An interesting lecture by Prof. Elma Blom from Utrecht
University, which I watched in preparation for a meeting about bilingualism at
the Efrat Educational Psychological Service. Prof. Blom specializes in language
development and multilingualism in the family and in education. The first part
of the lecture deals with language impairment in general, and the second part
deals with language impairment and bilingualism, and proposes two diagnostic
tools.
Part A: Language Impairment
About 5–7 percent of the population has a specific language
impairment. This impairment is heterogeneous and can affect different aspects
of language, such as phonology, word retrieval, speech, and articulation.
Grammar is usually affected as well. For example:
…and him say….he jump and smash….
Such errors are normal at age two, but not in an
eight-year-old English-speaking child. Children may also have difficulty with
pronouns — I, you, he, she — and/or with clitics, which are forms that
do not occur as independent words but attach to the beginning or end of a word,
such as the 's in she's or in England's flag, the 'll in we'll go, the n't in
don't. In each language, different aspects of grammar may be affected.
Most of the language difficulties of children appear in
production, but there are also problems in language comprehension. A child who
has problems both in production and comprehension is in a worse situation.
There is also a gap in vocabulary between children with language impairment and
children without language impairment, but the gap in grammar is larger. As
children with language impairment grow older, they close the gap in simple
grammatical rules, but they still struggle with complex grammatical rules.
Older children manage to avoid using complex grammar, and therefore their
difficulty is less noticeable. Children with language impairment also continue
to make progress in vocabulary.
Children with language impairment have greater difficulty
inflecting a regular verb in the past tense, for example dance–danced,
than typically developing children of the same age, and also more than
typically developing children at the same language level, that is, younger
children. But there is no difference between children with language impairment
and typically developing children in the inflection of irregular verbs such as go–went
or write–wrote. An irregular verb has to be memorized, and it does not
work according to a rule. It is like a word in the vocabulary. Typically
developing children are more successful at inflecting regular verbs than
irregular verbs. By contrast, among children with language impairment, there is
no difference between the inflection of regular and irregular verbs; they have
difficulty with both.
Some
researchers believe that the difficulty children with language impairments have
in inflecting regular verbs stems from an impairment in procedural learning. Procedural
learning belongs to the broad ability of “learning efficiency.” Procedures are
usually multi-step processes, such as tying shoelaces. Writing letters and
words is also a procedure, manifested in the motor movement used to write each
letter and to write the whole word. Long multiplication is also a procedure. In
many languages, there is a procedure for verb inflection, for example adding -ed
to verbs in English. An impairment in procedural learning should affect
inflections of regular verbs, but not inflections of irregular verbs, which do
not rely on a procedure and are related to vocabulary and declarative memory. In
other words, this approach would probably argue that the difficulty children
with language impairment have with inflecting irregular verbs is probably
related to difficulty acquiring vocabulary, and not to difficulty in procedural
learning.
Children with language impairment often also have additional cognitive impairments, in verbal working memory, executive control — especially inhibition and planning — and processing speed.
Part B: Language Impairment and Bilingualism
Do bilingual children usually also have a language delay?
Usage-based theory argues that every linguistic input
strengthens linguistic representations in long-term memory. Because the
linguistic input of bilingual children is divided between two languages, these
children may have language delays because they receive less linguistic input,
and for a shorter time, in each of the languages compared with their
monolingual peers. But if the input in both languages is balanced and rich,
bilingual children are less likely to show delays in language development.
Bilingual
children have an advantage over monolingual children in executive functions,
especially in inhibition, shifting, and working memory. This advantage
probably develops because in bilinguals, both languages are active all the
time. When they use one of the two languages, they have to actively suppress
the other language. This ability to suppress — to apply inhibition — trains
executive functions.
Blom and colleagues examined language and working memory in
four groups of five- and six-year-old children: monolingual children with
typical development, monolingual children with language impairment, bilingual
children with typical development, and bilingual children with language
impairment.
The first language task examined was receptive vocabulary —
that is, word comprehension, as distinct from word retrieval. In receptive
vocabulary tests, the child sees several pictures and is asked to point to an
object or action shown in one of the pictures. In this task, typically
developing monolingual children performed best. Monolingual children with
language impairment and typically developing bilingual children performed less
well, with no difference between them. Bilingual children with language impairment
performed worse than all the other groups. In other words, both language impairment and
bilingualism have a negative effect on receptive vocabulary. A bilingual
child who also has language impairment will have low receptive vocabulary for
both of these reasons together. Monolingual children with language impairment
were stronger in receptive vocabulary than bilingual children with language
impairment. It is not clear to me whether this study examined the vocabulary of
the bilingual children in both languages or only in one language. When both
languages are taken into account together, bilingual children are sometimes
found to have normal vocabulary.
Inflection
of regular verbs:
typically developing monolingual children were better than the other
three groups, which did not differ from one another.
Inflection
of irregular verbs:
typically developing monolingual children were best, followed by
typically developing bilingual children, while monolingual and bilingual
children with language impairment were lowest.
That is,
in the inflection of both regular and irregular verbs, there is a negative
effect of bilingualism and of language impairment, but there is no cumulative
effect of bilingualism and language impairment together: bilingual
children with language impairment do not perform worse than monolingual
children with language impairment.
Working
memory: the
bilingual children, both typically developing and language-impaired, were
better than the monolingual children, both typically developing and
language-impaired. In other words, it is possible that the simultaneous manipulation of two languages in
bilingual children trains and improves working memory.
In
summary, bilingual children with language impairment showed impairment in
vocabulary, but less so in grammar, and they had an advantage over monolingual
children in working memory. In terms of usage-based theory, it may be that
vocabulary is affected by the amount of input more than grammar is, because
once the child has acquired a rule or procedure, the child does not need
additional input in order to use the rule.
How can we distinguish between a language deficit related
only to bilingualism and a language impairment?
In the Netherlands, the percentage of children from
cultural minority groups who are in special education is higher than their
percentage in regular education. This is probably also the case in Israel. In a
document written by Nurit Yachimovich-Cohen from the Knesset Research and
Information Center, in the 2018–2019 school year, about 18% of immigrant
students studied in special education, compared with a rate of about 11.2% of
special-education students in Hebrew education as a whole. All over the world,
professionals have difficulty determining whether a bilingual child is
suffering only from gaps in the second language, which the child is still in
the process of acquiring, or whether the child’s difficulties also stem from a
language impairment. Some children are merely bilingual and are mistakenly
diagnosed as having language impairment, while other bilingual children do also
have language impairment but are not diagnosed as such.
Two tasks have been found that can distinguish between
bilingualism and language impairment.
The first task is nonword repetition. This task requires phonological
perception and discrimination between phonemes, together with short-term
memory. These skills are required for acquiring new words and for acquiring the
grammatical forms of words. This is why children with language impairment have
difficulty repeating nonwords. Nonwords are new for all children, regardless of
their linguistic background. Therefore, a nonword repetition test is very
useful in the assessment of bilingual children, and in the assessment of
children whose mother tongue is not spoken by the psychologist. A nonword
repetition test is less culturally biased than other language tasks.
Performance on it is less affected by experience in the second language than
performance on a standard language assessment. Nevertheless, children with a
larger vocabulary in a particular language may perform this task better in that
language, because even nonwords have a structure that fits a particular
language. For example, it is clear to us that blorked is a passive verb —
the pattern of blorked is identical to the pattern of blocked. In
other words, blorked is more of an English nonword than a nonword in
other languages.
Shula Chiat, who works at the University of London, created a nonword test that attempts to be as universal as possible and to include sounds and word structures that exist in many languages. Information about the test, called CR-NWR, the Crosslinguistic Nonword Repetition Test, can be read on this website.
Elma Blom examined the universal nonword repetition test
with the same four groups of five- and six-year-old children discussed earlier.
Monolingual and bilingual children with language impairment performed less well
on this test than monolingual and bilingual children without language
impairment. There was no difference between monolingual and bilingual children
with language impairment on this test. There was no difference between
monolingual and bilingual children without language impairment on this test.
Blom compared the universal test with a nonword repetition test that is
specific to Dutch, and found that the universal test was better for bilingual
children than the language-specific test.
The
second task that was found to distinguish between language impairment and
bilingualism is storytelling based on a picture, or narrative production.
Such a test was developed by Gagarina et al. (2012). In storytelling one can
distinguish between the macro level and the micro level. At the macro level, we
examine whether the child included all the components of the story: opening,
introduction of the characters, conflict, attempt to resolve the conflict, the
internal state of the characters, and the ending of the story. At the micro
level, we consider the vocabulary the child used, sentence length, grammar, and
syntax. Many studies have
found that children with language impairment have difficulty producing
narratives, both at the macro level and at the micro level. By contrast,
bilingual children without language impairment do not have difficulty at the
macro level, but do have difficulty at the micro level.
The diagnostic tool includes four stories, which make it
possible to assess both languages, and one model story. The examiner tells the
model story, the child answers questions about the story, and then the child
tells another story independently based on pictures. Story-memory tests in
which the child is asked to repeat a story assess memory. Here, the goal is to
assess language, not memory. The model story is intended to demonstrate the
task.
Elma Blom examined the narrative-production test with the
four groups of children. Only the macro level was examined. Monolingual and
bilingual children with language impairment performed less well than
monolingual and bilingual children without language impairment. There was no
difference in performance between monolingual and bilingual children with
language impairment. In other words, this test is sensitive to language impairments and is suitable for
both monolingual and bilingual children.
Bilingualism and Language Impairment: Theoretical and
Clinical Issues — Elma Blom
https://www.youtube.com/watch?v=dJSmxcJYwc4
Yachimovich-Cohen, Nurit. 2019. Immigrant
Students in Special Education — Data and Aspects of the Work of Placement and
Eligibility Committees. Knesset Research and Information Center.
https://fs.knesset.gov.il/globaldocs/MMM/f1782129-518a-e811-80de-00155d0a0b8d/2_f1782129-518a-e811-80de-00155d0a0b8d_11_12436.pdf
Gagarina, N. V., Klop, D., Kunnari, S.,
Tantele, K., Välimaa, T., Balčiūnienė, I., ... & Walters, J. (2012). MAIN:
Multilingual Assessment Instrument for Narratives. ZAS Papers in Linguistics,
56, 155–155.
https://doi.org/10.21248/zaspil.56.2019.414
Comments
Post a Comment