with parents and researchers.

    • Gestures coded for form and meaning with context-based interpretations.

Home Sign Mirrors Early Language Development

  • Key Points:

    • All children gesture; first signs appear at the same age as the first words in hearing children.

    • Two sign types:

    • Pointing (reference).

    • Characterizing signs (action/object categories).

    • Emergence of multi-sign combinations and consistent gesture order reflects early word combinations and conveys meaning.

Home Signs Are Abstract, Not Simple Pantomimes

  • Features:

    • Pointing and characterizing signs are abstract.

    • Characterizing signs stand for categories of actions across various objects and contexts.

Home Signs Can Refer to Things Not Present

  • Displacement:

    • Ability to refer to non-present entities is found in all human languages.

    • Home signers invent this capacity spontaneously.

Multi-Sign Expresses Complex Meanings

  • Example Sentence:

    • "Swordfish poke-in-chest dead long nose swim".

Linguistic Structure Emerges From Children, Not Parents

  • Comparison of Gestures:

    • Children's Home Sign:

    • New signs for new objects.

    • Sign combinations and systematic word order.

    • Structural complexity: emerges spontaneously, meaningful, language-like rules.

    • Parents' Gestures:

    • Rare new gestures, inconsistent, simple, unsystematic.

  • Conclusion: The linguistic structure emerges from the children themselves.

Part 4: Finding Structure in Input

Children Use Grammar Cues to Figure Out Word Meanings

  • Key Concept:

    • Children are predisposed to detect patterns in linguistic input.

    • Syntactic Bootstrapping:

    • Children use familiar grammar/syntax to infer meanings of new words (Fisher et al., 2020).

    • Examples of syntax leading to word identification:

      • "It's a blicket" indicates "blicket" is a noun.

      • Contexts like "I like the ___ " identify nouns, while "She ___ her lunch" identifies verbs.

Barbir et al. (2023): Hypothesis

  • Core Idea:

    • Infants keep track of which words appear in specific grammatical contexts and use familiar words to predict meanings of new words.

    • Example phrases to illustrate point:

    • "I like the banana. Where’s the apple? Want the bottle?" indicate nouns.

    • "She krads! She eats her lunch. She went home. She drinks some water." indicate verbs.

Barbir et al. (2023): Training Phase

  • Participants: 20-month-old French-learning infants.

  • Training: Short French video story with two made-up function words:

    • "ko" always precedes animate nouns (e.g., rabbit, chicken).

    • "ka" always precedes inanimate nouns (e.g., book, tractor).

  • Key Question: Can infants use "ko" and "ka" to predict whether a new word refers to something animate or inanimate?

Barbir et al. (2023): Test Phase

  • Procedure:

    • Infants observed a pair of objects (one animate, one inanimate) while hearing a sentence with either "ko" or "ka" + a novel noun (e.g., "bamoule").

  • Results:

    • Infants looked longer at the animate object when they heard "ko".

    • Interpretation: Infants learned that "ko" predicts animate nouns and "ka" predicts inanimate nouns, demonstrating sensitivity to distributional patterns in language input.

Conclusion: Three Pieces of the Language Learning Puzzle

  1. Timing matters: A sensitive period makes the brain especially receptive to language in childhood (findings from J&N, Hartshorne et al., Newport & Supalla).

  2. Quality input matters: Conversational interaction, beyond mere word exposure, shapes language skills and brain development (with insights from Romeo et al. and Huttenlocher et al.).

  3. Children are active learners: Children utilize innate structure-building (as seen in home sign) and powerful pattern-detection capabilities (as shown by Barbir et al.) in the language learning process.

  • Nativist–Empiricist Debate: It is reframed to emphasize not an either/or dichotomy but rather how each component interacts with the others.