This posting is the rest of my previous posting, vocabulary acquisition from birth to preschool (1). Before infants learn to communicate with words, their gestures signal an understanding of language (Lock, 1978; Namy & Waxman, 1998; Woodward & Hoyne, 1999). These referential gestures imply an understanding of others’ intentions, and are a nonverbal precursor of verbal reference (Golinkoff, Mervis, & Hirsh-Pasek, 1994). Truly communicative words arise when infants not only link spoken sounds with objects Vocabulary Acquisition: A Primer 3 and events in the environment, but also understand that words reference objects and concepts. An understanding of others’ intentions is essential for language acquisition. Joint attention serves as a means for referential word learning (Bloom, 2000; Tomasello, 2003). When the infant and the caregiver are jointly attending to an aspect of the environment, instigated perhaps with a communicative gesture like pointing, the caregiver labels salient objects and concepts. This joint sharing of attention allows the infant to attach the label to the object or concept in the environment. Words that are both familiar and novel to the infant thereby attain a real-world reference, and become true words.
When faced with the choice of labeling one object among a multitude of options that a caregiver is gesturing toward, infants’ word learning typically follows several lexical principles. The natural partitionshypothesis (Gentner, 1982) suggests that infants generally choose to label objects and concepts that are easily segmented from the environment; objects are discrete and more easily conceptualized than relational words.
Thus, English-speaking infants’ first symbolic words are often items and concepts that can be easily partitioned from the environment including concrete nouns, performatives (e.g., bye-bye), and nominals (e.g., breakfast), but generally include words from all parts of speech (Tomasello, 2003). Additional lexical principles include a whole-object constraint, where the child most often assigns a novel label to a whole object rather than to a part (Golinkoff et al., 1994; MacWhinney, 1998; Saylor & Sabbagh, 2004), and an exclusivity constraint, where a child will most often use a novel label for a novel object, instead of acquiring a second label for a known object (Bloom, 2000; Golinkoff et al., 1994; Masur, 1997).
From 18 to 24 months, infants are continuing to add a significant number of words to their vocabulary (Tomasello, 2003); some evidence of the beginning of a “vocabulary burst” around this time has been reported (Bates & Goodman, 1997; Mervis & Bertrand, 1995; Reznick & Goldfield, 1992). Although some infants demonstrate a “spurt” or “burst,” having been observed acquiring upwards of 10 new words in a 2- to 2½-week period, other infants show a more gradual increase in vocabulary (Mervis & Bertrand, 1995; Reznick & Goldfield, 1992).
Longitudinal estimates of word-learning rates vary greatly. The actual number of words known or acquired cannot feasibly be counted, so estimates of various kinds are used. The variability in reported rates of vocabulary acquisition depends to some degree on which procedures are used for calculating estimates. Biemiller and Slonim (2001) provided an estimate of 2.2 root words a day for children ages 1 through second grade. Templin (1957) provided an estimate of 5.1 new root words a day for 6-, 7-, and 8-year-olds. Carey (1978) provided an estimate of nine new words a day from 18 months to 6 years. Beck, McKeown, and Kucan (2002) conclude that the most commonly cited word-learning rate is seven new words a day, but this figure is an average of low- and high-ability word learners. Biemiller and Slonim (2001), for example, report that children in the lowest quartile for vocabulary size acquired only one word a day from age 1 to second grade, whereas children in the highest quartile acquired three new words a day. Even the most modest estimate hints at the enormous number of words that infants and young children are acquiring. At around the time children experience a vocabulary burst, they also demonstrate the ability to acquire novel labels rapidly (Mervis & Bertrand, 1995). Carey (1978) dubbed this process of word acquisition fast mapping, and found that children can acquire, or map, lexical information about a novel label in a single encounter. Carey and Bartlett (1978) explored fast mapping in a population of 3- and 4-year-olds, finding that although a more complete understanding of a word’s meaning develops over multiple exposures, children readily acquire partial knowledge about a word’s meaning and its referent from a single encounter.
Researchers have continued to explore fast mapping in younger populations of children. Eighteen-month-olds can fast-map novel gestures as object labels (Namy & Waxman, 1998). Two-year-olds will map novel words in a variety of implicit situations, using intention-reading
skills that allow the child to infer the meaning of a novel word from adults (Tomasello, 2003). Children as young as 2½ use familiarity and syntactical cues to map novel labels for parts of objects (Saylor & Sabbagh, 2004). Two- to 4-year-olds utilize contrast cues (presenting familiar and unfamiliar objects together) to fast-map labels to novel objects (Wilkinson, Ross, & Diamond, 2003). Two- and 3-year-olds fast-map novel adjectives to familiar nouns with strong syntactical and contrast cues (specifically cross-situational comparisons; see Mintz & Gleitman, 2002). Infants will even fast-map a novel label equally well when exposed directly (infant-directed speech) and indirectly (overhearing adult-directed speech) to a word (Akhtar, Jipson, & Callanan, 2001). In summary, young children acquire new words with and without direct instruction by utilizing various cues from the natural linguistic environment.
Around 18 months, while infants are continuing to add words to their vocabulary and gain motor control, gestures in conjunction with Vocabulary Acquisition: A Primer 5 single words arise; this combination generally signals the onset of twoword combinations (Lock, 1978). Initial two-word combinations are concrete constructions that do not have any syntactical meaning (e.g., “bottle Mommy”) (Tomasello, 2003). From 24 to 36 months, young children’s syntactical awareness begins to develop; their constructions become less concrete and reflect the types of constructions that are most prevalent in their caregivers’ speech (Huttenlocher, Haight, Bryk, Seltzer, & Lyons, 1991; Tomasello, 2003). A young child’s earliest understanding of syntax therefore appears to be centered around verbs and constructions that are commonly heard in his or her caregiver’s speech.
Additional exposure to speech paired with a growing corpus of experience with language allows children to abstract general rules about grammatical speech (Tomasello, 2003). Children’s ability to apply abstract syntactical rules to their speech signifies that they are able to construct grammatical phrases that are not merely reflections of the speech they have heard. In other words, an understanding of abstract rules of speech allows infants to become innovative with language. Children’s ability to be innovative with language and experiment with previously unheard words and constructions means that some innovations will contain rule overgeneralizations (e.g., “I go-ed to school”). Various environmental factors work to constrain language to conventional standards. Syntactic constraints, for example, restrict children to using commonly heard verbs in commonly heard constructions (e.g., “He made the rabbit disappear” rather than “He disappeared the rabbit”; Tomasello, 2003, p. 178). Additionally, a more complete understanding of verb subclasses (e.g., the difference between run and scamper) constrains a child’s innovations by limiting the applicable contexts (Brooks, Tomasello, Dodson, & Lewis, 1999). Whereas speech perception is constrained maturationally (e.g., limited cognitive abilities), language production appears to be constrained environmentally.
Thank you so much for your attention!
0 komentar:
Posting Komentar