Article Main Points

A 10-year-old nonverbal Greek boy, C.Z., who had been diagnosed with both bilateral sensorineural profound hear- ing loss and autism, was taught to use the Picture Exchange Communication System (PECS), with some modifications and extensions, over a 4-month intensive intervention period. C.Z.’s original communication and behavioral status as well as the PECS application process are presented, along with the communicative, language, and psychosocial outcomes follow- ing the intervention program. Follow-up data were collected 6 months post.

Little research has focused on the coexistence of hearingloss and autism in children (Gillberg & Steffenburg,1993; Gordon, 1991; Jure, Rapin, & Tuchman, 1991). Epidemiological studies reveal higher comorbidity rates than would be expected in the general population (Gordon; Jure et al.). As Konstantareas and Homatidis (1987) reported, au- ditory peculiarities and abnormalities in children with autism have often been noted, yet no systematic and reliable data have been collected on the frequency of hearing loss or peripheral ear problems in this population. Generally, hearing problems are reported more frequently in people with neurological and developmental disorders than in those without such diagnoses (van Schrojenstein Lantman-de Valk, 1997). Autism has also been reported to occur more often in children with viral- related diseases known to affect hearing, such as congenital ru- bella (Chess, 1971) and congenital cytomegalovirus infection (Stubbs, Ash, & Williams,1984).

The prevalence of autism is 30–40 cases in 100,000 (i.e., 0.3%–0.4%; Rutter, 2005) while the prevalence of hearing loss in the general population under 45 years of age is 4% (Nadol, 1993) and 2% of newborn infants experience bilateral hearing loss with 0.7% in the moderate to profound range (Conn- Wesson et al., 2000). Based on these studies, it can be con- cluded that the incidence of hearing loss in children with autism, as well as the incidence of autism in children with hear- ing loss, is greater than the one found in typically developing populations. Jure et al. (1991) found that 61 out of 1,150 chil- dren with hearing impairments met the criteria for a diagnosis of autism, or 5.3% of the population of individuals with hear-

ing impairments. In another Swedish study, Rosenhall, Nor- din, Sandstrom, Ahlsen, and Gillberg (1999) reported on au- diological examinations of 199 children with autism; 7.9% of them were found to exhibit mild to moderate hearing loss, and 3.5% had severe or profound hearing losses.

Despite the high comorbidity of autism and hearing loss, few studies have used complete audiometrical battery tests (i.e., combined measures of pure tone audiometry and tym- panometry) to assess the hearing status of children with autism (Smith, Miller, Stewart, Walter, & McConnell, 1988). The small number of studies may be explained by the observation that hearing assessment is very difficult in cases of combined autism and hearing impairment, as it is in all children with mul- tiple disabilities (Jure et al., 1991). On one hand, no in-depth hearing examinations are carried out on individuals who pre- sent behaviors associated with severe autism (Hayes & Gor- don, 1977). On the other hand, in many combined cases the diagnosis of hearing loss comes much earlier than the diagno- sis of autism (Roper, Arnold, & Monteiro, 2003). Roper et al. concluded that the characteristics of autism in deaf children re- semble those observed for hearing children. Nevertheless, when both conditions occur, the communicative and language problems are even more severe than with either of the two dis- orders in isolation (Rosenhall et al., 1999).

Although many specialized intervention methods are be- ing used with varying degrees of effectiveness for individuals with autism or hearing loss, no research is available regarding therapy in combined cases. As Jure et al. (1991) state, there are no school programs specifically designed for children with autism and hearing impairment. They suggested that the ideal solution for these children would be placement in a special class of a school for the hard-of-hearing, where a behavioral intervention program would be combined with sign language training.

Alternatively, a promising avenue for communication train- ing in the case of children with autism and hearing impairment may be combining a nonverbal, alternative communication ap- proach with the principles of behavioral analysis. Alternative augmentative communication systems have commonly been used to promote the communication skills of nonverbal chil-

FOCUS ON AUTISM AND OTHER DEVELOPMENTAL DISABILITIES VOLUME 22, NUMBER 1, SPRING 2007

PAGES 23–32

The Application of PECS in a Deaf Child With Autism: A Case Study Georgia A. Malandraki and Areti Okalidou

FOCUS ON AUTISM AND OTHER DEVELOPMENTAL DISABILITIES

24

dren with autism. The American Speech–Language–Hearing Association (1989) defines alternative augmentative commu- nication as an area of clinical practice that attempts to com- pensate (temporarily or permanently) for the disability of people with severe expressive communication disorders. The communication challenge is to assist verbal and nonverbal chil- dren with autism in developing functional communication and spontaneous (oral or sign) language use (Carr, Binkoff, Kolo- ginsky & Eddy, 1978; Charlop, Shreibman & Thibodeau, 1985; Lovaas, 1977) Most importantly, the acquisition of functional communication is considered a primary prognostic factor for long-term communication outcomes (Bondy & Frost, 1994).

A method that targets functional communication via non- verbal means is the Picture Exchange Communication System (PECS) developed by Frost and Bondy (1994) in the Dela- ware Autistic Program. PECS applies principles of applied be- havior analysis to a system of alternative and augmentative communication that focuses on self-initiated communicative behavior (Bondy & Frost, 1998). Although initially designed for young children with autism, PECS has been applied with success to people with other developmental disorders (Liddle, 2001) and across the age range, up to age 26 (Chambers & Rehfeldt, 2003). The basic idea of PECS is to provide indi- viduals who have limited communication abilities with a sys- tem via which they can interact with others, making their needs and desires known (Frost & Bondy, 1994, 1998). PECS users learn to approach a communicative partner and exchange, with him or her, the picture of a desired item for the item itself. By doing this, they learn to initiate spontaneous communication within a social context and to participate in the cause–effect act of communication (Frost & Bondy, 1994, 1998), an act that is fundamental to the whole communication process.

Although there are limited reports on the efficacy of PECS, its few documented implementations in individuals with autism spectrum disorders and other developmental dis- abilities have resulted in very encouraging communicative and social outcomes (Ganz & Simpson, 2004; Kravits, Kamps, Kemmerer, & Potucek, 2002; Liddle, 2001). Bondy and Frost (1994) indicated that 85 children with autism acquired func- tional communication via PECS. Bondy and Peterson (1990) reported that 59% of children with autism who used PECS for more than 1 year managed to acquire oral language as their sole means of communication. Finnegan and colleagues (1987) indicated that the use of PECS resulted in decreases in problem behaviors and increases in appropriate social behaviors. Furthermore, Charlop-Christy, Carpenter, Le, LeBlanc, and Kellet (2002) provided empirical evidence for the positive out- comes obtained by PECS in the acquisition of speech and com- municative-social behaviors and in the reduction of problem behaviors.

The clinical case reported in the present study concerns a child with multiple disabilities diagnosed with both autism and profound hearing loss. Because the child had never worn hear- ing aids and was nonverbal, PECS was chosen as the alterna-

tive augmentative communication system for this child’s com- munication training. The present study constitutes the first documented trial of the application of PECS in a combined case of autism and profound hearing loss, and also one of the few documented incidences of PECS application in Greece. It occurred within the framework of the child’s communication intervention program at school. The goal of the intervention was to establish functional communication for the child, who could not communicate his needs or feelings with verbal or nonverbal language. The goal of the study was to assess the ef- ficacy of PECS as a communication method for this school- age, nonverbal child with autism and hearing impairment.

Method

Participant

The study was approved by the Research Committee of the Technological and Educational Institute of Patras and by the National Foundation of Deaf and Hard of Hearing Individu- als of Southwestern Greece. Written consent was given by the participant’s parents. The participant, C.Z., was a 10-year-old Greek boy who had been diagnosed with both congenital pro- found hearing loss and autism.

A complete battery of audiological assessments (i.e., pure- tone audiometry, tympanometry, and auditory brainstem re- sponse [ABR]) was performed at the Department of Audi- ology and Neuro-Otology at a children’s hospital in Athens. Audiological results established the diagnosis of profound bi- lateral sensorineural hearing loss at the age of 2 and 6 years. At age 10, the diagnosis of autism was made by a psychologist at C.Z.’s school, based on the criteria in the Diagnostic and Statistical Manual of Mental Disorders (4th ed.; American Psy- chiatric Association, 1994). C.Z. had been enrolled for 3 years as a boarding student at a national school for children who are deaf and hard of hearing. He attended school regularly. The school followed the total communication method for deaf education, and C.Z. was instructed via Greek Sign Language, finger-spelling, and written language. Throughout school hours he had been receiving in-class instruction support by a paraprofessional who served as his tutor. C.Z.’s hearing status was monitored annually by the audiological center affiliated with the school. During his third year in school he was evalu- ated by the school’s psychologist, who conducted a couple of follow-up visits to monitor C.Z.’s behavioral status. An IQ as- sessment was not administered due to the lack of nonverbal standardized tests in this area in Greece. For three semesters, C.Z. had received speech and language therapy from three stu- dent clinicians completing supervised practica. At that time his speech and language therapy were focused on increasing lip- reading skills, without any signs of progress. His family history revealed that C.Z.’s parents were both deaf and communicated with each other through the Greek Sign Language. However, his 19-year-old sister was hearing and typically functioning.

VOLUME 22, NUMBER 1, SPRING 2007

25

Study Design

The study was composed of four phases:

1. an informal assessment; 2. the main intervention program (4 months), which was

highly intensive and allowed acquisition of PECS up to 6th phase;

3. the maintenance intervention program (4 months), which was less intensive, conducted by a single new trainer, and consisted of two sessions per week of 30 to 45 minutes each; and

4. the follow-up, which was a 2-hour informal observation by the main trainer that took place 6 months following cessation of the main intervention program.

Data were collected only for the main intervention program and are reported in this article.

Prior to the onset of training, communication skills were informally assessed via teacher and caregiver interviews and records review, in addition to direct observation and elicited responses. Data pertaining to the child’s communication pro- file before and after the intervention program were recorded. Also, gross observations on the behavioral and social patterns before and after the intervention were obtained. Following the establishment of baselines regarding the child’s communica- tive and psychosocial behaviors during the 1-week observation and the 3-day assessment period, PECS training began. Dur- ing Phases 1, 2, and 3, PECS training was highly structured, consisting of 3 to 5 sessions weekly, each of 30 to 45 minutes duration. Subsequently, during Phases 4 and 5, PECS training became more intensive, but sessions were not structured. For 1.5 months the trainers spent 8 to 10 hours daily in C.Z.’s en- vironment in order for him to apply the communication sys- tem in his daily living. Thus, the training staff participated in C.Z.’s daily activities, identified teachable moments, and used incidental teaching to stimulate communicative interactions. The following steps were taken to foster generalization: (a) change of trainers, (b) change of environment (therapy rooms), (c) remote placement of the communication book, and (d) expansion of therapy time to consecutive whole-day sessions in C.Z.’s living environment.

Settings

The settings in which the different phases of PECS were im- plemented included almost all of the common places of the boarding house, a speech therapy room, and C.Z.’s classroom at school. In the first phases of PECS, the therapy room used was a 10-foot by 16-foot room with one table and three chairs. Two of the chairs (for the child and the second trainer) were placed next to each other, and the third chair (for the main trainer) faced them from across the table. In subsequent phases of the program, sessions were held in a variety of places rep- resenting the actual living environments of the child, namely,

the common room, the playroom, the dining room, the bed- room, and the study room of the boarding house, as well as a few other bigger classrooms at the school. The bigger class- rooms enabled an increase in distance between the child, the trainers, and PECS materials, which was a program require- ment in advanced phases of PECS.

Trainers

None of the trainers had previous experience with PECS. The main trainer was a senior undergraduate student in the De- partment of Speech and Language Therapy of a local univer- sity (and one of the authors of the study), and the second trainer was a kindergarten teacher. The main trainer studied the PECS Training Manual (Frost & Bondy, 1994) and was supervised by a certified speech–language pathologist familiar with its implementation. The main trainer trained the second trainer via three 30-minute sessions of direct instruction and additional on-site observation. Subsequently, six additional trainers (five of C.Z.’s caregivers at the boarding house, plus his school teacher) were individually trained by the main trainer before participating in the advanced phases of the pro- gram. Each one had a 30-minute session of direct instruction with the trainer, and the entire procedure of each phase was explained during some demonstrative training sessions. Dur- ing each phase the additional trainers were also asked to come into the therapy room to observe the main trainer (on-site observation). Finally, after the acquisition of PECS (main in- tervention program), an additional trainer carried on a main- tenance program and received in-service training for 2 weeks, which included (a) three 30-minute sessions of direct instruc- tion and inspection of videotaped sessions, (b) on-site obser- vation, and (c) hands-on supervised training.

Materials

Before the intervention, trainers devoted a few days preparing and constructing the necessary materials (i.e., the pictures and the cards). These were constructed according to the require- ments of PECS (Frost & Bondy, 1994). The initial cards used were black-and-white pictures with drawings composed of thick black lines. Pictures were designed by hand by a semi- professional designer, then scanned and processed by the re- searchers with the software program Graphic Converter. All pictures were the same size (10 centimeters by 5 centimeters). C.Z. was familiar with almost all of the pictures chosen. Fig- ure 1 shows an example of the PECS pictures used.

Starting with the second phase, a communication book was introduced. This was a three-ring binder where all pic- tures, and subsequently word cards, were attached. Velcro was used to attach the pictures, the word-cards, a sentence strip, and cards with the phrases “I want” and “This is” in Greek. The picture cards were replaced by word cards of the same size, written in lower case letters. In subsequent phases these word cards were reduced to approximately half the original size.

FOCUS ON AUTISM AND OTHER DEVELOPMENTAL DISABILITIES

26

During the fourth phase, 66 small cards (2.5 centimeters square) were added to C.Z.’s communication book and orga- nized into six categories. Table 1 presents the items included in each category during this phase.

Informal Assessment

Descriptions of Behavior. Information on C.Z.’s be- havior patterns and psychosocial condition was gathered via teacher reports and direct observation for 7 consecutive days, 6 hours per day. During direct observation, all of C.Z.’s dis- crete behaviors were recorded in a notebook. The following behaviors were noted:

• Establishing fleeting eye contact when addressed by his teachers or caregivers for less than 5 seconds per occurrence.

• Stereotypical movements of clapping and hand-flapping conducted without a specific aim while standing, walking, or sitting on his favorite chair. The frequency of these movements was 30 per hour.

• Stereotypically walking back and forth the width of the main corridor to the kitchen without a specific aim, three times or more per half hour for 3.5 hours per day.

• Insistence on sitting specifically and exclusively on a particular chair, located in a secluded area of the children’s playroom, where he would not allow anyone else to sit.

• Not responding to requests by caregivers to enter and play in the children’s playroom.

Communication Skills. A questionnaire administered to teachers and caregivers indicated that C.Z. rarely communi- cated interactively. When he wanted something edible he com- municated by

• going toward the thing he wanted (the most frequent be- havior pattern);

• leading (taking an adult’s hand and guiding him or her to the desired item);

• using the sign for “come,” when he invited someone to do something for him or to give him something, without indicating anything else and without making eye contact; or

• screaming or moving his head (rare).

With respect to language comprehension, C.Z. was able to follow two one-step commands (“sit down” and “stand up”), which were given by the trainer using a hand gesture. He did not respond to his name auditorily, in writing, visuo- auditorily, or by finger-spelling. With respect to written lan- guage skills, C.Z. had a relatively sizable rote vocabulary, as he received formal instruction via a tutor-aided literacy program. When cued by modeling the type of task, he was able to read, copy, and write almost 150 words (mostly nouns), indepen- dently (30+) or with prompting (i.e., after the trainer had writ- ten the first letter or two). His reading ability was assessed by prompting him to match written words to pictures or actual objects. Upon picture presentation he could finger-spell 31 words without prompting and nearly 150 words with minimal prompting (i.e., after the trainer had finger-spelled the first one or two letters). The 31 words were: (a) eye, hair, mouth, nose, ears, leg, hands, and neck (body parts); (b) red, black, blue, white, green, purple, brown, light blue, yellow, and orange (col- ors); (c) cat, horse, goat, chicken, bird, and dog (animals); and (d) orange, apple, lemon, strawberry, banana, and pear (fruit). Reportedly, he had never used these written words for com- munication and was observed only once to exhibit sponta- neous writing: When 8 years old, he wrote the correct date of that particular day on the blackboard in his classroom. In sum, C.Z. did not communicate functionally with other people, hearing or deaf, by using oral or sign language, even though

FIGURE 1. Progression of card presentation during PECS training. The top figure indicates the initial type of card, the middle figure the type of card including the picture and the written word, and the bottom figure the card with just the written word on it.

VOLUME 22, NUMBER 1, SPRING 2007

27

he had been exposed to signing both at school and at the boarding house for 3 years and at home since birth. Further- more, his writing and finger-spelling skills remained unex- plored with respect to functional communication as he only used them upon request or during formal learning tasks dic- tated in class by his tutor. Evidently, C.Z. was not motivated to communicate with others, nor was he directly taught how to communicate. Therefore, the goal of the intervention was to establish functional communication. For that purpose, the researchers chose PECS, with a few adaptations made because C.Z. had a profound hearing loss.

Intervention Program

Assessment of Reinforcers. Prior to the actual inter- vention, an assessment of potential reinforcers was performed to identify the items C.Z. found the most desirable. Three ses- sions were devoted to the task of identifying toys and edibles. The procedure strictly followed the principles and guidelines of the PECS Training Manual (Frost & Bondy, 1994) by in- terviewing others who knew the child, observing the child in the natural environment, and conducting a formal reinforce- ment assessment using a multiple-without-replacement assess- ment procedure (DeLeon, Iwata, & Roscoe, 1997). For the category of toys, a total of eight items were chosen, in order of C.Z.’s preference: car, Playmobil©, truck, Play-Doh©, col- ored pencils, magic markers, teddy bear, and ball. C.Z. was hesitant to reach for edibles. He did not choose anything dur- ing two sessions, most likely because he had been chastised by his caregivers in the past for grabbing food. In view of this ap- parent difficulty, C.Z.’s caregivers were asked to provide their opinions regarding his favorite edibles. A total of nine edibles were chosen: banana, cookie, chocolate, potato chips, apple, pear, Coca-Cola©, orange juice, and water.

PECS Modifications

Training followed the guidelines described in the treatment protocol of the PECS Training Manual (Frost & Bondy, 1994). Except for the seventh (additional) phase, which was not targeted due to time limitations, PECS phases were ap- plied with some extensions and modifications (reported in Table 2) tailored to C.Z.’s hearing loss and writing ability.

Data Collection and Reliability

During the first three stages of PECS, data were recorded by two trainers using assessment forms adapted from the PECS Training Manual (Frost & Bondy, 1994). Initially, the main trainer independently collected data while the second trainer served as an observer. Then, at regular intervals (i.e., every 10 trials), the second trainer was asked to score over the relevant part at the assessment forms. Only responses receiving verifi- cation by the second trainer were considered valid. Fewer than

five disagreements occurred in this decision process, and in- terobserver agreement was 96.5%. In subsequent phases, the main trainer continued to be responsible for collecting data, and the second-trainer scoring procedure was followed by each of C.Z.’s caregivers (whoever was on duty at the time). Train- ing of caregivers took place at the third stage of the fourth phase.

The criterion for the trainers to move to the next PECS step was that C.Z. exhibited at least 80% success in perform- ing each session’s trials independently. The number of trials per scheduled session varied from 17 to 45, depending on the tasks used at each stage.

Results

Before the application of PECS, C.Z. had a limited, nonfunc- tional communication profile. After the intervention he achieved a much higher communication level, which was completely functional in social contexts. Table 3 provides the number of sessions required for mastery at each of the stages.

As shown in Table 3, the first phase (teaching the physi- cally assisted exchange) with its extension (written cards in- stead of pictures) for both categories lasted 9 sessions, the second phase (expanding spontaneity) lasted 11 sessions, the third phase (simultaneous discrimination of pictures) lasted 3 sessions, the fourth phase (building sentence structure) lasted 20 sessions, the fifth phase (responding to “What do you want?”) lasted 16 sessions, and the sixth phase (commenting in response to a question) lasted 4 sessions.

TABLE 1 Cards Created for Phase 4

Category Written words: Items

Edibles Chips, crescent rolls, chocolate, cookie, orange juice, Coca-Cola, water, apple, banana, pear, bread, orange, milk, yogurt

Toys Ball, car, truck, Playmobil, puzzle, teddy bear

Basic needs Sleep, food, toilet

School things Pencil, eraser, pencil sharpener, notebook, book, bag

Things for arts and crafts Colored pencils (10 words, one for each color, e.g., green pencil), magic markers (10 words, one for each color, e.g., yellow marker), scissors, glue, Play-Doh, yarn of all colors

Useful items Glass, plate, spoon, fork, straw

FOCUS ON AUTISM AND OTHER DEVELOPMENTAL DISABILITIES

28

In Phase 4 (sentence structure), the first stage (placing a selected card next to the “I want” card) and second stage (moving both the “I want” card and the target card, and hand- ing the strip over) were acquired in 3 sessions, but the third stage (use of a sentence strip to request far-remote items), which in essence constitutes the spontaneous use of the PECS program, was acquired in 20 days. The spontaneous use of the communication book was a very difficult task that needed sev- eral hours of training each day. Phase 5 began before Phase 4 was completely finished. Phase 5 was acquired 1 day before

Phase 4. Table 4 presents descriptions of communicative be- haviors before and after intervention.

During generalization of Phases 4 and 5, C.Z. started to use some of the signs the trainers had shown him with the cards. In particular, the first signs he made were for water, toi- let, food, and pencil, which he gestured during the time he was matching the cards on the sentence strip prior to handing them to the trainers. An unexpected outcome was the onset of vocalizations. C.Z. vocalized during the generalization of Phases 4 and 5, at exactly the moment he was choosing the

TABLE 2 PECS Modifications and Additions for Training a Child With Autism and Deafness

Intervention goals Changes of PECS training Analytic description of PECS changes

To facilitate functional use of writing

To facilitate focus on alternative sensory channels (vision and touch)

To effect multimodal communication mode

To effect spontaneous use of PECS

To accelerate acquisition of spontaneous use of PECS

Note. PECS = Picture Exchange Communication System (Frost & Bondy, 1994).

Modification: Gradual replacement of pictures by written words (Phase 1)

Modification: Replacement of all verbal praises, reassurances, and prompts by a multi-nonverbal communication mode

Addition: Multimodal communica- tion responses by the trainers (onset at Phase 4)

Addition: Multimodal communica- tion initiations by the trainers (Phases 5 and 6)

Addition: Dramatic increase of in- tervention time (onset at the third stage of Phase 4)

Modification: Preceding the third stage of Phase 4 (spontaneous requesting via PECS) by Phase 5 (responding to the question “What do you want?”)

At first, pictures were presented alone; then they were presented with written words underneath them; finally, they were com- pletely replaced by written words. The implementation of this extension occurred early on because during his evaluation C.Z. had shown increased facility with the recognition of many writ- ten words.

Throughout PECS training, all verbal praises, reassurances, and prompts were given via nonverbal means, namely, by simulta- neous usage of Greek Sign Language, intense facial expres- sions, and intense clapping, as well as praise touch (such as shoulder patting).

During and after Phase 4, the trainers’ responses following C.Z.’s returning the sentence strip to them involved more than verbal- ization of the sentence. They simultaneously expressed the sentence via Sign Language and with concurrent emphatic ar- ticulation while holding the strip very close to their mouths to facilitate lip-reading.

In Phases 5 and 6, written question forms replaced the verbal questions, “What do you want?” and “What is this?” The train- ers also expressed these questions in Sign Language and with emphatic articulation while holding the strip very close to their mouths to facilitate lip-reading.

The third stage of Phase 4 is a crucial stage because it targets spontaneous use of the communication book in everyday life. For the acquisition of this stage, the trainers decided to spend approximately 8 hrs per day at the boarding house. The spe- cific goals were to introduce the whole program in C.Z.’s every- day life and to instruct his caregivers in the use and functionality of the PECS program.

Due to time restrictions related to the contracting institution providing the speech therapy program, the order of Phase 4 (Stage 3) and Phase 5 was reversed in hope of enhancing com- pletion of the PECS program within the assigned time frame. It was hypothesized that an earlier introduction of Phase 5 would facilitate the acquisition of Stage 3 of Phase 4. Hence, the trainers decided to begin the application of Phase 5 after Stage 3 of Phase 4 was introduced but before it was acquired, that is, before C.Z. learned to express his needs spontaneously via the sentence strip “I want ___ .”

VOLUME 22, NUMBER 1, SPRING 2007

29

cards to put them on the sentence strip. These vocalizations seemed like an effort to read each word aloud. Until then no similar vocalizations had been observed, although he had been able to recognize the words for 2 years prior to the study.

The acquisition of a communication system was found to have a positive impact on C.Z.’s psychosocial functioning. Table 5 presents descriptions of psychosocial behaviors before and after intervention.

Another observed outcome was the change of other chil- dren’s behavior toward C.Z. According to his caregivers, the other children had always treated C.Z. very positively (even protectively), but after the application of PECS their behavior became even more positive. In particular, they began to play with him more and often hugged and kissed him. Many times they tried to be his trainers, with great success, by taking the card with the written question “What do you want?” and ask- ing him in the same way his trainers did.

Moreover, a positive psychosocial outcome of the PECS intervention was the observed change toward his mother. Prior to PECS, C.Z. had never expressed any anticipation of her arrival to take him home. Even when he did not see her for a long period of time, he reacted almost indifferently. For the first time, a few weeks before the intervention was com- pleted, he exhibited new behaviors toward his mother when

she came to take him home. At first he eagerly anticipated her arrival; then, when his mother came, he quickly grabbed his bag and ran toward her. He was smiling, and his mother re- ported that was the first time she had seen him so happy. She assured the trainers that his general behavior at home had im- proved as well.

Maintenance Program and Follow-Up Observation

During the 4-month maintenance program, C.Z.’s communi- cation book was available at all times and in all settings. Ac- cording to the additional trainers’ reports, during this period C.Z. continued to exhibit spontaneous usage of PECS, both to request and to comment.

Six months after the main intervention period, a 2-hour observation was arranged at the boarding house. The recorded behaviors suggested that the intervention continued to be ef- fective. Specifically, the trainers recorded the following: (a) C.Z. spontaneously requested five items by using his communica- tion book; (b) he spontaneously requested toilet and water through Greek Sign Language; (c) he understood two orders that were given to him through Greek Sign Language (“wash your hands” and “bring your bag”); and (d) he responded cor-

TABLE 3 Stages and Number of Sessions for PECS Acquisition

PECS phases Stage Sessions

Phase 1: Teaching the physically assisted exchange Exchange pictures for toys 4 Exchange written words for toys 2 Exchange pictures for edibles 1 Exchange written words for edibles 2

Phase 2: Expanding spontaneity Remove card from communication board 2 Increase distance between book and trainer 2 Increase distance between child and book 4 New communication partners 3

Phase 3: Simultaneous discrimination of pictures Exchange via picture discrimination 1 Correspondence checks 1 Use of small cards 1

Phase 4: Building sentence structure Placing card in the “I want ___ ” sentence frame 1 Constructing and delivering the sentence “I want + card” to trainer 2 Requesting remote, nonvisible objects 17a

Phase 5: Responding to “What do you want?”b Asking the child with zero-delayed pointing prompt 0.5 Asking the child with delayed pointing prompt 0.5 Asking without pointing prompt 15a

Phase 6: Commenting in response to a question Placing card in the “This is ___ ” sentence framec 3 Two-question discrimination 1

Note. PECS = Picture Exchange Communication System (Frost & Bondy, 1994). aAll-day sessions. bPhase 5 was introduced prior to Stage 3 of Phase 4. Following onset of the Stage 3 of Phase 4, C.Z. requested 15 items either as a response to the question or spontaneously. c The question “What do you see?” and the three-question discrimination stage were not used due to time limitations of the intervention program at the school.

FOCUS ON AUTISM AND OTHER DEVELOPMENTAL DISABILITIES

30

rectly to alternating written questions, “What do you want?” and “What is this?”

Furthermore, his caregivers commented that he used many more Greek Sign Language signs to communicate. He often smiled and was very affectionate with his trainers. He exhib- ited fewer stereotypical movements with hands. Most of the time he sat on the couches with other children; only once did he go toward his favorite chair, but did not sit on it.

Discussion

The primary goal of this study was to establish a functional communication system for a 10-year-old boy with profound hearing loss and autism, who was unable to communicate functionally using verbal or nonverbal language. The commu- nicative results described above show that this goal was met with adequate success. With 4 months of intensive interven- tion, C.Z. learned to use a functional communication tool (the PECS communication book) spontaneously.

Interestingly, the training period for PECS acquisition in this case was longer than for 3 hearing children with autism who were 3–8, 5–9, and 12 years old, all of whom acquired the six phases in 11 to 12 sessions (Charlop-Christy et al., 2002). This difference might be attributed to C.Z.’s multiple disabilities. However, C.Z.’s training period is similar to that reported for a 3-year-old boy with autism (Bondy & Frost,

1994). Hence, it seems that the pace for PECS acquisition de- pends on factors other than age or the presence of multiple disabilities. Because longitudinal data on C.Z.’s progress are lacking, one can only speculate that younger children would reach higher levels of language than older children. Nonethe- less the remarkable finding is that, overall, within one-third of a year, C.Z. achieved satisfactory command of PECS for com- munication purposes.

Additionally, the entire intervention led to very positive changes in C.Z.’s psychosocial development, elements which were not targeted by the intervention program. During the final phases of PECS, he started to socialize (i.e., play and in- teract with his peers) and ceased having an obsession for sit- ting on the isolated chair. This significant change may be attributed to increased communication skills, which play a crit- ical role in the development of socialization skills. Another outcome was the reduction of behaviors characteristic of autism, as C.Z. reduced the frequency of stereotypical move-

TABLE 4 Communication Behaviors Before and

After PECS Intervention

Before After

Going toward the things he wanted

Guiding someone to the things he wanted (lead- ing) without looking at the communication partner

Making a come sign with his hand without look- ing at the communica- tion partner

Rarely screaming or mov- ing his head

No responding to questions

Note. PECS = Picture Exchange Communication System (Frost & Bondy, 1994).

Spontaneously exchanging the sen- tence strip (with a complete three-word written sentence) for a particular item or skill (i.e., sponta- neously initiating communication)

Interacting with a communication partner in order to ask for some- thing and usually establishing eye contact before going to his com- munication book

Simultaneously beginning to use signs from Greek Sign Language and almost always looking at his communication partners

No screaming, no head moves

Responding to questions “What do you want?” and “What is this?”

TABLE 5 Psychosocial Profile of C.Z. Before and After

PECS Intervention

Before After

Rare eye contact lasting fewer than 5 seconds with train- ers and other communica- tion partners

Many stereotypical move- ments, especially clapping and flapping hands

A stereotypical walk from the width of the main corridor to the kitchen, approxi- mately six times per hour

A severe obsession with sit- ting on a particular isolated chair, on which he did not let anyone else sit

Almost complete indifference in the presence of other people (except for rare cases); flat affect

Refusal to enter and play in the children’s playroom

Note. PECS = Picture Exchange Communication System (Frost & Bondy, 1994).

Established eye contact for more than 15 seconds with trainers and other communicative part- ners during interactions

Reduction of all stereotypical movements

Stereotypical walk occurred only three or four times per day at most

Rarely sat on “his” chair any more; instead, preferred to sit on the common-room couches with the other children, and surprisingly, showed indiffer- ence when someone else sat on his chair

Became expressive with others and had body contact with them much more often; devel- oped a positive affect, having a smiling face and many times hugging others spontaneously

Entered playroom daily, thus be- ginning to interact with other children during play

VOLUME 22, NUMBER 1, SPRING 2007

31

ments. Bondy and Frost (1998), Finnegan et al. (1987), and Charlop-Christy et al. (2002) have documented the positive contribution of PECS to the alleviation of behavior manage- ment problems in individuals with autism. Increases in social behaviors and decreases in aberrant and stereotypical behav- iors mirror the findings of a Natural Aided Language inter- vention case study of an adolescent with autism (Cafiero, 2001).

Several complementary explanations can be posited for the behavioral and social outcomes of PECS as used with C.Z. First, after the intervention C.Z. could make his desires and needs known; this newly acquired communication ability may have reduced frustration and concomitant aberrant behaviors. Alternatively, C.Z.’s communication competency may have brought about a global behavioral change in his living and learning environment, which in turn may have had a positive impact on his behavior. Both adults (teachers, tutors, care- givers) and other children were able to interact with C.Z. via PECS. Hence, “his change” may have brought about “their change,” facilitating more change on C.Z.’s part, and so on. Such chaining of positive changes may have been conducive to interactions triggered by C.Z.’s newly acquired communica- tion ability. Pertaining to language skills, he started acquiring some Greek Sign Language concepts for the first time in his life, and 6 months later this ability had expanded. This acqui- sition can be compared to the acquisition of verbal language concepts by hearing children with autism after the application of PECS (Bondy & Frost, 2001). Like many of those children, who begin to speak and express themselves in their language by connecting the verbal word with the picture, C.Z. began to “speak” in his language, Greek Sign Language, by con- necting each sign with the written word. Thus in the case of C.Z., PECS may have functioned as the means to acquire sign language.

Moreover, the fact that C.Z. began to make vocalizations that he had never made before leads to the conclusion that the particular intervention program, although it does not directly target verbal language, encourages its acquisition. It should be noted that onset of vocalizations is rare in the case of a 10- year-old deaf child, especially for those with multiple disabili- ties. Nevertheless, vocalization is an expected outcome when PECS is used (Bondy & Frost, 1994; Bondy & Frost, 1998; Charlop-Christy et al., 2002; Ganz & Simpson, 2004). This newly acquired behavior was concomitant with PECS usage. Interestingly, C.Z.’s communication development encom- passed both language and speech areas in a rudimentary way, despite his nonverbal orientation.

The anecdotally reported array of unexpected positive changes—increases in social interactions, increases in vocali- zations, decreases in aberrant behaviors, emergence of sign language, and changes in peer responses and positive affect— resembles the “Pygmalion effect” (Rosenthal & Jacobson, 1992), according to which students perform better simply be- cause they are expected to do so. C.Z.’s communication change may have reinforced the staff ’s attitudes and expecta- tions toward C.Z., thereby increasing their interactions; the

latter change may then have been incorporated into peer atti- tudes toward C.Z. These changes deserve systematic future study to document PECS outcomes. An alternative interpre- tation is that C.Z.’s progress may not be attributed to PECS per se but to the intensiveness of the intervention, since the particular intervention that C.Z. received was more intensive than any of the other previous services.

One limitation of this study includes the scoring of C.Z.’s attempts in each session, which was conducted by both train- ers in unison and not independently. Anecdotally, all staff members of his boarding school who had known him for 3 years commented on his skill acquisition and verified that it was a dramatic change. Another limitation was that the data on his behavioral and social patterns collected at the beginning and end of the intervention were merely descriptive. This pit- fall occurred because the original focus of this study was the change of C.Z.’s communication profile. More quantifiable and continuous data on behavioral and social patterns within the intervention period might have given insight on how and when his behavior began changing. Finally, data on fidelity of the PECS implementation were not collected across the vari- ous trainers. It is possible that variations of implementation may have enhanced or inhibited C.Z.’s acquisition of func- tional spontaneous communication.

This study confirms the effectiveness of the Picture Ex- change Communication System, with a few adaptations, in the case of combined autism and profound hearing loss. It also re- ports the impact of communication on C.Z.’s social life and confirms the PECS creators’ findings that their system encour- ages the emergence of both speech and language production. Because this study reports only a single case, it is imperative that more research be conducted with individuals with autism and hearing impairment to validate the finding that PECS can facilitate the development of communication in children with multiple disabilities, even at older ages.

ABOUT THE AUTHORS

Georgia A. Malandraki, MA, is a PhD candidate in speech–language pathology at the University of Illinois at Urbana-Champaign. Her cur- rent interests include developing diagnostic and therapeutic procedures for individuals with swallowing and voice disorders, as well as studying the neural correlates of swallowing and voice in normal and disordered populations. Areti Okalidou, PhD, is an assistant professor of speech– language pathology at Macedonia University. Her current interests in- clude speech and communication development of children with hearing loss and other developmental disabilities, speech and communication following cochlear implantation, detection of children with commu- nicative impairments and intervention. Address: Areti Okalidou, De- partment of Educational and Social Policy, University of Macedonia, 156 Egnatia Str., PO Box 1591, Salonika 540 06, Greece; e-mail: [email protected]

AUTHORS’ NOTES

1. We would like to thank Helen Drosinou, Louisa Voniati, Anna Ser- petsidaki, the Dean, the teachers, and all the staff members of the

FOCUS ON AUTISM AND OTHER DEVELOPMENTAL DISABILITIES

32

boarding school for their kind assistance and dedication that they ex- hibited in C.Z.’s intervention program.

2. Authors are listed alphabetically.

REFERENCES

American Psychiatric Association. (1994). Diagnostic and statistical manual of mental disorders (4th ed.). Washington, DC: Author.

American Speech–Language–Hearing Association. (1989). Compe- tencies for speech–language pathologists providing services in aug- mentative communication. ASHA, 31, 107–110.

Bondy, A. S., & Frost, L. A. (1994). The picture exchange commu- nication system. Focus on Autistic Behavior, 9, 1–19.

Bondy, A. S., & Frost, L. A. (1998). The picture exchange commu- nication system. Seminars in Speech and Language, 19, 373–388.

Bondy, A. S., & Frost, L. A. (2001). The picture exchange commu- nication system. Behavior Modification, 25, 725–744.

Bondy, A., & Peterson, S. (1990). The point is not the point: Picture- exchange communication system with young students with autism. Paper presented at the annual meeting of the Association for Be- havioral Analysis, Nashville, TN.

Cafiero, J. M. (2001). The effect of an augmentative communication intervention on the communication, behavior and academic pro- gram of an adolescent with autism. Focus on Autism and Other De- velopmental Disorders, 16, 179–189

Carr, E. G, Binkoff J. A, Kologinsky, E., & Eddy, M. (1978). Acqui- sition of sign language by autistic children. I: Expressive labeling. Journal of Applied Behavioral Analysis, 11, 489–501.

Chambers, M., & Rehfeldt, R. (2003). Assessing the acquisition and generalization of two mand forms with adults with severe devel- opmental disabilities. Research in Developmental Disabilities, 24, 265–280.

Charlop-Christy, M. H., Carpenter, M., Le, L., LeBlanc, L. A., & Kellet, K. (2002). Using the picture exchange communication sys- tem (PECS) with children with autism: Assessment of PECS ac- quisition, speech, social-communicative behavior, and problem behavior. Journal of Applied Behavior Analysis, 35, 213–231.

Charlop, M. H., Shreibman, L., & Thibodeau, M. G. (1985). In- creasing spontaneous verbal responding in autistic children using a time delay procedure. Journal of Applied Behavior Analysis, 18, 155–166.

Chess, S. (1971). Autism in children with congenital rubella. Jour- nal of Autism and Childhood Schizophrenia, 1, 33–47.

Conn-Wesson, B., Vohr, B. R., Sininger, Y. S., Widen, J. E., Folsom, R. C., Gorga, M. P., & Norton, S. J. (2000). Identification of neonatal hearing impairment: infants with hearing loss. Ear and Hearing, 21, 488–507.

DeLeon, I. G., Iwata, B. A., & Roscoe, E. M. (1997). Displacement of leisure reinforcers by food during preference assessments. Jour- nal of Applied Behavior Analysis, 30, 475–484.

Finnegan, G., Mclaughlin, D., Ryan, L., Wachowiak, J., Swanson, C., & Bondy, A. (1987, May). The impact of alternative/augmentative communication systems on inappropriate behaviors. Paper presented at the convention of the Association for Behavior Analysis, Nash- ville, TN.

Frost, L. A., & Bondy, A. S. (1994). The picture exchange communi- cation system: Training manual. Newark, DE: Pyramid Educational Consultants.

Ganz, J., & Simpson, R. (2004). Effects on communicative request- ing and speech development of the picture exchange communica- tion system in children with characteristics of autism. Journal of Autism and Developmental Disorders, 34, 395–409.

Gillberg, C., & Steffenburg, S. (1993). Debate and argument: Reply to A. G. Gordon: The evidence for pathogenetic role of peripheral hearing deficits in autism is extremely limited. Journal of Child Psy- chology and Psychiatry, and Allied Disciplines, 34, 593–596.

Gordon A. G. (1991). Co-occurrence of deafness and infantile autism. American Journal of Psychiatry, 148(11), 1615.

Graphic Converter v.3.7.2, Lemke Software, Peine, Germany 1992- 1999, www.lemkesoft.com

Hayes, R. W., & Gordon, A. G. (1977). Auditory abnormalities in autistic children. Lancet, 2, 767.

Jure, R., Rapin, I., & Tuchman, R. F. (1991). Hearing-impaired autistic children. Developmental Medicine and Child Neurology, 33, 1062–1072.

Konstantareas, M. M., & Homatidis, S. (1987). Brief report: Ear in- fections in autistic and normal children. Journal of Autism and De- velopmental Disorders, 17, 585–594.

Kravits, T. R., Kamps, D. M., Kemmerer, K., & Potucek, J. (2002). Brief report: Increasing communication skills for an elementary- aged student with autism using the picture exchange communica- tion system. Journal of Autism and Developmental Disorders, 32, 225–230.

Liddle, K. (2001). Implementing the picture exchange communica- tion system (PECS). International Journal of Communication Dis- orders, 36, 391–395.

Lovaas, O. I. (1977). The autistic child: Language development through behavior modification. NY: Irvington Publishers.

Nadol, J.B. (1993) Hearing loss. The New England Journal of Medi- cine, 329, 1092–1102.

Roper, L., Arnold, P., & Monteiro, B. (2003). Co-occurrence of autism and deafness. Autism, 7, 245–253.

Rosenhall, U., Nordin, V., Sandstrom, M., Ahlsen, G., & Gillberg, C. (1999). Autism and hearing loss. Journal of Autism and Devel- opmental Disorders, 29, 349–357.

Rosenthal, R., & Jacobson, L. (1992). Pygmalion in the classroom: Teacher expectation and pupils’ intellectual development. New York: Irvington.

Smith, D. E. P., Miller, S. D., Stewart, M., Walter, T. L., & Mc- Connell, J. V. (1988). Conductive hearing loss in autistic, learn- ing-disabled and normal children. Journal of Autism and Developmental Disorders, 18, 53–65.

Stubbs, G. E., Ash, E., & Williams, C. P. S. (1984). Autism and con- genital cytomegalovirus. Journal of Autism and Developmental Dis- orders, 14, 183–189.

van Schrojenstein Lantman-de Valk, H. M. (1997). Prevalence and incidence of health problems in people with intellectual disability. Journal of Intellectual Disability Research, 41, 42–51.