lournal of Autism and Developrnental Disorders, VoL 11, No.

1, 1981

Sign Language and Autism

John D. Bonvillian
University of Virginia

Keith E. Nelson
The Pennsylvania State University

Jane Milnes Rhyne

University of Virginia

Research findings and issues in teaching sign language to nonspeaking

autistic children are reviewed. Data on over 100 children indicate that nearly
all autistic children learn receptive and expressive signs, and many learn to
combine signs. These children also exhibit marked improvement in adaptive
behaviors. Speech skills are acquired by fewer children and may be de-
veloped through simultaneous speech and sign training. Possible explana.
tions f o r these results are given, together with suggestions f o r future
research and data collection. Recommended innovations include exposure
to fluent signers and training in discourse and code-switching. Different
sign language teaching methods need to be investigated more fully, includ-
ing emphasis on training sign language within the children "s total environ-
ment and with greater staff and parental participation.

Disturbances in language and symbol-processing skills are among the most

salient characteristics of childhood autism (Churchill, 1978). Frequently the
lowest functioning autistic children have essentially no recognizable speech,
being either mute or limited to a small number of vocalizations. Some
autistic children exhibit echolalia, while others experience delay in language
acquisition, inappropriate use of language, or a defect in higher levels of
language functioning (Rutter, 1978). Because alleviating these deficits
would probably do much to enhance the children's development and adjust-
ment, much of the recent work with autistic children has focused on
training language skills.

125
0162-3257/81/0300-0125503.00/0 9 1981 Plenum Publishing Corporation
126 Bonvillian, Nelson, and Rhyne

During the past 15 years, language and communication training pro-

grams for autistic children with severe language deficits have been strongly
influenced by two major innovations in methods. The first, and more wide-
spread, is the behavior modification approach to speech training. The
second approach employs sign language, either alone or in combination
with speech. These approaches need not be considered as mutually exclu-
s i v e - m a n y investigators who use sign language training employ behavior
modification procedures as part of their language training, and future
complementary work can be expected. Other new programs do not appear
to have been as widely adopted.
Among the most systematic behavior modification projects have been
those conducted by Lovaas and his associates. Lovaas (Lovaas, 1977, 1978;
Lovaas, Koegel, Simmons, & Long, 1973) and others have documented the
efficacy of behavior modification programs in increasing speech through
positive reinforcement, and in decreasing inappropriate behaviors (e.g.,
tantrums, echolalia) through negative reinforcement or punishment. For
those autistic children who are echolalic, the behavior modification task
initially focuses on teaching the child to discriminate between those words
he should imitate and those he should not. With nonspeaking autistic
children, the initial emphasis is on teaching the child to imitate speech.
Overall, the behavior modification approach to training speech has proved
to be much more successful with echolalic than with mute autistic children
(Lovaas, 1977).
Despite the many substantial achievements obtained using the
behavior modification speech imitation approach, one telling criticism has
been the claim that mute autistic children who participated in such
programs only rarely learned to produce spontaneous rule-governed utter-
ances (Churchill, 1978; Schaeffer, Kollinzas, Musil, & McDowell, 1977).
Too often the final result of these programs was that the children learned to
repeat set phrases that were trained but failed to master the rules enabling
them to create new meaningful strings of words from their available lexi-
con. Nor did the children often employ their programmed language skills in
new, nonlaboratory locations. These speech imitation training programs
have been very time-consuming, sometimes entailing thousands of trials in
order to establish only minimal phonological or grammatical distinctions.
Finally, in order to maintain their progress, the children have frequently
required continued treatment in a highly structured environment.
The sign language programs are of more recent origin, but already
there are more than 20 studies reporting at least some success in fostering
communication skills in severely language-disturbed autistic children.
Actually, several different sign systems have been employed. Two of the
more commonly used systems consisted of (1) taking the signs of the lan-
guage of the deaf community, American Sign Language, and putting the
Sign Language and Autism 127

signs in English word order; and (2) using SEE or Signing Exact English
(Gustason, Pfetzing, & Zawoklow, 1972) signs, a system that attempts to
duplicate English syntax and morphology in the manual mode. The sign
language input in most cases has been augmented by the teacher or ther-
apist's simultaneous use of spoken English.
These studies of autistic children's acquisition of sign language have
demonstrated the strong potential effectiveness of manual communication
programs. At the same time, a number of essential studies have been left
undone. As just one example, in spite of the now widespread use of both
sign language and speech imitation procedures with autistic children, no
study has attempted to evaluate the language learning process in the two
modalities in carefully selected, comparable populations. The few small-
scale studies to date (e.g., Brady & Smouse, 1978; Leibovitz, Note 1; Baron
& Isensee, Note 2) have favored sign language or combined sign and speech
training over speech-only training, but longitudinal investigations of
language learning by many subjects in each of the different modes are
needed.
In the remainder of this paper, we shall review the findings of the
studies of sign language acquisition in autistic children.Our discussion of
these studies is focused on five major areas. (1) First is the pattern of sign
language acquisition, and how this pattern compares with the initial stages
of language acquisition in deaf and normal children. (2) Next, we examine
the progress made by some of the sign-trained children in learning to speak,
together with an analysis of those factors that appear to influence their
speech and sign language development. (3) Analysis of changes in the ap-
proaches to teaching manual communication is then followed by (4) a con-
sideration of discourse skills. (5) We c69clude the paper by discussing pos-
sible reasons why sign language programs are not more widely adopted.

SUMMARY OF RESEARCH FINDINGS

About 100 autistic subjects have participed in the 20 studies on sign

language or simultaneous sign and speech communication reviewed in this
report. The average initial age for the children was approximately 8 years,
with ages ranging from 3 89 to 23 years. In more than half of the studies the
subjects received some type of prior language training, with 9 studies in-
dicating unsuccessful attempts at teaching speech through operant pro-
cedures. Ability levels of subjects were most often reported in descriptive
terms and anecdotal accounts, rather than through the use of more ob-
jective measures. Subjects were described as mute, as echolalic, as pro-
ducing unintelligible vocalizations, or as otherwise limited in speech. In
addition to autistic, terms such as retarded, hyperactive, and withdrawn
128 Bonvillian, Nelson, and Rhyne

were used to describe subjects. Their behavioral profile indicated low

levels of self-care and high levels of self-stimulation, tantrums, and self-
destructive behaviors. Almost one-third of the subjects were institution-
alized, while the others remained at home and attended day school pro-
grams.
With a few exceptions (Leibovitz, Note !; Schaeffer et al., 1977;
Baron & Isensee, Note 2), input in sign alone was not provided. Instead,
simultaneous communication was employed. Many investigators also
directly employed operant techniques for reducing the frequency of inap-
propriate behaviors, and some employed additional components such as
auditory discrimination training, intrusion play, and imitation of body
movement (Benaroya, Wesley, Ogilvie, Klein, & Meaney, 1977), or activ-
ities to enhance body experience and intentional acts (Miller & Miller,
1973). Duration and intensity of treatment programs differed dramatically.
Results of these studies indicate that even fairly brief simultaneous
communication or sign language training can be an effective means of im-
proving communication skills in low-functioning autistic children. Despite
an extensive range in individual outcomes, almost every subject acquired
the ability to comprehend trained signs. A large majority also mastered pro-
duction of 5 or more signs, and maximum learning exceeded 350 signs. Pro-
ductive combinations were less frequently observed. Nevertheless, about
two-fifths of the children produced spontaneous combinations at least
occasionally, and in many cases the children moved to daily production of
many complex sign utterances. Comparable improvements in speech skills
were not as frequent (e.g., Bonvillian & Nelson, 1976; Webster, Konstan-
tareas, & Oxman, Note 3); many subjects acquired at least a limited spoken
vocabulary, while a smaller number also showed improvements in their
semantic and syntactic skills (e.g., Schaeffer et al., 1977; Yamada, Kriend-
ler, & Haimsohn, 1979; Creedon, Note 4). Communication skills were
generalized to adults besides the original trainers, to other environments
besides the training situation (e.g., Casey, 1978; Salvin, Routh, Foster, &
Lovejoy, 1977; Schaeffer et al., 1977; Webster, McPherson, Sloman,
Evans, & Kuchar, 1973), and for purposes not originally taught--for
example, to direct one's own behavior or to comment to oneself (Schaeffer
et al., 1977).
Overall, the level and rate of mastery for the subjects learning signs
appears to be greater than that for speech in comparable speech-oriented
operant programs. Subjects who had previously participated in speech-
oriented programs without success frequently demonstrated the ability to
communicate in signs after exposure to manual communication (e.g., Bon-
villian & Nelson, 1976; Carr, Binkoff, Kologinsky, & Eddy, 1978; Salvin et
al., 1977). Failure to grasp speech when it alone was used contrasts with
success in understanding signs when they were used as the sole mode of
Sign Language and Autism 129

communication (Baron & Isensee, Note 2; Dores & Carr, Note 5). Further-
more, signs were used as aids to communication after speech had finally
been exhibited. In one study, signs apparently were employed to assist
subjects in maintaining a train of thought when they were using only speech
(Schaeffer et al., 1977), and in another study the child spontaneously
produced signs as possible mediators for speech (Baron & Isensee, Note 2).
Because of the variations in training procedures and the diversity of
subjects and their individual achievements, we must wait for new studies to
tell us firmly what specific factors are related to eventual level of language
mastery. However, several trends in the data stand out. In general,
when children are relatively young and when program participation is rela-
tively lengthy, more progress is made. Not surprisingly, a child's ability
level prior to participation in a program is also an important predictor.
Children who receive low scores on indices of autistic symptom severity (in
other words, higher functioning autistic children) are more likely to make
substantial progress (Miller & Miller, 1973). However, the very low-func-
tioning autistic children do not appear to make as rapid progress. The
child's ability to imitate speech also has been positively associated with
eventual levels of language mastery, in both sign and speech (Carr, 1979;
Blascoe, Note 6). Finally, several investigators have observed informally
that those subjects who tended to signify their pretreatment desires through
pointing made good candidates for a manual language program.

PATTERNS OF SIGN LANGUAGE ACQUISITION

Autistic children sometimes comprehend more signs than they

produce spontaneously, just as most young normal and deaf children have
larger receptive than productive vocabularies. There is a resemblance also in
the content of the early vocabularies: autistic children's early vocabularies,
like normal children's, consist predominantly of concrete nouns and action
verbs. One difference is that although autistic children's vocabularies
typically continue to grow with increased training and exposure, they do not
grow nearly as quickly as those of young normal children learning to speak
or those of deaf children acquiring sign language. Early learning is fre-
quently very slow until the autistic child, as Carr et al. (1978) notes, seems
to acquire a "set"--that objects and actions have manual labels. After this,
progress is sometimes rapid.
In reporting the size of their autistic subjects' sign language vocabu-
laries, experimenters usually have expended little effort in probing the
subjects' knowledge of the signs they may have mastered--spontaneous
production of an object's or concept's name seems to have implied compre-
hension, and in other cases comprehension only has been trained and
130 Bonvillian, Nelson, and Rhyne

assessed. Moreover, it does not appear that experimenters often provide a

range of exemplars of a concept in their training and testing, making it
difficult to determine the semantic boundaries of the children's "mastered"
signs.
With some autistic children, when sign combinations are first gener-
ated, they appear to resemble structurally the early word combinations of
young normal children learning to speak. It is apparent that some of these
children also learn minimal rules for sign combinations (Carr & Kologinsky,
Note 7). However, after the initial success o f these early sign combinations,
most autistic children make only rather limited progress in terms of the
average length and complexity of their sign utterances. Occasionally, these
children will produce very long and complex sign sentences, indicating that
they have the conceptual and memory skills necessary for complex produc-
tions, but these productions generally appear infrequently.
Unfortunately, the absence of detailed records of most of the
children's sign language combinations makes it impossible to determine for
fairly fluent children whether there is sufficient regularity of syntax or com-
prehension of complex semantic aspects in the children's sign utterances to
credit them with these fundamentals of language. Moreover, to judge these
issues, reliable information also will have to be collected concerning the
syntactic structure (and consistency) of signed input utterances. What inves-
tigators need to do is move beyond the widespread current practice of
focusing primarily on vocabulary size, and instead include records of the
children's combinations together with the input context and nonlinguistic
context in which the sign utterances are produced. Such records should
enable us to discuss what semantic relations autistic children frequently
encode, as well as the pragmatic aspects of their language. In addition, these
records would be helpful in determining which training strategies are most
effective with different autistic children. Finally, by emphasizing the
children's sign combinations, we might be able to note specific higher order
language defects that only become apparent when complex utterances are
produced. Identification of such higher order difficulties may then, in turn,
lead to more sophisticated language training tailored to the individual
child.

FACTORS INFLUENCING SPEECH AND SIGN ACQUISITION

Many autistic children who learn initially to communicate via manual

signs also make progress in acquiring spoken language skills. For most of
these children, it appears that their teachers or therapists produce signs in
English word order while simultaneously saying the words. Apparently,
signing and speaking at the same time does not exceed the processing limits
Sign Language and Autism 131

of these children. Rather, the two input modes may function in a redundant
manner, with remembered signs being used to cue forgotten words and vice
versa (Schaeffer et al., 1977). A second possible explanation for the effec-
tiveness of simultaneous communication in facilitating speech is that the
signs may help the child distinguish where a word starts and stops (Yamada
et al., 1979). Moreover, when words are difficult to say or similar in sound,
the child might be better able to distinguish among the words in simul-
taneous communication. For example, the child probably would not sign
"This is a hat" when he should sign "This is a cat" (Yamada et al., 1979),
and the salient differences between the signs may help the child keep
straight the more subtle acoustical differences between " h a t " and " c a t "
and other related words in speech. A related possibility is that autistic
children might come to perceive words in terms of phonemes only after they
learn some correlations between hand and voice demonstrated by their
teachers.
When some nonspeaking autistic children trained through
simultaneous communication fail to make appreciable gains in speech while
concurrently acquiring communication skills in sign language, this single-
mode progress may result from overselective attention to multiple stimuli
(Lovaas & Schreibman, 1971; Lovaas, Schreibman, Koegel, & Rehm, 1971).
When speech is paired with gestures, these children may respond to the
visual stimuli provided by gestures and not to the auditory stimuli.
Because sign language is a visual-motor communication system, it may
bypass many of the difficulties that autistic children have with auditory-
vocal processing. The motor, kinesthetic, and visual systems in autistic
children appear to be relatively more intact than their auditory-vocal sys-
tems (O'Connor, 1971). One may also account for the children's mastery of
signs by noting the ease with which teachers can mold the child's fingers and
guide his hand through the appropriate movements (Bonvillian & Nelson,
1978, in press). As the child is instructed in signs, visual feedback is avail-
able to him as he is able to see both his hand movementsand those of his
teacher. Furthermore, signs can be slowed and stopped, thus providing the
child more time to study the specific hand configurations involved. The fact
that a number of signs are iconic, in that they resemble the actions or
objects for which they stand, may also assist the child in his learning. This
observable tie between the sign and its referent might be especially helpful
to children with symbolic processing difficulties, especially during the initial
stages of language acquisition. Although iconicity is not an essential aspect
since autistic children learn many noniconic signs, the iconic signs appear to
be easier to learn and to constitute a considerable portion of sign-trained
autistic children's early vocabularies (Konstantareas, Oxman, & Webster,
1978; Schwam, in press; Brown, Note 8). Finally, the success of manual
communication with these children may stem from differences in neuro-
132 Bonvillian, Nelson, and Rhyne

logical processing of a visual-motor language. By using a visuomotor lan-

guage, such as one of the sign languages, the children may be able to take
advantage of less damaged areas of their left hemisphere (cf. Hauser,
DeLong, & Rosman, 1975) or relatively intact areas of their right. A related
point is that right hemispheric processing of signs has previously been noted
in some deaf and hearing subjects (McKeever/Hoemann, Florian, & Van
Deventer, 1976; Poizner & Battison, in press; Poizner & Lane, 1979).
As noted earlier, speech-oriented operant programs typically are more
successful with echolalic children than with mute children. There is some
evidence (e.g., Blascoe, Note 6) indicating that simultaneous sign and
speech programs can lead to decreased echolalia and increased auditory
comprehension, together with enhanced communication skills in both
modes. However, it is appropriate to point out that one of the few cases in
which sign language training proved ineffective was with an echolalic child.
After failing to progress from a limited repertoire of receptive signs to spon-
taneous expressive signing, this child made rapid progress in a speech-
oriented operant program, evidencing both improved language and
improved social behavior (Konstantareas, Oxman, & Webster, 1977). In
this particular case, the subject's echolalic verbalizations may have inter-
fered with his signing; his mastery of signs and of speech might possibly
have been more expeditious had the signs and speech each been presented
alone rather than together.

TEACHING OR TRAINING STRATEGIES

An essential, unresolved question is what would be the most effective

sign language teaching strategy to employ with nonspeaking children?
Many of the sign language training programs have been embedded within a
behavior modification framework, at least in their early stages, in that the
experimenters in these studies reinforced correct signing or approximations
to signs by food, tokens, or praise. Furthermore, the sign training was often
conducted in distinct language sessions that constituted only a small portion
of a subject's daily or weekly schedule. With this approach, object
exemplars or pictures of referents were often present when a sign was
initially being taught. Yet another strategy for teaching signs, which has
been explored but which has not been adequately tested, would be to do
nearly all sign-teaching in context throughout the course of the child's day,
focusing on the child's apparent desires or interests from moment to
moment and situation to situation. This approach would appear to mirror
more closely the normal child's pattern of language development and might
result in more spontaneous signing by the child (Schaeffer, 1978).
Sign Language and Autism 133

A second aspect of sign language training that needs to be considered

more fully is whether to combine sign input with spoken language. Most
sign language programs have relied on simultaneous language presentation
in both modes, with the reason frequently given for this procedure being
that it might facilitate the nonspeaking child's transition to speech. How-
ever, there are few data to support this widely held view, and compelling
arguments can be made that such a procedure might hinder some autistic
children (above section; Bonvillian & Nelson, 1978; Leibovitz, Note 1).
Briefly, a number of autistic children not only react abnormally to speech
(Condon, 1975) but have special difficulties with bi- or multimodal process-
ing. For these children a program that focused on manual input alone might
lead to the more rapid development of communication skills.

DISCOURSE AND DISCOURSE THERAPY

From many points of view, observers have noted that the autistic child
needs to learn appropriate, contextually adjusted uses of language. For
example, investigators working within a social learning point of view have
recently begun to address theoretically how one can teach language skills in
a way that will generalize beyond the circumscribed training settings. Thus
Hart and Rogers-Warren (1978) describe a "milieu approach" for
"bridging" language use across different settings, Lovaas (1978) and
Schopler (1978) include both the family setting and the parent-as-practi-
tioner in their current programs, and Schiefelbusch (1978) comments that
"pragmatic behaviors are no doubt an essential part of effective communi-
cation and must be included in the design for training" (p. 10). And from a
much different perspective, Menyuk (1978) reviews language use limitations
and argues that the frequent tendencies for autistic children to display
echolalia, neologisms, and related responses may represent "the retrieval of
unanalyzed wholes in response to a particular situation" (p. 106).
We agree that discourse skills need to be an essential target for therapy
and education with many autistic children. Further, when children learn
sign language in combination with some receptive and productive skills in
spoken language, the child ideally needs to learn code-switching discourse
skills that allow the child to adjust to conversational partners who vary in
their awareness of and degree of mastery of sign language. For sign
language programs or any other variety of language programs, however, no
detailed and plausible blueprints for training discourse skills are yet avail-
able. Here we offer simply a few ideas about components that may prove
important in discourse training: (1) Even at very low levels of productive
sign language skill for the child, include contingent language training where
134 Bonviilian, Nelson, and Rhyne

adults seize opportunities as they arise, in any context, to engage in conver-

sation on events and topics that the chiM selects for attention. (2) In
focused, one-to-one, limited-time training sessions it may help for the child
to have sessions in more than one setting and with two or more trainers. (3)
Goals should include the child's skill in shifting linguistic encodings of the
same referents (objects, people, tools) when the situational meaning of the
referents shift. (4) Explicitly teach the child discourse routines. By adult
modeling and by direct instruction a child might learn to assess the com-
municative abilities in different modes of another person before attempting
to get messages across. (5) The child, particularly when sign language and
other nonspeech modes are employed, should have exposure to conversa-
tion in which two or more fluent communicators provide a display of
varied, smooth discourse (cf. Bonvillian & Nelson, 1978). (6) When the
child has an urgent message he is eagerly trying to communicate, initial
comprehension failure by an adult should lead to a strategy of patient,
imaginative attempts to help the child get his meaning across.

PARENTAL AND INSTITUTIONAL INVOLVEMENT

Parents of autistic children frequently are reluctant to give permission

for their nonspeaking child's participation in a sign language training
program. However, although speech progress might suffer, the data so far
show no such effect. Moreover, a larger percentage of nonspeaking, low-
functioning, autistic children might make the transition successfully from
sign communication to speech if clues about speech-specific deficits would
be taken seriously and parents would begin sign language training earlier in
their children's lives, rather than after years of failure in speech-oriented
programs. In addition, some authors have implied that motor difficulties in
language-limited children would commonly imperil sign language acquisi-
tion and that initial stages of learning a sign language would be so complex
that meaningful use of signs or sign phrases necessarily would be postponed
(e.g., Hollis & Carrier, 1978); both these implications appear to be mislead-
ing as applied to the majority of autistic children (see Alpert, 1980).
Another reason that sign language programs for autistic children have
not been more widely embraced rests with a misperception by institutional
staff members and the children's parents as to the difficulty of learning to
communicate manually. The children usually master only several new
signs each week, and most of their sign utterances are neither long nor struc-
turally complex. Furthermore, the actual physical production rate of the
signs is also much slower than that of deaf signers (Fristoe, Note 9). Because
of these factors, most staff members at institutions and parents of autistic
children would probably need to devote only a few minutes each week to
Sign Language and Autism 135

sign training. And in light of the findings that parental participation or

involvement can play an effective role in facilitating their autistic child's de-
velopment (Casey, 1978; Schopler & Reichler, 1971), steps should be taken
to convey to the parents the ease with which they might acquire many useful
elements of a manual communication system. At the same time, it should
also be remembered that the optimal effectiveness levels of sign language
programs are not likely to be reached until some teachers and parents
become fluent signers with each other and with the autistic children
(Bonvillian & Nelson, 1978).

REFERENCE NOTES

1. Leibovitz, S. Sign versus speech in the imitation learning o f a mute autistic child. Un-
published master's thesis, School of Human Communication Disorders, McGill University,
1976.
2. Baron, N. S., & lsensee, L. M. Effectiveness o f manual versus spoken language with an
autistic chiM. Unpublished manuscript, Brown University, 1976.
3. Webster, C. D., Konstantareas, M. M., & Oxman, J. Simultaneous communication with
severely dysfunctional nonverbal children: A n alternative to speech training. Unpublished
working paper in Child Development and Care, University of Victoria, 1976.
4. Creedon, M. P. Language development in nonverbal autistic children using a simultaneous
communication system. Paper presented at the biennial meeting of the Society for Research
in Child Development, Philadelphia, 1973.
5. Dotes, P., & Carr, E. G. Teaching sign language to psychotic children: Assessing simul-
taneous communication. Paper presented at the meeting of the American Psychological
Association, New York, September 1979.
6. Blascoe, M. Echolalia reduction and language acquisition through total communication
with an autistic child. Unpublished manuscript, Rockford Public Schools, Rockford,
Illinois, 1979.
7. Cart, E. G., & Kologinsky, E. Teaching psychotic children to use sign language: Develop-
ment o f descriptive generative sentences. Paper presented at the meeting of the Association
for the Advancement of Behavior Therapy, Chicago, November 1978.
8. Brown, R. Why are signed languages easier to learn than spoken languages? Paper pre-
sented at the National Symposium on Sign Language Research and Teaching, Chicago
1977.
9. Fristoe, M. Non-speech communication f o r handicapped persons using the visual modality.
Proceedings of the Adelphi University Conference on Total Communication for the
Severely Handicapped, October 1979.

REFERENCES

Alpert, C. Procedures for determining the optimal nonspeech mode with the autistic child. In
R. L. Schiefelbusch (Ed.), Nonspeech language and communication. Baltimore: Uni-
versity Park Press, 1980.
Benaroya, S., Wesley, S., Ogilvie, H., Klein, L. S., & Meaney, M. Sign language and multi-
sensory input training of children with communication and related developmental
disorders. Journal o f Autism and Childhood Schizophrenia, 1977, 7, 23-31.
Bonvillian, J. D., & Nelson, K. E. Sign language acquisition in a mute autistic boy. Journal
o f Speech and Hearing Disorders, 1976, 41, 339-347.
136 Bonvillian, Nelson, and Rhyne

Bonvillian, J. D., & Nelson, K. E. Development of sign language in autistic children and other
language-handicappedindividuals. In P. Siple (Ed.), Understanding language through
sign language research. New York: Academi c Press, 1978.
Bonvillian, J. D., & Nelson, K. E. Exceptional cases of language acquisition. In K. E. Nelson
(Ed.), Children's language (Vol. 3). New York: Gardner Press (Halsted/Wiley), in press.
Brady, D. O., & Smouse, A. D. A simultaneous comparison of three methods for language
training with an autistic child: An experimental single case analysis. Journal of Autism
and Childhood Schizophrenia, 1978, 8, 271-279.
Carr, E. G. Teaching autistic children to use sign language: Some research issues. Journal of
Autism and Developmental Disorders, 1979, 9, 345-359.
Carr, E. G., Binkoff, J. A., Kologinsky, E., & Eddy, M. Acquisition of sign language by
autistic children. I. Expressive labelling. Journal of Applied Behavior Analysis,
1978, ll, 489-501.
Casey, L. O. Development of communicative behavior in autistic children: A parent program
using manual signs. Journal of Autism and Childhood Schizophrenia, 1978, 8, 45-59.
Churchill, D. W. Language of autistic children. Washington, D.C.: V. H. Winston, 1978.
Condon, W. S. Multiple response to sound in dysfunctional children. Journal of Autism and
Childhood Schizophrenia, 1975, 5, 37-56.
Gustason, G., Pfetzing, D., & Zawoklow, E. Signing Exact English. Rossmoor, California:
Modern Signs Press, 1972.
Hart, B., & Rogers-Warren, A. A milieu approach to teaching language. In R. L. Schiefel-
busch (Ed.), Language intervention strategies. Baltimore: University Park Press, 1978.
Hauser, S. L., DeLong, G. R., & Rosman, N. P. Pneumographic findings in the infantile
autism syndrome. Brain, 1975, 98, 667-688.
Hollis, J. H., & Carrier, J. K. Intervention strategies for nonspeech children. In R. L.
Schiefelbusch (Ed.), Language intervention strategies. Baltimore: University Park
Press, 1978.
Konstantareas, M. M., Oxman, J., & Webster, C. D. Simultaneous communication with
autistic and other severely dysfunctional nonverbal children. Journal of Communica-
tions Disorders, 1977, 10, 267-282.
Konstantareas, M. M., Oxman, J., & Webster, C. D. Iconicity: Effects on the acquisition of
sign language by autistic and other severely dysfunctional children. In P. Siple (Ed.),
Understanding language through sign language research. New York: Academic Press,
1978.
Lovaas, O. I. The autistic child: Language development through behavior modification.
New York: Irvington, 1977.
Lovaas, O. I. Parents as therapists. In M. Rutter & E. Schopler (Eds.), Autism: A reappraisal
of concepts and treatment. New York: Plenum, 1978.
Lovaas, O. I., Koegel, R., Simmons, J. Q., & Long, J. S. Some generalizations and follow-up
measures on autistic children in behavior therapy. Journal of Applied Behavior Anal-
ysis, 1973, 6, 131-166.
Lovaas, O. I., & Schreibman, L. Stimulus overselectivity of autistic children in a two stimulus
situation. BehaviourResearch and Therapy, 1971, 9, 305-310.
Lovaas, O. I., Schreibman, L., Koegel, R., & Rehm, R. Selective responding by autistic
children to multiple sensory input. Journal of Abnormal Psychology, 1971, 77, 211-222.
McKeever, W. F., Hoemann, H. W., Florian, V. A., & VanDeventer, A. D. Evidence of
minimal cerebral asymmetries for the processing of English words and American Sign
Language stimuli in the congenitally deaf. Neuropsychologia, 1976, 14, 413-423.
Menyuk, P. Language: What's wrong and why. In M. Rutter & E. Schopler (Eds.), Autism:
A reappraisalof concepts and treatment. New York: Plenum, 1978.
Miller, A., & Miller, E. E. Cognitive-developmental training with elevated boards and sign
language. Journal of A utism and Childhood Schizophrenia, 1973, 3, 65-85.
O'Connor, N. Visual perception in autistic children. In M. Rutter (Ed.), Infantile autism:
Concepts, characteristicsand treatment. Edinburgh: Churchill, Livingstone, 1971.
Poizner, H., & Battison, R. Cerebral asymmetry for sign language: Clinical and experimental
evidence. Languages, in press.
Sign Language and Autism 137

Poizner, H., & Lane, H. Cerebral asymmetry in the perception of American Sign Language.
Brain and Language, 1979, 7, 210-226.
Rutter, M. Diagnosis and definition. In M. Rutter & E. Schopler (Eds.), Autism: A reap-
praisal of concepts and treatment. New York: Plenum, 1978.
Salvin, A., Routh, D. K., Foster, R. E., & Lovejoy, K. M. Acquisition of modified American
Sign Language by a mute autistic child. Journal of Autism and Childhood Schizo-
phrenia, 1977, 7, 359-371.
Schaeffer, B. Teaching spontaneous sign language to nonverbal children: Theory and method.
Sign Language Studies, 1978, 21, 317-352.
Schaeffer, B., Kollinzas, G., Musil, A., & McDowell, P. Spontaneous language for autistic
children through signed speech. Sign Language Studies, 1977, 17, 287-328.
Schiefelbusch, R. L. Introduction. In R. L. Scbiefelbusch (Ed.), Bases of language interven-
tion. Baltimore: University Park Press, 1978.
Schopler, E. Changing parental involvement in behavioral treatment. In M. Rutter & E.
Schopler (Eds.), Autism: A reappraisal of concepts and treatment. New York:
Plenum, 1978.
Schopler, E., & Reichler, R. J. Parents as cotherapists in the treatment of psychotic children.
Journal of Autism and Childhood Schizophrenia, 1971, 1, 87-102.
Schwam, E. Signs and strategies: The interactive l~rocesses of sign language learning. In
K. E. Nelson (Ed.), Children's language (Vol. 4). New York: Gardner Press, in press.
Webster, C. D., McPherson, H., Sloman, K., Evans, M. A., & Kuchar, E. Communicating
with an autistic boy by gestures. Journal of Autism and Childhood Schizophrenia,
1973, 3, 337-346.
Yamada, J., Kriendler, J., & Haimsohn, M. The use of simultaneous communication in a
language intervention program for an autistic child: A case study. UCLA Working
Papers in Cognitive Linguistics, 1979, 1, 63-92.