Teaching Sign Language to Hearing Children as a Possible

Factor in Cognitive Enhancement

O. Capirci
A. Cattani
Institute of Psychology, Italian National Research Council
P. Rossini
Gruppo SILIS
V. Volterra
Institute of Psychology, Italian National Research Council

Downloaded from https://academic.oup.com/jdsde/article/3/2/135/393210 by guest on 31 October 2023

We describe an educational experience designed to teach Ital- this linguistic feature? Bellugi and her colleagues stud-
ian Sign Language (LIS) to a group of hearing children. The ied visual-spatial cognition in deaf signers, comparing
hypothesis underlying this experience was that learning a vis-
their performance to that of hearing, nonsigning chil-
ual-gestural language such as LIS may improve children's at-
tentional abilities, visual discrimination, and spatial memory. dren on a battery of visual-spatial tests (Bellugi,
To examine this hypothesis, we conducted two studies. The O'Grady, Lillo-Martin, O'Grady, van Hoek, & Corina,
first involved an educational experience lasting two years 1990). In tests of spatial construction, spatial organiza-
with a group of hearing children attending a Sign Language
tion, and facial recognition, deaf signing children were
class from first to second grade. The Raven PM 47 TEST
was administered at the beginning and at the end of each markedly ahead of the hearing nonsigning children and
school year to children attending the LIS classes and to a far in advance of their chronological norms. Similar re-
control group of children enrolled in the same school but not sults were reported by Chovan, Waldron, and Rose
exposed to LIS. The second study involved an educational (1988), indicating that deaf middle school and high
experience in first grade. The Raven PM 47 and Corsi's
block-tapping tests were administered at the beginning and
school students had faster responses in visual cognition
at the end of the school year to the children attending the tasks than their hearing peers.
LIS classes, to children enrolled in the same school but at- To clarify the relation between familial deafness
tending an English class, and to children not exposed to a and intelligence, Zweibel (1987) examined the intellec-
second language. We found that in both studies the LIS
group performed better than the other groups. These results
tual abilities of 243 children, each with familial deaf-
suggest that learning a sign language may lead to a cognitive ness. The Snijders-Oomen Nonverbal Intelligence Test
advancement in hearing children. (SON) and the Goodenough-Harris Human Figure
Drawing Test were administered both to deaf children
Sign languages used by deaf people employ sophisti- with deaf parents or deaf siblings (suggesting genetic
cated ways of representing space. Do signers develop deafness) and to deaf children with hearing parents and
other, nonlinguistic, visual-spatial abilities as a result of hearing siblings. Zweibel found that, in both tests, deaf
children with deaf parents scored significantly higher
This research was supported by funds from CNR Progretto Finalizzato
than deaf children with hearing parents but deaf sib-
"FATMA" and CNR Targeted Project "Beni Culturali." We thank Anna
Atbertoni and Barbara Grosso for assistance in data collection; Jana M. lings, according to scores on the SON and Figure
Iverson, Elena Pizzuto, and Judy Reilly for assistance with the English
Drawing tests. Furthermore, the latter group did not
version of this article; the director and staff of the Villanova di Guidonia
(RM) School for collaboration; and the children and their families for differ from deaf children with all-hearing families
participating in this study. Correspondence should be sent to Olga Capi- (nongenetic deafness). The main conclusion of the
rci, Istituto di Psicologia CNR, Reparto di Neuropsicologia del Lin-
guaggio e Sordita, via Nomentana, 56, 00161 Rome, Italy (e-mail: study was that genetic background makes no difference
volterra@kant.irmkant.rm.cnr.it). in intelligence. Zweibel suggested that these results
© 1998 Oxford University Press were best interpreted in terms of manual communica-
136 Journal of Deaf Studies and Deaf Education 3:2 Spring 1998

tion use in the home, which increased the ability of in which the two modes were separated and presented
deaf children to absorb messages and stimuli, thus as two different languages were introduced only when
leading to subsequent enhancement in cognitive devel- the children had acquired a basic ASL vocabulary.
opment. The aim of this article is to evaluate the effects of
Finally, a recent study by Parasnis, Samar, Bettger, sign language instruction in hearing children. Specifi-
and Sathe (1996) compared deaf nonsigning children cally, we hypothesize that hearing children's experience
with hearing controls on five tests that measured visual with sign language in the early school years may en-
spatial skills. Deaf and hearing children did not differ hance performance in the domain of nonverbal cogni-
in their performance, suggesting that exposure to sign tive skills such as visual perception, visual discrimina-
language and not to deafness itself determines differ- tion, and spatial memory. Two studies are reported
ences in visual spatial skills. here. The first study describes an educational experi-

Downloaded from https://academic.oup.com/jdsde/article/3/2/135/393210 by guest on 31 October 2023

Few studies have focused on the use of sign lan- ence lasting two years with a group of hearing children
guage by normally developing hearing children. In attending a sign language class in first and second
most cases sign language has been used with hearing grade matched to a group of hearing children not
children exhibiting particular pathologies such as exposed to sign language. All of the children were
Down syndrome, (Acosta, 1981) or autism (Konstant- given the Raven PM 47 test that measured visual-
areas, 1984; see Bonvillian & Miller, 1995, for a recent spatial skills at four time points. This study investi-
review of Sign Communication Training with mentally gated whether the performance of children who were
retarded children). Research in this area has shown that exposed to sign language was different from the perfor-
the use of signs improves the communicative skills of mance of hearing children not exposed to sign language
these children. on this test of visual-spatial cognition.
A limited number of studies have been carried out The second study describes a similar educational
in situations where bilingualism comes naturally in a experience of teaching sign language to hearing chil-
family context: hearing children of deaf parents who dren attending the first grade. The children attending
acquire sign language together with spoken language. the sign language class were matched to two control
Often bilingual children show very rapid language de- groups: (1) hearing children enrolled in the same
velopment in both languages (Capirci, Montanari, & school but exposed to an English course; and (2) hear-
Volterra, in press; Griffith, 1985; Orlansky & Bonvil- ing children not exposed to any foreign language. All
lian, 1985; Prinz & Prinz, 1981). children were given the same test as in the first study
In particular, a study by Daniels (1993) shows that and a Corsi's block-tapping test that measured spatial
bilingual-bimodal children achieve higher scores on memory at two time points. The second study investi-
the Peabody Picture Vocabulary Test (PPVT), sug- gated whether the performance of children exposed to
gesting that knowledge of American Sign Language sign language was different from the performance of
(ASL) has a positive effect on the acquisition of En- children exposed to English and to children not ex-
glish by hearing children. posed to any foreign language on tests of visual-spatial
Only one study (Daniels, 1994) reports a pilot proj- cognition and spatial memory.
ect of teaching sign language to hearing children of
hearing parents in a school context. Daniels demon-
Study 1: Method
strated that preschoolers who learned sign language
showed a greater understanding of English vocabulary. Subjects. Twenty-eight children from two first-year ele-
At the end of the year, the children attending ASL les- mentary school classes participated in a longitudinal
sons achieved significantly higher scores in the PPVT two-year study. Half of the children attended a course
vocabulary comprehension test than their peers who in Italian Sign Language (LIS group), while the other
did not take part in the project. The teaching method half had no such experience (control group). The after-
employed was basically bimodal, supplying the corre- noon program in sign language was voluntary. All chil-
sponding sign for each word. Additional brief phrases dren in the class were given a choice of activities: music,
 Hearing Children Learning Sign Language 137

gymnastics, or sign language. Half the children (14) Table 1 Course outlines
chose to participate in the sign language program. The Outlines Description
remaining 14 participated in music or gymnastics pro- 1-year course
grams. The two groups came from families living in Name signs
the same neighborhood and of the same low-middle- Fingerspelling letters and syllables
Geometric blocks triangles, squares, circles,
class background, and with the exception of Sign Lan-
rectangles
guage class, they were enrolled in the same school pro- Number from 1 to 10
gram. All children (28) are from monolingual Italian- Geometrical figure
speaking families and they had no experience with deaf Colors primary
culture or sign language. At the beginning of the Animals
Family members father, mother, brother, and
course, the mean age of the children in the LIS group
sister

Downloaded from https://academic.oup.com/jdsde/article/3/2/135/393210 by guest on 31 October 2023

was 6.6 years, and the mean age of the children in the Meals objects and dishes
control group was 6.5 years. Narration comprehension of short fairy
tales and real life events
Pretended play individually and as a group
Procedure. The LIS course was held in the afternoon,
2-year course
one hour a week (for seven months in the first year and Fingerspelling first and last name
for eight months in the second year) on the school Number to 30
premises by a deaf teacher whose first language was Colors complementary, light, and dark
sign language. All of the lessons were video-recorded Family all members, e.g., grandmother,
uncle, cousin
and transcribed by an experimenter. Children in the
Geometrical drawing two or more figures spatially
LIS group also worked for an hour each week with a located
hearing teacher who had knowledge of LIS. Months and days
The educational experience is based on the follow- Picture story "The Balloon"
Pretend play e.g., "The doctor and the
ing methodological principles: (1) presenting LIS
patient"
through a native signer interacting with the children Fairy tales playing as a group
exclusively in this language; (2) offering children the
opportunity to experiment with LIS in familiar con-
texts; (3) never translating from one language to an- lution for each problem requires logical nonverbal rea-
other, only stimulating children to capture analogies; soning. Children's responses were scored by totaling
(4) improving the development of comprehension the number of correct responses.
skills, especially in the first stage; and (5) letting chil-
dren spontaneously develop their production skills.
Results
The LIS program lessons developed in the two
years of the course are outlined in Table 1. As shown in Figure 1, the two groups attained very
At the beginning and end of each of these two aca- similar scores in the first test, but considerable differ-
demic years, all of the children in our sample were ences had already emerged by the end of the first and
given the Raven PM 47 Test (Raven, 1949). This test the second years, with the children attending the LIS
measures visual perception and level of mental devel- course showing evident gains. While the line for the
opment. It consists of a series (36 color pictures) of in- children who did not attend the course rises fairly
creasingly difficult matrixes, each with one piece miss- steadily, the graph for the LIS group shows no such
ing. The subject must select the correct piece to stability, reaching maximus corresponding to the two
complete the matrix from six alternatives. For each end-of-course tests and a stationary state during the
trial, a single stimulus picture is presented above the summer break.
six response-choice pictures. The subject's task is to In Figure 1 two dotted lines represent the 50th and
point to the one picture that fits in the stimulus picture. 90th percentile scores per age obtained by normative
There is no time limit for responding. The correct so- values of a sample of French children (Bourdier, 1964).
138 Journal of Deaf Studies and Deaf Education 3:2 Spring 1998

-•— us
-O— Control

••«— N . C 90°

••«— N.C 50°

Downloaded from https://academic.oup.com/jdsde/article/3/2/135/393210 by guest on 31 October 2023

Begin. 1st yr End 1st yr Begin. 2nd yr End 2nd yr

N.C. 90° a Nomutive Control 90° percentile scores

N . C 50° a Normative Control 50° percentile scores

Figure 1 RAVEN PM 47. Mean scores in the four trials for the two groups.

Table 2 Raven PM 47, mean scores and standard deviations for the two groups for
each year
Group Beginning 1st yr End 1st yr p Beginning 2nd yr End 2nd yr p
LIS 16.0 (3.5) 21.4(3.8) .003* 21.6(4.0) 24.6 (4.3) .01**
Control 15.7(3.0) 16.9 (4.6) NS 19.2 (5.0) 20.9 (5.8) NS
•Highly significant.
"Significant.

These norms were developed from 784 children from within-subjects factor. The dependent variable was the
different cities in France. In Figure 1 we present four raw score of correct responses. This analysis yielded
time points for the following ages: 6.6; 7.0; 7.6; 8.0. As significant Group, F{\, 26) = 4.179, MSE = 48.71,
is evident, the control children's performance on the p < .05, and Trials, F{\ 78) = 26.886, MSE = 8.67,
Raven test fell within the 50th percentile score of the p < .0001, effects, and the Group X Trials interaction
French sample across all of the time points. Children approached significance, F{Z, 78) = 2.68, MSE = 8.67,
who attended the LIS course were initially in the 50th p = .053. The main effect for Group reflected higher
percentile, like the control group, but their mean per- performance at test for the LIS group, and the main
formance was close to the 90th percentile at the end of effect for Trials reflected increasing scores across trials
both years of the LIS course. for both LIS and control subjects.
We calculated the mean scores attained by the LIS We also examined differences between the two
group and by the Control group at both administra- groups for individual trials. Group t tests showed no
tions of the Raven PM 47 (Table 2). To determine significant group differences in performance on the
whether the two groups differed in performance from first trial (/(26) = .233, ns), as expected since our inten-
the beginning of the first year to the end of the second tion was to have similar groups. On the second trial,
year, a 2 X 4 analysis of variance (ANOVA) was con- LIS group's performance was significantly better than
ducted, with Group (LIS course and no course) as the that of the control group (/ (26) = 2.76, p < .01). On
between-subjects factor and Trials (time point) as the the third trial there was a trend toward better perfor-
 Hearing Children Learning Sign Language 139

mance in the LIS group (f(26) = 1.38, p = .09). How- Procedure. The LIS course lasted five months during
ever, this may be due to the fact that the control group the first grade with the same procedure as Study 1.
caught up between the end of the first year LIS course The English course was held in the afternoon two
and the beginning of the second year LIS course (about hours a week. Before the beginning and soon after the
4—5 months). Finally, on the fourth trial, at the end of last lesson of the course, all the children in our sample
the second year, performance was again significantly were given two tests: the Raven PM 47 Test (Raven,
different in the expected direction (f(26) = 1.94, 1949), as in Study 1, and the Corsi's block-tapping test
to examine visual and spatial memory (Corsi, 1972; for
These results indicate that exposure and participa- Italian data see Orsini, Grossi, Capitani, Laiacona,
tion in a sign language program enhances nonverbal Papagno, & Vallar, 1987).
cognitive development. The data also show that the In the Corsi test, nine identical small white cubes

Downloaded from https://academic.oup.com/jdsde/article/3/2/135/393210 by guest on 31 October 2023

control group caught up to some extent over the sum- ( 4 X 4 X 4 cm) are arranged irregularly on a small
mer vacation, while the experimental group shows a wooden board (26 X 32 cm). The sides of cubes facing
performance plateau. This plateau suggests that the ac- the examiner are numbered from 1 to 9. The examiner
celerated growth in nonverbal cognition was strictly re- taps a certain number of blocks (digits) in a particular
lated to the sign langage course. sequence, and the subject is required to tap out the
same pattern immediately afterwards. The test begins
with a sequence of two units and then increasingly long
Study 2: Method sequences are presented; each time a maximum of five
equal unit sequences is tapped out. In the present ar-
The results of the first study led us to conduct a second ticle, we used the procedure suggested by Orsini (1994)
experiment to determine whether the acquisition of and Orsini, Maggiore, and Rotondaro (1996) with the
LIS or the exposure to a second language enhances vis- variation that the test was administered with sequences
ual discrimination and recognition of spatial relations. of from 2 to 5 digits (for a total of 20 items). The child's
To address this possibility, we added as control a group score is the number of items correctly reproduced.
of children attending an English course as second
language. In addition, in order to explore whether sign
language per se has specific effects on children's ability Results
to process and memorize visual stimuli trhough space,
we included an additional task that tapped short-term Data from individual subjects on each of the two tests
spatial memory. were analyzed to determine whether group differences
existed in performance at the beginning and the end of
the academic year.
Subjects. The 49 first-grade children who participated
in this study came from three classes in the same
school. All 20 children from one class attended a course Raven PM 47
in LIS, the 20 children of the other class attended
a course in English language, and the remaining 9 Figure 2 and Table 3 show the mean scores obtained by
children had no second language exposure. The three the three groups of children at the beginning (first
groups came from families living in the same neighbor- trial) and at the end (second trial) of the school year.
hood and of the same low-middle-class background. It can be seen that on the first trial, the three groups
With the exception of Sign or English Language class, performed similarly, with no difference among means.
they were enrolled in the same school program. All On the second trial, all groups showed an increase in
children (49) are of monolingual Italian-speaking fami- their Raven's score, with children attending the LIS
lies and they had no experience with deaf culture or course showing a clear gain. The 50th and 90th percen-
sign language. tile per age obtained from normative data available on
140 Journal of Deaf Studies and Deaf Education 3:2 Spring 1998

us
English

Control
N.C 90°
N.C. 50°

Downloaded from https://academic.oup.com/jdsde/article/3/2/135/393210 by guest on 31 October 2023

Begin. 1st yr End 1st yr

N . C 90° = Normative Control 90° percentile scores

N . C 50° = Normative Control 50° percentile scores

Figure 2 RAVEN PM 47. Mean scores in the two trials for the three groups.

Table 3 Raven PM 47, mean scores and standard significant differences between the LIS group vs. the
deviations for the three groups
English and the control group, f\l, 46) = 13.33, p <
Group Beginning 1st yr End 1st yr P .01, and no difference between the English group and
LIS 18.3 (5.3) 22.2 (5.2) .001** the control group, F(\, 46) < l,ns.
English 17.5(4.1) 18.8 (3.8) NS
Control 17.1(3.4) 18.9 (2.9) NS
**Highly significant. Corsi's Block-Tapping Test

The data in Figure 3 and Table 4 show that children

this test on French children are also shown in Figure 2 who attended the LIS course and control children were
(Bourdier, 1964). equivalent on the first trial, with no difference among
As it is apparent in the figure, performance on the means. Performance in both groups increased on
Raven PM 47 by children enrolled in the English the second trial, with children who attended the LIS
course and by the control children was close to the 50th course showing again a clear gain. Children who at-
percentile score in both observations. Children who at- tended the English course showed no such enhance-
tended the LIS course performed similarly to the other ment, but equal performance across the two trials.
children on the first trial. The same children at the end A 3 X 2 ANOVA was carried out with Group (LIS
of the course performed better, approaching the 90th course, English course, and no course) as the between-
percentile of the French sample. subjects factor and with Trials (beginning vs. end of
A simple effects analysis was carried out on the data course) as the within-subjects factor. The dependent
from the first trial to determine whether significant variable was the number of correct responses. The
differences existed among the groups (LIS course, En- analysis yielded a significant main effect and a Group
glish course, and no course). This analysis revealed no X Trial interaction that approached significance,
significant difference, f\2, 46) < 1, ns. In contrast, on F(Z, 46) = 2.66, MSE = 4.51,/) = .08. The main effect
the second trial, there was a significant difference for Trials, F{\, 46) = 8.13, MSE = 4.51, p< .01, indi-
among the groups, f\2, 46) = 3.4, p < .05. Pairwise cated a higher performance at the end of academic year.
comparisons carried out on the second trial revealed A simple effects analysis indicated that this interaction
 Hearing Children Learning Sign Language 141

us
English

Control

Downloaded from https://academic.oup.com/jdsde/article/3/2/135/393210 by guest on 31 October 2023

Begin. 1st yr End 1st yr

Figure 3 CORSI SPAN. Mean scores in the two trials for the three groups.

Table 4 Corsi span, mean scores and standard deviations In our first study, the finding that hearing children
for the three groups attending the LIS course performed better relative to
Group Beginning 1st yr End 1st yr P control children and in advance of their chronological
LIS 11.5(3.8) 14.1 (3.4) .001** norms on the Raven PM 47 is consistent with the idea
English 12.8 (3.7) 12.8(3.1) NS that experience with sign language enhances nonverbal
Control 11.2(1.8) 12.6 (3.5) NS
cognitive skills such as visual and perceptual discrimi-
**Highly significant. nation and recognition of spatial relations (Emmorey,
Kosslyn, & Bellugi, 1993). Attendance of the LIS
was solely due to an increase in the performance of course appears to promote faster development in non-
children who attended the LIS course, F\\, 46) = verbal cognition: children learning LIS reach the level
14.99, MSE = 4.51, p<. 001. achieved by their schoolmates not attending the LIS
course almost one year earlier.
In the second study, we examined whether the ac-
Conclusion
quisition of LIS or the exposure to a second language
The results of the educational experience reported enhances visual discrimination and memory of spatial
here show that hearing children who learn sign lan- relations. We found that children who attended the
guage as a second language in the early school years LIS course showed enhanced performance on the Ra-
improve more rapidly on tests of visual-spatial cogni- ven PM 47 compared to controls, that is, to children
tion and spatial memory than their schoolmates not at- attending an English course and to chronological
tending a sign language course. norms. Similar findings were obtained on the Corsi's
Our findings are consistent with previous studies block-tapping test.
reported on signing (Bellugi et al, 1990; Zweibel, In summary, these results suggest that exposure to
1987) and nonsigning (Parasnis et al., 1996) deaf chil- sign language among hearing children may be an im-
dren. Specifically, our results support the position that portant factor in the enhancement of visual-spatial
it is exposure to a visuo-gestural language per se that is cognition.
responsible for enhanced visual-spatial abilities. In the present article our focus was on cognitive
142 Journal of Deaf Studies and Deaf Education 3:2 Spring 1998

improvement as a consequence of sign language teach- Chovan, J. D , Waldron, M. B., & Rose, S. (1988). Response la-
tency measurements to visual cognitive tasks by normal
ing. While the data concerning sign language acquisi- hearing and deaf subjects. Perceptual and Motor Skills, 67,
tion by the same group of children are reported else- 179-184.
where (Capirci, Cattani, Rossini, & Volterra, 1997), we Corsi, P. M. (1972). Human memory and the medial temporal region
would briefly like to mention here only the major goals of the brain. Unpublished Ph.D. dissertation, McGill Uni-
versity.
achieved through this educational experience in the Daniels, M. (1993). ASL as a possible factor in the acquisition of
linguistic domain. English for hearing children. Sign Language Studies, 78,
Hearing children attending the LIS course showed 25-30.
Daniels, M. (1994). Words more powerful than sound. Sign Lan-
an increasing interest in sign as an alternative to spoken
guage Studies, 83, 155-166.
communication, reached a basic competence in LIS, Emmorey, K., Kosslyn, S. M., & Bellugi, U. (1993). Visual imag-
and displayed a new, spontaneous sensitivity towards ery and visual-spatial language: Enhanced imagery abilities

Downloaded from https://academic.oup.com/jdsde/article/3/2/135/393210 by guest on 31 October 2023

the culture and communicative modes of someone in deaf and hearing ASL signers. Cognition, 46, 139-181.
Griffith, P. (1985). Mode switching and mode finding in a hear-
differing from them: they always communicated with
ing child of deaf parents. Sign Language Studies, 48,
their deaf teacher through the appropriate visual- 195-222.
gestural modality rather than in speech. Konstantareas, M. (1984). Sign language as a communication
prothesis with language impaired children. Journal ojAutism
These findings suggest that it would be extremely
and Developmental Disorders, 14(1), 9-25.
useful to offer sign language as a second language to Orlansky, M., & Bonvillian, J. (1985). Sign language acquisition:
hearing children for linguistic as well as cognitive Language development in children of deaf parents and im-
reasons. plications for other population. Merrill-Palmer Quarterly,
31(2), 127-143.
Orsini, A. (1994). Corsi's block-tapping test: standardization and
References concurrent validity with Wisc-R for children aged 11 to 16.
Perceptual and Motor Skills, 79, 1547-1554.
Acosta, L. (1981). Instructor use of total communication: Effects of Orsini, A., Grossi, D., Capitani, E., Laiacona, M., Papagno,
preschool Down's syndrome children's vocabulary acquisition and G., & Vallar, G. (1987). Verbal and spatial immediate mem-
attempted verbalizations. PhD dissertation, University of ory span: normative data from 1355 adults and 1112 chil-
Iowa. dren. Italian Journal of Neurological Sciences, 8, 539-548.
Bellugi, U , O'Grady, L., Lillo-Martin, D , O'Grady, M., van Orsini, A., Maggiore, B., & Rotondaro, C. (1996). II test di Corsi
Hoek, K., & Corina, D. (1990). Enhancement of spatial cog- come tredicesimo subtest della Wisc-R. Bollettino di Psico-
nition in hearing and deaf children. In V. Volterra & C. Ert- logia Applicata, 217, 35-44.
ing (Eds.), From gesture to language in hearing children and Parasnis, I., Samar, V, Bettger, J. G., & Sathe, K. (1996). Does
deaf children. New York: Springer Verlag. deafness lead to enhancement of visual spatial cognition in
Bonvillian, J. D., & Miller, A. J. (1995). Everything old is new children? Negative evidence from deaf nonsigners. Journal
again: Observations from the nineteenth century about sign ofDeaf Studies and Deaf Education, 1(2), 145-152.
communication training with mentally retarded children. Prinz, P., & Prinz, E. (1981). Acquisition of ASL and spoken
Sign Language Studies, 88, 245-254. English by a hearing child of a deaf mother and a hearing
Bourdier, G. (1964). Progressive matrices (1947). Sets A, Ab, B. father: Phase II, early combinatorial patterns. Sign Language
Board and Book Forms. London: Lewis. Studies, 30, 78-88.
Capirci, O., Cattani, A., Rossini, P., & Volterra, V. (1997). La Raven, J. C. (1949). Progressive matrices (1947). Sets A, Ab, B.
lingua dei segni come seconda lingua nella scuola ele- Board and Book Forms. London: Lewis.
mentare. Psicologia clinica dello sviluppo, 1(2), 301-311. Zweibel, A. (1987). More on the effects of early manual commu-
Capirci, 0., Montanari, S., & Volterra, V. (in press). Gestures, nication on the cognitive development of deaf children.
signs, and words in early language development. In J. Iver- American Annals of the Deaf, 132, 16-20.
son & S. Goldin-Meadow (Eds.), Gesture-speech relationships
in childhood. San Francisco: Jossey-Bass.