ED 030 235 By -Greenough. Diane Comprehension and Imitation of Sentences by Institutionalized Trainable Mentally Retarded Children as a Function of Transformational Complexity. Michigan Univ., Ann Arbor:: Cent& for Research on Language and Language Behavior.; Mount Pleasant State Home and Training School, Mich. Spans Agency -Office of Education (DHEW), Washington. D.C. Bureau of Research. Bureau No -BR -6 -1784 Pub Date 68 Contract -OEC -3 -6 -061784 -0508 Note -18p. EDRS Price MF -$0.25 HC -$1.00 Descriptors -Cognitive Processes. Complexity Level, *Comprehension. Deep Structure, *Exceptional Child Research. *Imi.tation, Institutionalized (Persons), Kernel Sentences, Language Develcpment, *Mentally Handicapped. Mongolism, Negative Forms (Language), Pictorial Stimuli, Sentence Structure, Surface Structure, Trainable Mentally 'Handicapped, *Transformations (Language), Verbal Stimuli The ability of 31 institutionalized trainable mentally retarded mongoloid and nonmongoloid subiects (mean 10=27.5) to comprehend and imitate verbally presented sentences of varying levels of grammatical complexity was examined. Eight stimulus sentences were spoken for four pairs of pictures, simple declarative or kernel, negative, passive, and negative passive. Subfects were asked to select one of a pair of pictures for each sentence and then to repeat the sentences. They corr.ctly comprehended kernel sentences significantly more often than other kinds of sentences. Among nonmongoloid subfects, imitalion of sentences to whick one optional transformation had been applied was significantly better than .imitation of sentences modified by two potential transformations. Chronological age was significantly associated with imitation ability in nonmongoloids (p .04); 10 and imitation ability were significantly associated in mongoloids (p .05), indicating that the ability to store verbal material for immediate recall is associated with general intellectual or cognitive abilities. (Author/RJ) 110 etIMPINT R PIA IIMR 48 EC 003 891
20
Embed
for four pairs of pictures, of a pair of pictures for each ... · sentences of varying levels of grammatical complexity was. examined. Eight stimulus. sentences were spoken. for four
This document is posted to help you gain knowledge. Please leave a comment to let me know what you think about it! Share it to your friends and learn new things together.
Transcript
ED 030 235By -Greenough. DianeComprehension and Imitation of Sentences by Institutionalized Trainable Mentally Retarded Children as aFunction of Transformational Complexity.
Michigan Univ., Ann Arbor:: Cent& for Research on Language and Language Behavior.; Mount Pleasant StateHome and Training School, Mich.
Spans Agency -Office of Education (DHEW), Washington. D.C. Bureau of Research.
Bureau No -BR -6 -1784Pub Date 68Contract -OEC -3 -6 -061784 -0508Note -18p.EDRS Price MF -$0.25 HC -$1.00Descriptors -Cognitive Processes. Complexity Level, *Comprehension. Deep Structure, *Exceptional Child
Research. *Imi.tation, Institutionalized (Persons), Kernel Sentences, Language Develcpment, *Mentally
The ability of 31 institutionalized trainable mentally retarded mongoloid andnonmongoloid subiects (mean 10=27.5) to comprehend and imitate verbally presentedsentences of varying levels of grammatical complexity was examined. Eight stimulussentences were spoken for four pairs of pictures, simple declarative or kernel,
negative, passive, and negative passive. Subfects were asked to select one of a pairof pictures for each sentence and then to repeat the sentences. They corr.ctlycomprehended kernel sentences significantly more often than other kinds ofsentences. Among nonmongoloid subfects, imitalion of sentences to whick one optionaltransformation had been applied was significantly better than .imitation of sentencesmodified by two potential transformations. Chronological age was significantlyassociated with imitation ability in nonmongoloids (p .04); 10 and imitation ability weresignificantly associated in mongoloids (p .05), indicating that the ability to storeverbal material for immediate recall is associated with general intellectual or cognitiveabilities. (Author/RJ)
110 etIMPINT R PIA IIMR
48 EC 003 891
de.49- 4'foz..51
COMPREHENSION AND IMITATION OF SENTENCES BY INSTITUTIONALIZED
TRAINABLE MENTALLY RETARDED CHILDREN AS
A FUNCTION OF TRANSFORMATIONAL COMPLEXITYr
11 !tel !! ;...tuftCD CP = CD
r"
owl
re Fio-40
Diane Greenough 3Cr C) ..II
Center for %search on Language and Language Behavior rn
The University of Michiganrvi CM XJP el ""74.418
ZCA Fl re
5731 Vnn 4CP FC:3: G;....4 =
Mount Pleasant State Home and Training School e-0mc rn
The ability of 31 trainable mentally retarded (TMR) mongoloid4= c-en NO 4
.4
and nonmongoloid Ss ()CIA = 27.5) to comprehend and imitate verbally DC
presented sentences of varying levels of grammatical complexity was RR= a
examined in two studies. Comprehension ability was tested by ask- CD A; cr. ring each S to indicate which of 2 pictures was being described in rn
the sentence spoken by E. Eight stimulus sentences were spoken by
E for each pair of pictures - -simple declarative or kernel (K), neg-. 11.1"153CCD 7C 4
ative (p), passive (P), and negative passive (NP) for each picture. c) .....mc
S's ability to imitate sentences of varying levels of complexity was4"
tested by having him repeat the sentences used in the comprehension
task, with passive forms truncated so that stimuli would be nearly
equal in length. Ss correctly comprehended kernel skntences signi-
ficantly more often than chance expectancy. Among nonmongoloid Ss,
imitation of sentences to which one optional transformation had
been applied was significantly better than imitation of sentences
modifidd by tworoptional transformations. CA was significantly
associated with imitation ability in nonmongoloid Ss; IQ and imita-
tion ability were significantly associated in mongoloid Ss. Results
are discussed in terms of competence and performance variables which
might affect the ability of TMR Ss to deal with verbal stimuli.
The theory of transformational grammar states that the knowledge which a
speaker has of his language may best be described in terms of a deep or base
structure component plus an additional set of operations called transformations
(Chomsky, 1965). It is not claimed that such a model describes how the speak-
er generates the sentences he uses, but rather that it describes the intrinsic
knowledge that a speaker must have to use the language correctly (Chomsky,
1965). It appears likely, however, that the grammatical complexity of a sen-
tence is inversely related to the ease with which it can be uttered and under-
stood by speakers of the language. Results of studies in whi h Ss indicated
the truth-value of a sentence in relation to a picture presented by E have
been offered as evidence that speed of comprehdnsion is inversely proportional
269
-71,
Greenough 2
to the number of transformations separating a sentence from its base struccure.
In experiments with normal adults (Gough, 1965, 1966; Slobin, 1966) and chil-
dren (Slobin, 1966), it was found that latencies of correct responses increased
in the following order: kernel-passive-negative-negative passive. Simple
declarative (kernel) sentences result from obligatory transformations of the
base structure (Chomsky, 1965) and are thought to be the simplest sentence
forms. Addition of the semantic element of negation apparently creates more
difficulty for comprehension than do grammatical transformations. Thus,
Slobin and Cough found slowest comprehension rates when the negative (semantic)
transformatior was applied. Slobin also reported that chronological age (CA)
interacts significantly with syntactic structure--suggesting developmental
changes in linguistic competence. IQ and reaction time to sentences did not
correlate significantly.
Semmel and Greenough (1968) tested the generalizability of the findings
of Gough and Slobin to trainable mentally retatded (TMR) mongoloid children.
Ss were able to comprehend kernel sentences at levels significantly beyond
chance expectancy. They seemed to treat simple negative strings as affirma-
tives and so they were correct significantly less often than would be ex-
pected by chance. Their responses to passive and negative-passive strings
appeared to be random. Imitation of simple declarative sentences (strings
whose base structure had been modified by obligatory transfomations only)
was significantly better than imitation of sentences to whSch optional trans-
formations (negative and/or passive) had been applied. All strings to which
optional transformations had been applied were imitated equally well. IQ was
highly correlated with imitation ability.
In the present studies the comprehension and imitation abilities of in-
stitutionalized TMR children were tested. It was predicted that the institu-
tionalized Ss would not be able to process syntactically or semantically cm-_plex sentences so as to understand their content (comprehension) or store them
efficiently in immediate memory for later retrieval (imitation). In both the
comprehension and imitation tasks, Ss would presumably perform much better on
K (kernel) sentences than on N (negative), P (passive), or NP (negative-pas-
sive) sentences. The negation marker was stressed in the reading of N and NP
sentences in the comprehension task, in order to reduce the likelihood that
Ss were missing this cue and treating the strings as affirmative statements.
270
Greenough 3
Both mongoloid and nonmongoloid Ss were used so that possible effects of
biological differences between Ss on comprehension and imitation abilities, as
well as effects of factors associated with institutionalization, might be dis-
covered.
MethodStudy I: Comprehension
Subjects,. Ss were 31 children ranging in CA from 7-7 to 20-8, ell resi-
dents of the Mt.-Pleasant State Home and Training School and classified as
trainable in a recent educational evaluation program there. Ss were selected
so that the mean CA and IQ in the mongoloid and nonmongoloid subsamples would
be similar. Peabody Picture Vocabulary Test scores from tests administered in
1966 served as measures of IQ in the selection of Ss. There were 16 mongoloids
(RCA
= 1573, s.d. = 12.4 mos..'
XIQ
= 26, s.d. = 2.90) and 15 nonmongoloids
(RCA
= 1575, s.d. = 11.7 mos..'
RIQ
29, s.d. = 2.53) in the total sample.
There were five females in each of the two subgroups.
Procedure. Four pairs of pictures, each pair representing both aspects
of a reversible situation, were used as stimuli. The pictures were brightly
colored and were drawn on 8 1/2 x 11-inch white paper. The following situa-
tions were depicted: (a) Boy pushing girl in wheelbarrow; girl pushing boy in