LING0001 -- Lesson 23 -- First Language Acquisition (2023)

Phases of language acquisition in children

In almost all cases, children's language development follows a predictable pattern. However, there are big differences in the age at which children reach a certain milestone. In addition, in general, the development of each child is characterized by the gradual acquisition of certain skills: thus, the "correct" use of English verb inflection develops over a period of a year or more, beginning with a stage when inflected verbs are always omitted, and ends in a phase in which they are almost always used correctly.

There are also many different ways to characterize the developmental process. On the production side, one way of naming the stages is as follows, with the main focus being on the unfolding of lexical and syntactic knowledge:

Vocalizations in the first year of life

At birth, an infant's vocal tract resembles in some respects that of a monkey rather than that of an adult human. Compare the infant vocal tract diagram shown at left withDiagrams of adult humans and apes.

In particular, the tip of the velum reaches or overlaps the tip of the epiglottis. As the baby grows, the tract gradually remodels into the adult pattern.

During the first two months of life, the infant's vocalizations are mainly expressions of discomfort (crying and fussing), along with sounds that arise as the by-product of reflexes or vegetative actions, such as coughing, sucking, swallowing, and belching. There are some non-reflective and non-disturbing sounds made with the veil lowered and the mouth closed or nearly closed, giving the impression of a syllabic nasal or a nasalized vowel.

During the period of about 2 to 4 months, babies begin to make "comfort sounds," usually in response to a pleasant interaction with a caregiver. Early comfort sounds may be grunts or sighs, with later versions more like vowel "cooing". The vocal tract is held in a fixed position. Initially, comfort sounds are short and produced individually, but later appear in series separated by glottal stops. Laughter appears about 4 months.

During the 4 to 7 month period, infants often engage in "voice play", manipulating pitch (to produce "queen" and "growl"), loudness (to produce "whoops"), and also manipulating the closure of the tract, to produce nasal rubbing sounds mumbling, "raspberries" and "puffing".

Around seven months of age, “canonical babble” occurs: Babies begin to utter prolonged sounds that are rhythmically cut into syllable-like sequences by the oral articulations as the jaws, lips, and tongue open and close. The range of sounds produced can be heard in both attack and slide. Fricatives, affricates, and liquids are heard less often, and clusters are even less common. Vowels tend to be low and open, at least initially.

Repeated sequences such as [bababa] or [nanana] are often produced, as are "variegated" sequences in which the properties of the consonant articulations are varied. Varied sequences are initially rare and later become more common.

Both vocal play and babble are more commonly produced in interactions with caregivers, but babies also produce them when they are alone.

No other animal babbles like that. It has often been hypothesized that voice play and babbling function to 'practice' language-like gestures, helping the child to gain control of the motor systems involved and to learn the acoustic consequences of various gestures.

One Word Stage (holophrastic)

Babies begin to utter recognizable words around ten months. Some word-like vocalizations that do not correlate well with words in the local language may be used consistently by certain infants to express certain emotional states: an infant is said to have used thisexpressing pleasure and someone else is said to have used itto express "sorrow or discomfort". Most often, recognizable words are used in a context that appears to involve naming: "duck," when the child bangs a toy duck on the edge of the bathtub; "sweep" while the child sweeps with the broom; “Car” when the child looks out the living room window at the passing cars on the street; "Dad" when the child hears the bell.

Young children often use words either very narrowly or very broadly: "bottle" is used only for plastic bottles; "Teddy" is only used for a specific bear; "Dog" is used for lambs, cats and cows, and dogs; "Kick" is used for pushing and flapping, as well as kicking. Thissub-extensionseexcessive extensionsevolve and change over time with use by an individual child.

Perception vs. Production

Clever experiments have shown that by the age of 4 to 9 months, most babies can demonstrate (e.g. by looking in the right direction) that they have understood a few words, often before they even start babbling. In fact, the development of phonological skills begins even earlier. Newborns can distinguish speech from non-speech and can also distinguish between speech sounds (e.g., [t] vs. [d] or [t] vs. [k]); A few months after birth, babies can distinguish speech in their native language from speech in other languages.

Early verbal interactions with mothers, fathers, and other caregivers are almost certainly important in building and reinforcing these early skills long before the child shows any signs of language proficiency.

vocabulary development rate

First, babies gradually add active vocabulary. Here are measurements of active vocabulary development in two studies. Nelson's study was based on diaries kept by mothers of the language of all their children, while Fenson's study was based on asking mothers to compare words against a list to indicate which words they thought their children were producing.

Vocabulary acquisition often occurs in leaps and bounds in the second year. First words are acquired at a rate of 1-3 per week (as measured by production diaries); In many cases, after learning 40 or more words, the rate can suddenly increase to 8 to 10 new words per week. However, some children show a more stable activity rate in these early stages. From the third year, vocabulary acquisition accelerates significantly: A plausible estimate would be an average of 10 words per day in preschool and primary school.

Gender differences in vocabulary acquisition

In a context of enormous individual variation, girls tend to learn more words faster than boys; but the difference disappears over time.

Svetlana Lutchmaya, Simon Baron-Cohen and Peter Raggat ("Fetal testosterone and vocabulary size in 18- and 24-month-old infants",behavior and development of children24:418-424, 2002) found that in a sample of 18-month-old boys, the mean vocabulary was 41.8 words (range 0 to 222, standard deviation 50.1), while the mean for girls was 86.8 (Range from 2 to 318, standard deviation 83.2). At 24 months, the difference narrowed to a mean boy score of 196.8 (range 0 to 414, standard deviation 126.8) versus a mean girl score of 275.1 (range 15 to 415, SD = 121.6). In other words, the girls' lead in mean scores shrank from 86.8/41.8 = 2.1 to 275.1/196.8 = 1.5.

Over time, the difference disappears completely and then reappears in the opposite direction, with males having a larger average vocabulary during their college years (though again related to the within-group variation, which is much larger than the differences between males and females). Women). ). Here is Table 6 by JanetShibleyHyde and MarciaC.Linn, "Gender differences in verbal ability: a meta-analysis",Psychological Bulletin104:1 53-69 (1988).

perception vs. production again

Benedict (1979) asked mothers to keep a diary listing not only the words the children produced but also the words they used to give signs of understanding. Their results show that when children produced 10 words, they understood an estimated 60 words; and an average of five months elapsed between understanding 50 words and producing 50 words.

All of these methods (mother's diaries and checklists) are likely to underestimate the number of words about young children who actually know something, although they may also overestimate the number of words to which they attribute adult meanings.

Combining Words: The Rise of Syntax

Word combinations appear in the second year. New combinations (where we can be sure that the result will not be treated as a single word) appear sporadically as early as 14 months. At 18 months, 11% of parents say their child combines words often and 46% say they sometimes combine words. At 25 months, almost all children sometimes combine words, but about 20% still do not "often".

First statements of different units

In some cases, the first utterances of multiple entities can be viewed as concatenations of individual naming actions that might as well have taken place alone: ​​"mama" and "hat" can be combined as "mama-hat"; “shirt” and “wet” can be combined as “wet shirt”. However, these combinations tend to appear in an order that corresponds to the language being learned:

1. puppy heartbeat
2. Ken Water (for "Ken is drinking water")
3. spank puppy

Some combinations with certain closed-class morphemes also occur: "my train", "in there", etc. However, these are closed-class words, such as pronouns and prepositions, that have their own semantic content, which does not differ significantly from words of the open class differs. The more purely grammatical morphemes - verb inflections and verb auxiliaries, noun determiners, complementers, etc. - are usually absent.

Since the first multiple unit utterances are almost always two morphemes - two is the first number after one! -- this period is sometimes referred to as the "two word stage". Soon, however, children sometimes begin to produce utterances with more than two items, and it is unclear whether the time when most utterances contain one or two lexical items should really be treated as a separate phase.

In the early multi-word phase, children who are asked to repeat sentences can simply omit determiners, verbal modal and auxiliary verbs, verb inflections, etc., and often pronouns as well. The same pattern can be seen in her own impromptu speeches:

1. "I can see a cow" repeated as "See cow" (Eva at 25 months)
2. "The dog will bite" repeated as "dog bite" (Adam at 28 months)
3. Kathryn doesn't like celery (Kathryn at 22 months)
4. Doll toy truck (Allison at 22 months)
5. Pig Says Oink (Claire at 25 months)
6. Do you want a lady to get chocolate (Daniel at 23 months)
7. "Where is Dad going?" repeated as "Is Daddy going?" (Daniel at 23 months)
8. "Car goes?" means "Where is the car going?" (Jem at 21 months)

The pattern of omitting most grammatical/functional morphemes is referred to as "telegraphic", and hence the initial stage of various words is also sometimes referred to as "telegraphic stage".

Acquisition of grammatical elements and corresponding structures

Around the age of two, children begin to use grammatical elements. In English, this includes finite auxiliary verbs ('is', 'was'), tenses and agreement affixes ('-ed' and '-s'), nominative pronouns ('I', 'she'), complementers ('that', 'wo ’) and modifiers (‘the’, ‘the’). The process is usually somewhat gradual, in which more telegraphic patterns alternate with adult or adult-like forms, sometimes in adjacent utterances:

1. She's gone. your school is gone (Domenico at 24 months)
2. He kicks a beach ball. She goes up the stairs there. (Jem at 24 months).
3. I tease mom. I'm teasing Mom. (Holly at 24 months)
4. I have this. I have 'Nana. (Olivia aged 27 months).
5. I get this little one. I want that. (Betty at 30 months).
6. Mom isn't done yet, is she? (Olivia at 36 months).

Over the course of a year or a year and a half, sentences become longer, grammatical elements are less often omitted and misspelled, and sentences with multiple subordinate clauses become more common.

perception vs. production again

Several studies have shown that children who regularly omit grammatical elements in their speech, however, expect these elements in what they hear from adults, in the sense that their sentence comprehension is affected when grammatical elements are absent or missing.

backward progress

Morphological inflections often include a regular case ("walk/walked", "open/opened") and some irregular or extraordinary cases ("go/went", "throw/threw", "hold/hold"). Initially, these words are used in their root form. When inflections are added, regular and irregular patterns are found. At a certain point it is common for children to overgeneralize the regular case and produce forms like "brought", "went"; "Feet", "mice", etc. At this stage, the child's speech can become realless correctby adult standards than before, because of over-regulation.

This over-regulation, like most other aspects of grammar in early childhood development, is typically correction-resistant:

CHILD: My teacher held the rabbits and we pet them. ADULT: You said your teacher held the bunnies. CHILD: Yes. ADULT: What did you say? CHILD: She held the rabbits and we pet them. ADULT: You said she was holding her? CHILD: No, she kept them loose.

More information

A good place to start for more information on children's language acquisition is theKINDERCMU website where you can get information on how to download the raw materials for child language research and also research a specialized child languagebibliography.

A recent article in NYT Magazine (Paul Tough, "What does it take to make a student", 26.11.2006) discusses at length some well-known studies on social class differences in language acquisition (Betty Hart and Todd Risley, "Significant Differences in the Everyday Experience of Young American Children" (1995); Betty Hart, "A natural history of early language experience",Issues in early childhood special education, 20(1), 2000; "The first catastrophe: The 30 million word gap",American educator, 27(1) pp. 4-9, 2003). The summary of the 2003 article:

By the age of 3, children from privileged families have heard 30 million more words than children from disadvantaged families. Longitudinal data from 42 families examined what accounted for the large differences in vocabulary growth rates. The children were found to be similar to their parents in stature, activity level, vocabulary, language, and interaction style. Follow-up data showed that 3-year performance measures predicted third-grade school performance.

42 is not a very large sample and there are many other questions that need to be asked, but this work suggests that we should be concerned about the possible lasting effects of cultural differences on children's language environments.

Another recent study that comes to the same conclusion is Martha J. Farah et al. (“Child poverty: specific associations with neurocognitive development", Brain Research 1110(1) 166-174, September 2006). Prof. Farah and coworkers "managed a set of tasks aimed at assessing specific neurocognitive systems for healthy children of low and intermediate SES [socioeconomic status], who were examined by a physician history and matched for age, sex and ethnicity".

Cowardly. 1. Effect sizes, measured in standard deviations of the separation between the performance of the low and intermediate SES groups, on the composite measures of the seven different neurocognitive systems assessed in this study. Black bars represent effect sizes for statistically significant effects; gray bars represent effect sizes for non-significant effects.

All participants in this study were African American girls between the ages of 10 and 13. As the graph above shows, the difference in performance in the “Language” domain of the test battery between girls with medium and low SES represented an effect size of about 0.95.

There were two language-related tasks:

Peabody Picture Vokabeltest (PPVT)
This is a standardized vocabulary test for children between the ages of 2.5 and 18. With each attempt, the child hears a word and has to choose the corresponding picture from four options.
Grammar Reception Test (TROG)
In this task of combining sentences and figures, devised by Bishop (1982), the child hears a sentence and must choose from a set of four pictures the picture that represents the sentence. Its lexico-semantic requirements are negligible since the vocabulary is simple and a pre-test ensures that subjects know the meaning of the small set of words appearing in the test.

This finding is consistent with a persistent effect of differences such as those in the Hart & Risly study, although other explanations are possible.