J Autism Dev Disord (2008) 38:926–940

DOI 10.1007/s10803-007-0463-4

ORIGINAL PAPER

Effects of the Use of Visual Strategies in Play Groups for Children

with Autism Spectrum Disorders and their Peers
Jennifer B. Ganz Æ Margaret M. Flores

Published online: 12 October 2007

Ó Springer Science+Business Media, LLC 2007

Abstract The purpose of this study was to investigate the Visual information is thought to be more easily pro-
impact of visual strategies with preschool children with cessed by individuals with ASD than auditory information
autism spectrum disorders (ASD) and their peers during (Schopler et al. 1995). This has been demonstrated via
play group sessions. A changing-criterion design was higher performance on non-verbal intelligence tests than
implemented with three preschool-aged children with ASD verbal ones (Scherer et al. 2001) and the presence of
while they participated in play groups with four typically- hyperlexia, or advanced sight word recognition, in many
developing peers. Results indicated improvements in the individuals with ASD (Nation et al. 2006). Further, high-
use of script phrases, context-related comments, and functioning adults with ASD have noted that they respond
intervals in which speech occurred for all three partici- better to visual information and think visually and are often
pants. Results regarding unscripted phrases, responses, and agitated by auditory input that would not bother most
use of prompts were variable and are discussed. others (Grandin and Scariano 1986; Willey 1999).
Visually-based interventions include visual (e.g., pic-
Keywords Autism
Scripts
Verbal communication
ture, written) schedules, visually organized tasks, written
Visual strategies
Unscripted speech or pictorial scripts, rule reminder cards, and visual task
analyses (Simpson et al. 2008). Such interventions may be
more effective than auditory-based instruction for indi-
Deficits in social and communication skills are primary viduals with ASD (Bryan and Gast 2000; Simpson et al.
characteristics of individuals with autism spectrum disor- 2008). Visual schedules have been found to decrease
ders (ASD) (Baron-Cohen 2004). In particular, such off-task (Bryan and Gast 2000) and other disruptive
individuals may not frequently respond to peers or adults, behaviors (Dooley et al. 2001), and to improve social
initiate interactions, or speak spontaneously (Janzen 2003; skills (Morrison et al. 2002), following directions (Dooley
Krantz and McClannahan 1993). The use of visual strate- et al. 2001), and task engagement (Bryan and Gast 2000;
gies within the context of heterogeneous play groups (i.e., Dettmer et al. 2000; MacDuff et al. 1993; Massey and
groups that include both children with disabilities and Wheeler 2000). Written scripts, with or without pictures,
typically-developing children) is a strategy that may have been reported in a small number of published studies
improve such skills in children with ASD. to be effective with individuals with ASD. They have
been used to decrease aggressive behavior (Sasso et al.
1990) and to increase bids for attention (Krantz and
McClannahan 1998), conversational skills (Charlop-
J. B. Ganz
M. M. Flores
Department of Special Education, University of Texas at San Christy and Kelso 2003), social interactions (Stevenson
Antonio, San Antonio, TX, USA et al. 2000) and play abilities (Gonzalez-Lopez and
Kamps 1997; Johnston et al. 2003; Kamps et al. 1992;
J. B. Ganz (&)
Pierce and Schreibman 1995; Pierce and Schreibman
Department of ILT, UTSA, One UTSA Circle, San Antonio,
TX 78249, USA 1997; Strain et al. 1979), imitation (Pierce and Schreibman
e-mail: [email protected] 1995), and other communication skills (Krantz and

123
J Autism Dev Disord (2008) 38:926–940 927

McClannahan 1993; Loveland and Tunali 1991; Sarokoff out-of-town trips or other summer activities. Thus, ulti-
et al. 2001). mately, the group structures were different than those
Play groups that include both children with ASD and their intended and the two groups differed in that one included
typically-developing peers may be an effective means to one child with ASD, Dario, and two typically-developing
improve play and communication skills in children with peers while the other included two children with ASD, Max
ASD, however, only a few empirical studies have been and Eddie, and one typically-developing peer.
published on this subject. Use of the Integrated Play Group Criteria for participants with ASD included: (a) must be
model have resulted in improved and increased social-play between 3 and 6 years of age, (b) must have a current
skills (Lantz et al. 2004; Wolfberg and Schuler 1993; Yang diagnosis of autism or pervasive developmental disorder,
et al. 2003), functional play with toys (Wolfberg and Schuler not otherwise specified (PDD-NOS) given by a medical
1993), pretend play (Yang et al. 2003; Zurcher et al. 2001), practitioner (e.g., physician, child psychiatrist) or school
joint attention (Zurcher et al. 2001), verbal interactions diagnostician or school psychologist independently of this
(Zurcher et al. 2001), and decreased stereotyped activities research, and (c) must be able to speak in understandable
and solitary play (Wolfberg and Schuler 1993). phrases of one or more words. Participants with ASD were
The purpose of this study was to investigate the impact recruited through an online group including members of
on communication of visual strategies with preschool the local chapter of the Autism Society of America. An
children with ASD and their peers during play group ses- email was sent to all members of this group describing the
sions. The current study contributes to the small literature project and criteria for participation. Diagnoses were con-
basis for visually-based interventions and expands on firmed via a review of the children’s school records, which
previous research by implementing visual cues within the were provided by their mothers. Evaluation records were
context of dramatic play activities that include typically- sparse and included no intelligence testing. The Gilliam
developing peer models, which has not been examined Autism Rating Scale (GARS; Gilliam 1995) was completed
previously. Specific research questions included the by each child’s mother prior to the beginning of the study
following. Would participants’ use of script phrases and by the first author following the study to confirm their
increase in stepwise fashion along with the implementation diagnoses and provide more detailed information regarding
of scripts and peer instruction cards and would these results the participants’ social and communication skills. Two
generalize to two additional play themes? Would partici- additional participants with ASD were considered, but
pants’ use of unscripted phrases increase over the were unable to participate due to the dates during which the
implementation of scripts and peer instruction cards? study took place during the summer or schedule conflicts
Would participants’ use of context-related comments with other services they were receiving.
increase over the implementation of scripts and peer
instruction cards? Would participants’ use of responses to
peers and adults increase over the implementation of scripts Dario
and peer instruction cards? Would participants’ use of
speech increase over the implementation of scripts and peer Dario was a Hispanic boy who was 4 years, 6 months old
instruction cards? Would the levels of prompts used during at the time of this study. He was currently enrolled in the
the implementation of scripts be decreased over the course preschool in which the study took place and had an assis-
of implementation and generalization activities? tant with him at all times in the classroom he regularly
attended. This study took place during the summer. In
addition to 30 h of preschool each week, he received
Method occupational therapy and speech therapy each for 45 min
per week at school and behavioral therapy at home for
Participants approximately 20 h per week, implemented by his mother.
During the previous school year, in addition to speech and
Two play groups were formed. Initially, we intended to occupational therapies and behavioral therapy at home,
form two play groups, each with four children, two of Dario was enrolled in a public school preschool classroom
whom were to have ASD and two of whom were to be for children with disabilities for 15 h per week.
typically-developing. However, we encountered numerous Dario was diagnosed with high-functioning autism at
difficulties in recruiting the required number of children, two-and-a-half-years of age. According to a GARS (Gilliam
possibly due to the occurrence of this project during the 1995) completed by his mother and one completed by the
summer months. We received several inquiries from par- first author following completion of the study, Dario scored
ents of children who were subsequently unable to in the ‘‘average’’ range regarding probability of autism.
participate due to the co-occurrence of the play groups and According to the first author’s ratings, he scored in the

123
928 J Autism Dev Disord (2008) 38:926–940

range of ‘‘average’’ for the stereotyped behaviors and (Gilliam 1995) completed by his mother, Max scored in the
social interaction subtests, and ‘‘very high’’ for the com- ‘‘low’’ range regarding probability of autism and according
munication subtest regarding probability of autism. In to a GARS completed by the first author following com-
terms of stereotyped behaviors, Dario (a) often avoided eye pletion of the study, Max scored in the ‘‘below average’’
contact with peers and adults, (b) frequently engaged in range regarding probability of autism. According to the
visual self-stimulatory behaviors (e.g., repetitively throw- first author’s ratings, he scored in the range of ‘‘below
ing paper and coins in the air to watch them fall, average’’ for the stereotyped behaviors and social interac-
repetitively turning over tubes of liquids to watch them tion subtests, and ‘‘average’’ for the communication subtest
swirl around), and (c) often made repetitive, unintelligible regarding probability of autism. In terms of stereotyped
vocalizations. Dario’s communication was characterized behaviors, Max often avoided eye contact with peers and
by (a) frequent repetitions of words and sounds, immedi- adults, frequently walked on tip-toe, and occasionally
ately after hearing them and after delays; (b) unusual, engaged in repetitive tensing of his body (i.e., seated with
gravelly vocal quality; (c) difficulty responding to one-step legs stiff and straight out in front, arms bent at 90° with
directions (e.g., ‘‘sit in the chair’’); (d) use of single-word hands in fists, and face in a grimace). Max’s communica-
utterances; (e) failure to initiate verbal interactions with tion was characterized by (a) occasional repetitions of
peers or adults (e.g., requesting items, commenting, con- words and sounds, immediately after hearing them and
versations); and (f) difficulty answering simple questions after delays; (b) some difficulty responding to one-step
(e.g., although he was able to identify colors and numbers, directions (e.g., ‘‘sit by Ash’’); (c) use of complete sen-
he infrequently did so when asked). Dario’s social skills tences; (d) some difficulty answering questions about
were delayed as well. He (a) did not demonstrate a change books read-aloud; and (e) some difficulty initiating verbal
in affect when praised, (b) did not imitate peers or adults, interactions with peers or adults. Though Max frequently
(c) played alone and did not play with toys as they were made context-appropriate comments, he often ‘‘talked to
meant to be played with, (d) sometimes tried to leave the himself’’ rather than getting the attention of a peer or adult
play area in avoidance of his peers and adults, (e) occa- via eye contact, proximity, or saying someone’s name.
sionally lined objects up, and (f) failed to respond to play Max’s social skills were characterized by (a) occasional
initiations made by peers and adults. Dario’s mother withdrawal in group situations, (b) some difficulty imitat-
reported that he appeared to develop normally until ing peers in play situations, (c) a tendency to ‘‘look
approximately 10 months of age when he underwent a through’’ people who spoke directly to him, (d) difficulty
catheterization for a bladder and kidney problem. While he adjusting to changed routines and new situations and set-
was making some consonant and vowel sounds prior to tings, and (e) occasional tantrums in response to other
10 months, afterward, he no longer did. Dario had received children taking toys he wanted and adults asking him to do
no diagnoses other than high-functioning autism. His things he did not want to do. Max’s mother reported that he
hearing had been tested and was within normal ranges. demonstrated delays in speech, lack of imitation skills
Although he had been diagnosed with high-functioning (e.g., waving goodbye), compulsive behaviors and extreme
autism, his GARS score and behavior in play group indi- difficulty with transitions, and that he appeared to be deaf
cated that he appeared to have autism, particularly in when he was under 3 years of age. Max’s score on the
consideration of his difficulties communicating. GARS and our observations indicated that Max was high-
functioning, particularly considering his ability to com-
municate effectively in several-word phrases and frequent
Max social initiation.

Max was a Caucasian boy who was 4 years, 5 months old

at the time of this study. He was not enrolled in the pre- Eddie
school in which the study took place; however, he received
11 h of behavioral therapy per week at a private clinic Eddie was a Caucasian boy who was 4 years, 7 months old
during the summer. During the previous school year, Max at the time of this study. He was not enrolled in the pre-
was enrolled in a public school preschool classroom for school in which the study took place; however, he received
children with disabilities for 15 h per week, which included 30 h of behavioral therapy per week, which included 45–
approximately 30 min of speech therapy per week, and 90 min of direct one-on-one instruction per day, at a pri-
received 5 h of behavioral therapy per week at a private vate behaviorally-based school during the summer and
clinic. school year. In addition, Eddie received 1 h of private
Max was diagnosed with a speech delay at 2 years of speech therapy and 2 h of private occupational therapy per
age and autism at 3 years of age. According to a GARS week and 45 min of private physical therapy per month.

123
J Autism Dev Disord (2008) 38:926–940 929

7 below average
Eddie was diagnosed with strabismus and far-sighted-

Developmental
ness and was diagnosed with PDD-NOS at 16 months of

9 average
8 average
age. According to a GARS (Gilliam 1995) completed by his
mother, Eddie scored in the ‘‘low’’ range regarding prob-
ability of autism and according to a GARS completed by
the first author following completion of the study, Eddie

Social interaction

7 below average
scored in the ‘‘below average’’ range regarding probability
of autism. According to the first author’s ratings, he scored

10 average

9 average
in the range of ‘‘very low’’ for the stereotyped behaviors
subtest, and ‘‘average’’ for the communication and social
interaction subtests regarding probability of autism.
Regarding stereotyped behaviors, Eddie (a) often avoided

Communication
eye contact with peers and adults, (b) lined up and arranged

17 very high
10 average
10 average
toys instead of playing with them appropriately, and (c)
made high-pitched, repetitive vocalizations occasionally.
Eddie’s communication was characterized by (a) occa-
sional repetitions of words and sounds, immediately after

Stereotyped behaviors
hearing them and after delays; (b) an unusually high-pit-

GARS scores and Probability of Autism

ched voice for his age; (c) some difficulty responding to

7 below average
one-step directions; (d) use of complete sentences; (e) a

3 very low
tendency to look away or ignore an adult or peer who

9 average
spoke to him; (f) difficulty answering questions about
books read-aloud; and (g) failure to initiate verbal inter-
actions with peers or adults. Though Eddie occasionally

Table 1 Participants’ GARS autism quotients and subtest standard scores and demographic information

87 below average
82 below average
made context-appropriate comments, he often ‘‘talked to

Autism quotient
himself’’ rather than getting the attention of a peer or adult.

108 average
Eddie’s social skills were characterized by (a) frequent
withdrawal and attempts to escape in group situations, (b)
lack of imitation of peers or adults in play situations, (c) a
tendency to ‘‘look through’’ people who spoke directly to
Caucasian
Caucasian
him, (d) difficulty adjusting to changed routines and new
Hispanic

situations and settings, and (e) frequent crying in response

Race

to other children taking toys he wanted and adults asking

him to do things he did not want to do. Eddie’s mother
4 years, 6 months
4 years, 5 months
4 years, 7 months

reported that he had frequent ear infections and had tubes

implanted in his ears and that he appeared to be deaf when
he was under 3 years of age, but that his hearing had been
tested and was normal. Eddie’s score on the GARS and our
Age

observations indicated that Eddie was high-functioning,

particularly considering his ability to communicate effec-
High-functioning autism

PDD-NOS, strabismus,

tively in several-word phrases.

and far-sightedness

Table 1 provides GARS autism quotients and subtest

scores, according to the first author’s ratings, as well as
some demographic information for each participant.
Diagnoses

Developmental subtest scores were taken from each of the

Autism

participants’ mothers’ ratings.

Name (pseudonym)

Typically-Developing Peers

Four typically-developing peers participated in the play

Eddie
Dario

groups as peer models. Each play group consisted of

Max

three children. Dario’s group consisted of him and two

123
930 J Autism Dev Disord (2008) 38:926–940

typically-developing 4-year-old Caucasian girls, Judith clothes (i.e., vest, hat, bandana), stuffed animal puppets
and Kyla who were currently enrolled in the preschool in (i.e., two ducks, two pigs, two dogs, and two cows), small
which the play group took place. Max and Eddie’s group plastic animals (i.e., cows, horses, pigs, sheep, ducks,
included one typically-developing child. Geri, a 5-year-old chickens, geese, and a dog), and three toy tractors. Mate-
Caucasian girl, participated for the first 2 weeks. Ash, a rials for the ‘‘birthday party’’ included a wooden birthday
4-year-old Caucasian boy, participated for the final cake with decorations (i.e., candles, strawberries, candies)
2 weeks. Neither Geri nor Ash was currently attending the and cake slices that could be attached with hook-and-loop
preschool, though Ash had attended until a month- fasteners, party hats, plates, cups, utensils, balloons, and
and-a-half prior to the study. Preference was given to gift bags with tissue paper. Materials for the ‘‘bakery’’
children currently attending the preschool when choosing included a basket with a variety of toy breads (e.g., bread
typically-developing peers. Criteria for typically-developing sticks, buns, loaves, cupcakes), the wooden birthday cake,
participants included: (a) must be between 4 and 5 years old, toy baking utensils (e.g., spoons, cake pan, pastry cutter,
(b) must not have an identified disability or suspected delays, rolling pin, muffin tin, cookie cutters, bowls, spatula), and a
(c) must be able to speak in intelligible sentences of four or toy oven. Max and Eddie’s play group participated in all
more words, (d) should be children with good attendance four themes, though data was collected only during the first
and who attend full-time (criterion d applied only to those three and Dario’s play group participated in all themes
children attending the preschool but the necessity for good except ‘‘birthday party’’ because he was ill and was absent
attendance during the study was explained to parents of from preschool for the entire week (week 3) that theme
children who did not previously attend the preschool). occurred.
The researchers developed scripts to encourage theme-
related speech in the children with ASD during each play
Setting, Play Themes, and Materials group session. The scripts that the participants with ASD
used during intervention were printed on 2½ by 7½ inch
This study took place in a private preschool for typically- white paper mounted on 3½ by 8½ inch colored paper. The
developing children, though the preschool did enroll some script cards were laminated and each had a 1–2 inch picture
children with disabilities and developmental delays. Spe- (i.e., line drawing or photo) corresponding with the words
cifically, the play groups took place in a separate, spare written on them, which were between 1 and 6 word phra-
classroom, which was connected to a classroom for infants. ses. In addition, during intervention, the typically-
That is, while they were two separate classrooms, the play developing peers were introduced to peer instruction cards
group classroom and the infant room were connected by a printed on 8½ by 5½ inch paper, which included pictures
diaper-changing area and had only a half-wall between (i.e., line drawings or photos) and written instructions on
them. Therefore, if the infants were crying, it was easily how to interact with their peers with ASD. Scripts and peer
heard in the play group classroom. During baseline and instruction cards are described below in the script
intervention, the children played in an enclosed play area instruction section.
measuring about 55 square feet, surrounded on two sides
by child-sized furniture, approximately 2–3 feet tall, and on
two sides by walls. Procedure
Three play themes were implemented for each group.
All of the children participated in an ‘‘ocean theme park’’ The play groups occurred for 4 weeks, for 30 min a day,
theme for intervention. Max and Eddie’s play group par- for 4–5 days each week. Data collection for Max and Eddie
ticipated in farm and birthday party themes for occurred during the first 3 weeks of their play group. Dario
generalization. Dario’s group participated in farm and was ill during the entire third week of his play group, so his
bakery themes for generalization. These were chosen with data were collected during weeks 1, 2, and 4. Data were
input from the participants’ mothers regarding their pre- collected in 5-min sessions during thematic play. Data
ferred activities (e.g., the participants lived near an ocean- collection sessions occurred 1–3 times per day. A changing
themed amusement park and had visited) and those with criterion single-subject design was used to demonstrate
which they were familiar (e.g., one participant frequently collateral changes in the use of scripted phrases in relation
engaged in echolalia related to birthday parties). Materials to changes in task requirements (Hartmann and Hall 1976).
available for the ‘‘ocean theme park’’ included stuffed- Play themes were introduced approximately one per week;
animal puppets (i.e., whale, dolphin, fish, crab, turtle, and however, all criterion changes were made during the first
octopus), a cash register with play money and tickets for thematic play unit, ‘‘ocean theme park’’, for all three par-
the park, a blow-up child’s pool, play food, plates, cups, ticipants; thus, this theme took slightly longer than a week
and utensils. Materials for the ‘‘farm’’ included farm-hand to complete. Each participant progressed to the next

123
J Autism Dev Disord (2008) 38:926–940 931

criterion when he met the current criterion for three consec- with the adult. While the peer instruction cards were posted
utive sessions. The criteria for each phase began at one on the walls within the play area, beyond initial daily
interval (i.e., one script card was implemented within a single instruction, the peer instruction cards were not pointed out
interval) and increased by one interval for Dario and Eddie by the adults during play group sessions.
due to the low amounts of speech used during baseline. The The lengths of the scripts were individualized for each
criteria for each phase began at two intervals (i.e., two script participant based on his verbal and reading abilities.
cards were implemented, each during a separate interval) and Furthermore, because Max used more context related and
increased by two intervals for the second criterion and by one unscripted phrases during baseline, he was introduced to
for the third criterion for Max because he spoke more than the twice as many script phrases during the first two criteria.
other two participants during baseline. All sessions were Script instruction for the participants with ASD took place
videotaped to enable data collection. within the 15 min prior to the play group meeting in which
Visual inspection was used to interpret the graphed results. each new play theme (i.e., prior to intervention and gen-
Additionally, percentage of nonoverlapping data (PND; eralization) was introduced. In addition, immediately prior
Scruggs and Mastropieri 1998) were calculated for each to each criterion change in intervention, new scripts were
participant for each category of data. PND is calculated by introduced. That is, within the 15 min prior to play group,
dividing the number of intervention data points that are higher Dario and Eddie were each initially taught one script
than the highest baseline data point by the total number of phrase at a time, then a second script phrase was taught to
intervention data points and multiplying the result by 100. each participant prior to the increase of the criterion to two
According to Scruggs and Mastropieri, PND scores over 90% intervals, then a third script phrase was taught immediately
indicate very effective strategies, scores between 70% and prior to the increase to the criterion to three intervals.
90% are considered effective, those between 50% and 70% Within the 15 min immediately prior to each generalization
are questionable, and those lower than 50% are not effective. play theme, they were each taught the three script phrases
relevant to those themes. Max’s instruction followed the
same pattern, however, he was first taught 2 phrases, then
Baseline two more (criterion = 4 intervals), then one more (crite-
rion = 5 intervals). Each script was practiced with each
Data were collected for all participants during the ‘‘ocean participant at least five times or until he was able to recite it
theme park’’ theme. The children were told they were ‘‘here independently upon being shown the script. Samples of
to play’’ and were shown the materials and were given some script phrases for each participant for each theme are
suggestions for what they could do with each item (e.g., ‘‘the illustrated in Table 2.
cash register is for buying and selling park tickets’’). If any
children left the play area, they were returned and told that it
was time to play. If any children tried to engage the adults in Script Intervention
conversation, they were reminded to talk to and play with
their friends and their questions were answered briefly. Geri Each participant with ASD participated in the ‘‘ocean
participated in Max and Eddie’s play group during baseline theme park’’ play theme for intervention. Procedures in
and script intervention. place during baseline continued. In addition, during each
5-min session, typically-developing peers were reminded
of the directions on the peer instruction cards. Script
Script Instruction phrases were introduced during intervention based on the
current criterion level for each participant (e.g., when the
Following baseline, the typically-developing peers were criterion was one phrase/interval, the script card was shown
told that their friends with ASD would need their help to the participant at least once during that session). Criteria
learning to talk and play. Then, within the 15 min prior to were increased when the participant had met the current
intervention and generalization for each new play theme, level for three consecutive sessions. No more than one
the typically-developing peers were introduced to one peer script phrase was held in front of a participant per 20-s
instruction card. The peer instruction cards included the interval; however, the script cards were placed in the play
following instructions: (a) to get your friend’s attention, area near relevant materials. If participants were already
talk to him and hand him a toy; (b) play with the toys your verbally interacting with peers or spontaneously and
friend is playing with; and (c) when your friend talks, you independently recited script phrases, these interactions
say something too. The typically-developing peers were were not interrupted. Each script card was used once per
shown a peer instruction card, told what it said, asked to session and no more than one script card was prompted
verbally repeat it, and then they role-played the instruction during an interval. Scripts were shown to the participants

123
932 J Autism Dev Disord (2008) 38:926–940

Table 2 Samples of script phrases for each participant

Theme 1 (Ocean theme park) Theme 2 (Farm) Theme 3 (Dario: Bakery; Max and Eddie: Birthday)

Dario A fish! A duck. Rolls, please.

Max Look at the dolphins. The horse is running. Let’s decorate the cake.
Eddie Do you see the purple fish? ‘‘Neigh, neigh,’’ says the horse. Blow out the candles.

according to the following prompt sequence/levels, waiting seven percent of Dario’s intervention and generalization
5 s between each step: sessions were observed for procedural integrity, which was
100% for each session. Twenty-five percent of Max’s
1. Held script card up in front of participant, about 2–3
intervention and generalization sessions were observed for
feet away from face,
procedural integrity, which was 100% for each session.
2. Hand-over-hand helped participant point to the card,
Twenty-seven percent of Eddie’s intervention and gener-
3. Verbally modeled the words written on the card,
alization sessions were observed for procedural integrity,
4. Told the participant to, ‘‘Say _____ [what the card
which was 100% for each session.
says].’’
Prompts were delivered by the first and second authors
and a trained undergraduate research assistant. Prompting Response Definitions
stopped when the participant recited the script phrase. If
the participant did not respond after all of the steps were Data were collected via interval recording. Intervals were
completed, the instructor waited at least one interval before each 20 s long and there were 15 intervals per session.
trying again. Whenever possible, the scripts were shown to Each interval was observed to determine whether or not the
the participants when they were already in close proximity participant used script phrases and/or unscripted phrases.
or playing with the related materials. Both scripted and unscripted phrases could be recorded as
occurring during a single interval; however, the frequency
of these statements was not recorded. Only the occurrence
Generalization or non-occurrence of each within the interval was recorded.
Script phrases were recorded as occurring when an interval
Procedures implemented during intervention were the same included words, phrases, or sentences that were word-for-
as those implemented during generalization, however, word the same as the phrases listed on the participants’
prompting decreased and occurred at less intrusive levels. written scripts or within one word of the written scripts.
The final criterion for intervention was used to determine Any other intelligible words, phrases, or sentences were
how many script phrases to implement in the following two recorded as unscripted phrases. In addition, intervals were
generalization play themes. The participants were taught all recorded as including context-related comments whenever
of the script phrases for each new theme at one time, unlike a script phrase was used and when an unscripted phrase
during intervention, when participants learned only one or used during an interval related to the current theme, an
two phrases until mastery of that criterion level. Dario’s play action performed by a peer or adult, or something said by a
group participated in the ‘‘farm’’ and ‘‘bakery’’ during gen- peer or adult during the play group.
eralization. Max and Eddie’s play group participated in the Intervals were also recorded as including responses
‘‘farm’’ and ‘‘birthday party’’ during generalization. Ash’s when the participant answered a peer’s or adult’s question
participation began during the ‘‘farm’’ play theme. within 5 s or immediately imitated a peer’s or adult’s
speech. Percentages of intervals in which each of the above
response definitions occurred were calculated for each
Procedural Integrity session. In addition, the percentage of intervals in which
speech occurred was calculated for each session and
Procedural integrity was observed based on the prompt included all intervals in which any of the above response
sequence described above in the script intervention section. definitions occurred. That is, each session was analyzed
Instructors were rated overall for the session to determine if overall for the occurrence or non-occurrence of any verbal
the steps were implemented correctly and if the instructors communication and this was calculated as a percentage of
waited 5 s between each step. Procedural integrity was the total possible intervals (i.e., an interval that contained
calculated by dividing the number of steps implemented the coding of the occurrence of any of the above response
correctly by the total possible steps completed. Twenty- definitions was coded as an occurrence of speech). Finally,

123
J Autism Dev Disord (2008) 38:926–940 933

the level of prompt required was recorded each time a during the third criterion level (33%), 35% (range = 33–
participant used a script phrase, according to the defini- 40%) during the first generalization play theme (33%), and
tions of each prompt level given above in the script 44% (range = 33–53%) during the second generalization
intervention section. play theme (33%). Eddie’s use of script phrases during
intervention averaged 0% of the intervals observed during
baseline, and met or exceeded criterion levels for each
Inter-Observer Reliability session. His use of script phrases averaged 7% (range =
7%) during the first criterion level (7%), 20% (range =
Inter-observer reliability was collected by the first and 13–27%) during the second criterion level (13%), 25%
second authors. Both have terminal degrees and experience (range = 20–27%) during the third criterion level (20%),
in conducting and collecting data on single-subject 31% (range = 27–40%) during the first generalization play
research. Reliability was calculated by dividing the number theme (20%), and 22% (range = 20–27%) during the sec-
of agreements by the total number of agreements plus ond generalization play theme (20%). There was no
disagreements, multiplied by 100 (Kazdin 1982). An overlap between baseline and intervention data points for
agreement was scored when both observers independently any of the participants, thus, the PND was 100%, indicating
recorded the same code in an interval (e.g., script phrase a very effective treatment.
[S], unscripted phrase [U], context-related comment [C],
responses [R]). Reliability was assessed during baseline
sessions for 33% of Dario’s, Max’s, and Eddie’s sessions Unscripted Phrases
and was assessed at 95%, 92%, and 89% agreement
respectively. Reliability was assessed during intervention Figure 1 illustrates each participant’s results for unscripted
and generalization sessions for 33% of Dario’s, 29% of phrases. Dario’s use of unscripted phrases averaged 9%
Max’s, and 33% of Eddie’s sessions and was assessed at (range = 0–13%) of the intervals observed during baseline.
mean agreements of 98%, 89%, and 89% respectively. His use of unscripted phrases during intervention averaged
Agreement was not assessed on prompting level. 2% (range = 0–7%) during the first criterion level, 5%
(range = 0–7%) during the second criterion level, 7%
(range = 0–13%) during the third criterion level, 2%
Results (range = 0–7%) during the first generalization play theme,
and 4% (range = 0–13%) during the second generalization
Script Phrases play theme. Max’s use of unscripted phrases averaged 40%
(range = 33–47%) of the intervals observed during base-
Figure 1 illustrates each participant’s results for script line. His use of unscripted phrases during intervention
phrases. A functional relation was demonstrated between averaged 89% (range = 87–93%) during the first criterion
the intervention and the participants’ behaviors. There was level, 71% (range = 60–80%) during the second criterion
a successive increase in scripted verbalizations and all but level, 53% (range = 47–60%) during the third criterion
one data point met or exceeded the established criterion for level, 49% (range = 13–80%) during the first generaliza-
all participants. Dario’s use of script phrases averaged 0% tion play theme, and 60% (range = 40–73%) during the
of the intervals observed during baseline, and met or second generalization play theme. Eddie’s use of unscrip-
exceeded criterion levels for each intervention and gener- ted phrases averaged 24% (range = 13–33%) of the
alization session. His use of script phrases during intervals observed during baseline. His use of unscripted
intervention averaged 7% (range = 7%) during the first phrases during intervention averaged 16% (range = 7–
criterion level (7%), 27% (range = 13–40%) during the 20%) during the first criterion level, 9% (range = 0–27%)
second criterion level (13%), 25% (range = 20–27%) dur- during the second criterion level, 27% (range = 0–40%)
ing the third criterion level (20%), 27% (range = 27%) during the third criterion level, 40% (range = 33–47%)
during the first generalization play theme (20%), and 20% during the first generalization play theme, and 29%
(range = 20%) during the second generalization play (range = 7–40%) during the second generalization play
theme (20%). Max’s use of script phrases averaged 0% of theme.
the intervals observed during baseline, and met or exceeded To supplement the visual analysis, PND calculated for
criterion levels for all intervention and generalization ses- Dario reveals a score of 0% for unscripted phrases, a score
sions except one. His use of script phrases during of 82% for Max’s unscripted phrases, and a score of 22%
intervention averaged 22% (range = 20–27%) during the for Eddie’s unscripted phrases. The PND for unscripted
first criterion level (13%), 29% (range = 20–33%) during phrases suggests an ineffective treatment for Dario and
the second criterion level (27%), 33% (range = 33%) Eddie and an effective treatment for Max.

123
934 J Autism Dev Disord (2008) 38:926–940

Fig. 1 Script phrases, context- baseline 1 interval 2 intervals 3intervals generalization

100%
related comments, and

Percentage of Intervals Observed

unscripted phrases 90%

80%

70%

60% Script phrases

50% Context-related comments
40% Unscripted phrases

30%

20%

10%

0%
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18

Session Number Dario

baseline 2 4 intervals 5 generalization

1 00% intervals intervals
Percentage of Intervals Observed

90%

80%

70%

60% Script phrases

50% Context-related comments
40% Unscripted phrases
30%

20%

10%

0%
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20

Session Number Max

baseline 1 interval 2 3 generalization
100% intervals intervals
Percentage of Intervals Observed

90%

80%

70%

60% Script phrases

50% Context related phrases
40% Unscripted phrases
30%

20%

10%

0%
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18

Session Number Eddie

Context-Related Comments across all participants. Dario’s use of context-related com-

ments averaged 0% of the intervals observed during baseline.
Figure 1 illustrates each participant’s results for context- His use of context-related comments during intervention
related comments. A functional relation was demonstrated averaged 7% (range = 7%) during the first criterion level,
between the intervention and context-related comments 29% (range = 13–40%) during the second criterion level,
across all participants. There were only 2 overlapping data 25% (range = 20–27%) during the third criterion level, 27%
points (6%) between baseline and intervention. There was a (range = 27%) during the first generalization play theme, and
successive increase in the use of context-related comments 20% (range = 20%) during the second generalization play

123
J Autism Dev Disord (2008) 38:926–940 935

theme. Max’s use of context-related comments averaged 9% To supplement the visual analysis, PND calculated for
(range = 0–20%) of the intervals observed during baseline. Dario reveals a score of 0% for responses, a score of 73% for
His use of context-related comments during intervention Max’s responses, and a score of 22% for Eddie’s responses.
averaged 82% (range = 73–93%) during the first criterion The PND for responses suggests an ineffective treatment
level, 85% (range = 80–93%) during the second criterion for Dario and Eddie and an effective treatment for Max.
level, 73% (range = 67–80%) during the third criterion level,
71% (range = 47–93%) during the first generalization play
theme, and 76% (range = 60–87%) during the second gen- Intervals in which Any Speech Occurred
eralization play theme. Eddie’s use of context-related
comments averaged 5% (range = 0–7%) of the intervals Figure 2 illustrates each participant’s results for intervals
observed during baseline. His use of context-related com- with any speech. Intervals in which Dario used any speech
ments during intervention averaged 11% (range = 7–13%) averaged 11% (range = 0–20%) of the intervals observed
during the first criterion level, 25% (range = 20–27%) during during baseline. His use of speech during intervention aver-
the second criterion level, 38% (range = 27–60%) during the aged 9% (range = 7–13%) during the first criterion level,
third criterion level, 42% (range = 33–53%) during the first 29% (range = 13–40%) during the second criterion level,
generalization play theme, and 47% (range = 20–60%) 31% (range = 20–40%) during the third criterion level, 27%
during the second generalization play theme. (range = 20–33%) during the first generalization play theme,
To supplement the visual analysis, PND calculated for and 24% (range = 20–33%) during the second generalization
Dario reveals a score of 100% for context-related phrases, a play theme. Intervals in which Max used any speech aver-
score of 100% for Max’s context-related phrases, and a aged 42% (range = 40–47%) of the intervals observed
score of 89% for Eddie’s context-related phrases. The PND during baseline. His use of speech during intervention aver-
for context-related phrases suggests a very effective treat- aged 95% (range = 93–100%) during the first criterion level,
ment for Dario and Max and an effective treatment for 89% (range = 80–93%) during the second criterion level,
Eddie. 78% (range = 73–80%) during the third criterion level, 75%
(range = 60–93%) during the first generalization play theme,
and 82% (range = 60–93%) during the second generalization
Responses play theme. Intervals in which Eddie used any speech aver-
aged 24% (range = 13–33%) of the intervals observed
Figure 2 illustrates each participant’s results for responses. during baseline. His use of speech during intervention aver-
Dario’s use of responses averaged 2% (range = 0–7%) of aged 24% (range = 13–33%) during the first criterion level,
the intervals observed during baseline. His use of responses 33% (range = 20–53%) during the second criterion level,
during intervention averaged 0% during the first criterion 62% (range = 27–80%) during the third criterion level, 71%
level, 0% during the second criterion level, 7% (range = 0– (range = 67–73%) during the first generalization play theme,
13%) during the third criterion level, 0% during the first and 53% (range = 27–73%) during the second generalization
generalization play theme, and 2% (range = 0–7%) during play theme.
the second generalization play theme. Max’s use of To supplement the visual analysis, PND calculated for
responses averaged 4% (range = 0–13%) of the intervals Dario reveals a score of 44% for use of any speech, a score
observed during baseline. His use of responses during of 100% for Max’s use of any speech, and a score of 33%
intervention averaged 22% (range = 13–33%) during the for Eddie’s use of any speech. The PND for use of any
first criterion level, 26% (range = 13–40%) during the speech suggests an ineffective treatment for Dario and
second criterion level, 22% (range = 7–40%) during the Eddie and a very effective treatment for Max.
third criterion level, 20% (range = 20%) during the first
generalization play theme, and 11% (range = 0–20%)
during the second generalization play theme. Eddie’s use of Prompts
responses averaged 4% (range = 0–13%) of the intervals
observed during baseline. His use of responses during Figure 3 illustrates the highest and lowest levels of
intervention averaged 4% (range = 0–13%) during the first prompts given to each participant in each session. Dario’s
criterion level, 0% during the second criterion level, 25% prompts ranged from level 1 to level 4 during the first
(range = 0–47%) during the third criterion level, 22% criterion level for intervention, from level 0 to level 4 in the
(range = 7–47%) during the first generalization play second criterion level, from 0 to 3 in the third criterion
theme, and 11% (range = 0–27%) during the second gen- level, from 0 to 3 in the first generalization play theme, and
eralization play theme. from 2 to 4 in the final generalization theme. Max’s

123
936 J Autism Dev Disord (2008) 38:926–940

Fig. 2 Responses and intervals baseline 1 interval 2 3 generalization

with any speech 100%
90%

Percent age of Int ervals O bserved

80%
70%
60%
Responses
50%
Intervals with Any Speech
40%
30%
20%
10%
0%
1 3 4 6 7 9 10 12 13 15 16 18
Session Number Dario

baseline 2 intervals 4 5 generalization

100%
90%
Percent age of Int ervals O bserved

80%
70%
60%
Responses
50%
Intervals with Any Speech
40%
30%
20%
10%
0%
1 3 4 6 7 9 11 12 14 15 17 18 20
Max
Session Number

baseline 1 interval 2 3 generalization

100%

90%
Percent age of Int ervals O bserved

80%

70%
60%
Responses
50%
Intervals with Any Speech
40%
30%

20%
10%

0%
1 3 4 6 7 9 10 12 13 15 16 18
Eddie
Session Number

prompts ranged from level 0 to level 2 during the first to 3 in the first generalization play theme, and from 1 to 4
criterion level for intervention, from level 0 to level 4 in the in the final generalization theme.
second criterion level, from 1 to 3 in the third criterion
level, from 1 to 3 in the first generalization play theme, and
from 0 to 3 in the final generalization theme. Eddie’s Discussion
prompts were all at level 1 during the first criterion level
for intervention, from level 1 to level 4 in the second cri- Consistent with the research literature in the area of het-
terion level, from 1 to 4 in the third criterion level, from 1 erogeneous play groups, the participants’ play behaviors

123
J Autism Dev Disord (2008) 38:926–940 937

Fig. 3 Levels of prompts 1 phrase 2 phrases 3 phrases generalization

4

Level of Prom pt
Lowest Level of Prompt
2
Highest Level of Prompt

0
4 5 6 7 8 9 10 11 12 13 14 15 16 17 18
Session Number
Dario

2 4 phrases 5 generalization
4 phrases phrases

3
Level of Prompt

Lowest Level of Prompt

2
Highest Level of Prompt

0
4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20
Session Number Max

1 phrase 2 phrases 3 phrases generalization

4

3
Level of Prom pt

Lowest Level of Prompt

2
Highest Level of Prompt

0
4 5 6 7 8 9 10 11 12 13 14 15 16 17 18
Session Number
Eddie

increased with regard to their context-related language which visually-based interventions led to increases in
(Lantz et al. 2004; Wolfberg and Schuler 1993; Yang et al. communication, language, and social skills (Charlop-
2003; Zurcher et al. 2001). This study extended the line of Christy and Kelso 2003; Gonzalez-Lopez and Kamps 1997;
research on heterogeneous play groups by including a Johnston et al. 2003; Kamps et. al. 1992; Krantz and
visually-based language intervention. A functional relation McClannahan 1998; Krantz and McClannahan 1993;
was demonstrated between the visually-based intervention Loveland and Tunali 1991; Pierce and Schreibman 1997;
and play-related language across play themes as well as for Pierce and Schreibman 1995; Sasso et al. 1990; Sarokoff
all participants. This is consistent with previous research in et al. 2001; Stevenson et al. 2000; Strain et al. 1979). This

123
938 J Autism Dev Disord (2008) 38:926–940

study extended the lines of research in the areas of integrated higher quantity and quality than the comments he made
play and visually-based interventions by demonstrating during baseline.
increases in language by implementing components of the two Responses to peers and adults were recorded for each
interventions. participant to determine if the peer instruction cards used
Script phrases systematically increased for all three with the typically-developing peers resulted in increased
participants as the script intervention was implemented. responding from the children with ASD. PND results for
PND results for scripted phrases for all three participants responses indicate that this is not an effective intervention
indicate that this is a very effective intervention. Also for two of the participants, Dario and Eddie. Dario, who
importantly, each participant occasionally exceeded crite- spoke the least of the three participants with ASD, only
rion levels during intervention phases. This occurred at responded three times, once during baseline, once during
times when participants demonstrated immediate echolalia the final criterion level of intervention, and once during
that overlapped use of the same script phrase into following generalization. Dario’s responses included saying, ‘‘no,’’
intervals, at time due to delayed echolalia occurring within and echoing words said by peers or adults. Max did dem-
later intervals in the same session, or when participants onstrate a modest increase in responses from baseline to
independently picked up script cards and recited their intervention and generalization and Eddie demonstrated a
scripts. Independent use of scripts in preschool children modest increase in responses in the final criterion increase
with ASD is encouraging, particularly when considering of intervention and during generalization. This may be due
the viability of the use of this strategy with older individ- to the difference in the makeup of their play group.
uals with ASD and the possibility of using scripts Because they had only one typically-developing peer, that
inconspicuously with peers, rather than relying on prompts child was more likely to interact with his or her playmates
and cues from adults. Results regarding script phrases were with ASD. In addition, Max frequently used context-related
maintained during all generalization sessions. comments during intervention and generalization, which
Unscripted phrases remained at low rates from baseline may have encouraged Geri and Ash to interact with him.
through intervention and generalization for Dario, Similarly, Eddie spoke more and made more context-
increased from baseline to intervention then slightly related comments during the final criterion level and during
decreased during generalization for Max, and remained generalization, which also may have encouraged more
approximately the same for Eddie. Further, PND results for initiations from Geri and Ash. It is also possible that the
unscripted phrases for two participants indicate that this is difference in makeup of the two groups may have impacted
an ineffective intervention and indicate that it was an in- the participants’ performance. Placing two children with
effective intervention for one participant, Max. However, ASD in the same group provides fewer socially-competent
changes in the amount of context-appropriate comments peer models than placing one child with ASD in a group
may be more important to consider in terms of the abilities with two socially-competent models; however, in this
of children with ASD to ‘‘fit in’’ in play situations with example, Max and Eddie were high-functioning and Max,
their peers. PND results for context-related phrases for the in particular, initiated communication with his peer with
three participants indicate that this is a very effective or ASD and typically-developing peers.
effective intervention. In particular, though Max and Eddie Intervals in which any speech occurred increased for all
exhibited moderate rates of unscripted phrases during participants throughout intervention and generalization;
baseline, those statements were rarely related to the play however, PND results for use of any speech indicate that
theme, actions of peers or adults in the room, or related to this is a very effective intervention for only one participant,
what peers or adults in the play area were talking about. Max, and an ineffective treatment for the other participants.
Context-related comments increased steadily from baseline Dario’s speech increased during the second criterion level
to intervention and to generalization for Dario, closely of intervention and those levels were maintained during the
following his progress in script phrases. Context-related third criterion level and generalization. Max’s speech
comments were also low for Max and Eddie during base- increased and was maintained throughout intervention and
line and increased during intervention and generalization. generalization, however, observers noted that while he
While Max and Eddie’s context-related comments were not spoke more often than the other two participants with ASD,
as closely matched with script phrases as were Dario’s, he did not always speak to anyone in initial sessions of
there is a clear relationship between an increase in script intervention, instead speaking to himself, though this
phrases and an increase in context-related comments. In improved, particularly during generalization. Eddie’s
fact, Max’s context-related comments, script phrases, and speech increased during the final criterion level and was
unscripted phrases all increased dramatically following maintained through generalization.
intervention implementation, demonstrating that the speech Ideally, prompts would have been decreased to level 0
he used during intervention and generalization was of (self-prompt) or 1 (held script card in front of participant)

123
J Autism Dev Disord (2008) 38:926–940 939

in all sessions by the end of intervention. This did not Acknowledgments This study was funded by a grant from the
occur for any of the participants with ASD. However, both College of Education and Human Development at the University of
Texas at San Antonio. The authors wish to express their gratitude to
Dario and Max did, at times, self-prompt by picking up the the parents of the participants for allowing us to play with their
script phrase cards and reciting the scripts (sessions 9, 11, children and to the Director, Alissa Levey Baugh, and staff at the
12, and 14 for Dario and sessions 5–7, 9, 11, and 20 for Block and Dreeben School for Young Children for allowing us to
Max). While Eddie did not self-prompt with the scripts, he implement the project on their site and for their ongoing support
throughout the project and to Heidi Hawbecker who assisted in video
usually required no more than a level 1 prompt. It is recording and implementation of the play groups.
possible that with a longer intervention or with older
participants, prompts may be decreased to all level 0 or 1
prompts. References
This study was limited in that it only involved three
children with ASD and four typically-developing peers. In Baron-Cohen, S. (2004). Autism: Research into causes and interven-
addition, results may only be generalized to children of a tion. Pediatric Rehabilitation, 7, 73–78.
Bryan, L. C., & Gast, D. L. (2000). Teaching on-task and on-schedule
similar age and similar communication skills. Dario’s behaviors to high-functioning children with autism via picture
peers, Judith and Kyla, initiated toward him infrequently, activity schedules. Journal of Autism & Developmental Disor-
which may have limited his opportunities to communicate. ders, 30, 553–567.
This may have been due to two factors, the makeup of his Charlop-Christy, M. H., & Kelso, S. E. (2003). Teaching children
with autism conversational speech using a cue card/written script
play group and his communication abilities. Judith and program. Education and Treatment of Children, 26, 108–127.
Kyla were close friends prior to this play group and easily Dettmer, S., Simpson, R. L., Myles, B. S., & Ganz, J. B. (2000). The
interacted with each other, but did not easily include Dario, use of visual supports to facilitate transitions of students with
possibly due to his infrequent use of speech, difficulty autism. Focus on Autism & Other Developmental Disabilities,
15, 163–169.
understanding what he did say, and his infrequent attempts Dooley, P., Wilczenski, F. L., & Torem, C. (2001). Using an activity
to interact with them, despite visual and verbal reminders schedule to smooth school transitions. Journal of Positive
each session regarding their jobs listed on the peer Behavior Interventions, 3, 57–61.
instruction cards. Gilliam, J. E. (1995). Gilliam autism rating scale. Austin: PRO-ED,
Inc.
Future research may expand this intervention for use Gonzalez-Lopez, A., & Kamps, D. (1997). Social skills training to
with older individuals with ASD (e.g., elementary-aged, increase social interactions between children with autism and
high school students, adults) and to individuals with less their typical peers. Focus on Autism and Other Developmental
verbal ability or more. Future research may also investigate Disabilities, 12, 12–14.
Grandin, T., & Scariano, M. M. (1986). Emergence: Labeled autistic.
how to best implement integrated play groups between New York: Warner Books.
preschoolers with ASD and typically-developing peers, Hartmann, D., & Hall, R. V. (1976). The changing criterion design.
particularly how to increase the initiations of typically- Journal of Applied Behavior Analysis, 9, 527–532.
developing children to their peers with ASD (e.g., via self- Janzen, J. E. (2003). Understanding the nature of autism: A guide to
the autism spectrum disorders (2nd ed.). San Antonio: Therapy
monitoring procedures or reinforcement), providing train- Skill Builders.
ing on autism to desensitize typically-developing peers to Johnston, S., Nelson, C., Evans, J., & Palazolo, K. (2003). The use of
unfamiliar characteristics and behaviors, and on what types visual supports in teaching young children with autism spectrum
of visual strategies are most effective with typically- disorder to initiate interactions. Augmentative and Alternative
Communication, 19, 86–103.
developing peers to increase their initiations toward peers Kamps, D., Leonard, B., Vernon, S., Dugan, E., Delquadri, J., Gerson,
with ASD. Further, one-on-one pairs that include one child B., Wade, L., & Folk, L. (1992). Teaching social skills to
with ASD and one typically-developing child may be most students with autism to increase peer interactions in an integrated
effective and efficient in increasing interactions. As inter- first grade classroom. Journal of Applied Behavior Analysis, 25,
281–288.
actions with the child with ASD and a typically-developing Kazdin, A. E. (1982). Single-case research designs: Methods for
peer increase, perhaps the number of play group members clinical and applied settings. New York: Oxford University
could be increased successively to resemble more a typical Press.
play situation found in inclusive settings. A longer imple- Krantz, P., & McClannahan, L. (1993). Teaching children with autism
to initiate to peers: Effects of a script-fading procedure. Journal
mentation of visual scripts could lead to decreased prompts of Applied Behavior Analysis, 26, 121–132.
and independent use of scripts. Further, future studies may Krantz, P. J., & McClannahan, L. E., (1998). Social interaction
investigate the use of generic phrases that could be used in skills for children with autism: A script-fading procedure for
multiple contexts. Finally, investigations of the use of beginning readers. Journal of Applied Behavior Analysis, 31,
191–202.
visual cues may have wide application in a variety of Lantz, J. F., Nelson, J. M., & Loftin, R. I. (2004). Guiding children
contexts in which individuals with ASD are integrated with with autism in play: Applying the integrated play group model in
their peers. school settings. TEACHING Exceptional Children, 31, 8–14.

123
940 J Autism Dev Disord (2008) 38:926–940

Loveland, K., & Tunali, B. (1991). Social scripts for conversational Scherer, M., Pierce, K. L., Paredes, S., Kisacky, K. L., Ingersoll, B., &
interactions in autism and Down’s syndrome. Journal of Autism Schreibman, L. (2001). Enhancing conversational skills in children
and Developmental Disorders, 21, 177–186. with autism via video technology. Which is better, ‘Self’ or ‘Other’
MacDuff, G. S., Krantz, P. J., & McClannahan, L. E. (1993). as a model? Behavior Modification, 25, 140–158.
Teaching children with autism to use photographic activity Schopler, E., Mesibov, G. B., & Hearsey, K. (1995). Structured teaching
schedules: Maintenance and generalization of complex response in the TEACCH system. In E. Schopler & G. B. Mesibov (Eds.),
chains. Journal of Applied Behavior Analysis, 26, 89–97. Learning and cognition in autism (pp. 243–267). New York:
Massey, N. G., & Wheeler, J. J. (2000). Acquisition and generaliza- Plenum.
tion of activity schedules and their effects on task engagement in Scruggs, T. E., & Mastropieri, M. A. (1998). Summarizing single-
a young child with autism in an inclusive pre-school classroom. subject research: Issues and applications. Behavior Modification,
Education and Training in Mental Retardation and Develop- 22, 221–242.
mental Disabilities, 35, 326–335. Simpson, R. L., Myles, B. S., & Ganz, J. B. (2008). Efficacious
Morrison, R. S., Sainato, D. M., BenChaaban, D., & Endo, S. (2002). interventions and treatments for learners with autism spectrum
Increasing play skills of children with autism using activity disorders. In R. L. Simpson & B. S. Myles (Eds.), Educating
schedules and correspondence training. Journal of Early Inter- children and youth with autism: Strategies for effective practice
vention, 25, 58–72. (2nd ed., pp. 477–512). Austin, TX: Pro-Ed, Inc.
Nation, K., Clarke, P., Wright, B., & Williams, C. (2006). Patterns of Stevenson, C. L., Krantz, P. J., & McClannahan, L. E. (2000). Social
reading ability in children with autism spectrum disorder. interaction skills for children with autism: A script-fading
Journal of Autism and Developmental Disorders, 36, 911–919. procedure for nonreaders. Behavioral Interventions, 15, 1–20.
Pierce, K., & Schreibman, L. (1995). Increasing complex social Strain, P., Kerr, M., & Ragland, E. (1979). Effects of peer-mediated
behaviors in children with autism: Effects of peer-implemented social initiations and prompting/reinforcement procedures on the
pivotal response training. Journal of Applied Behavior Analysis, social behavior of autistic children. Journal of Autism and
28, 285–295. Developmental Disorders, 9, 41–54.
Pierce, K., & Schreibman, L. (1997). Multiple peer use of pivotal Willey, L. H. (1999). Pretending to be normal: Living with
response training to increase social behaviors of classmates with Asperger’s syndrome. London: Jessica Kingsley Publishers.
autism: Results from trained and untrained peers. Journal of Wolfberg, P. J., & Schuler, A. L. (1993). Integrated play groups: A
Applied Behavior Analysis, 30, 157–160. model for promoting the social and cognitive dimensions of play
Sarokoff, R. A., Taylor, B. A., & Poulson, C. L., (2001). Teaching in children with autism. Journal of Autism and Developmental
children with autism to engage in conversational exchanges: Disorders, 23, 467–489.
Script fading with embedded textual stimuli. Journal of Applied Yang, T., Wolfberg, P. J., Wu, S., & Hwu, P. (2003). Supporting
Behavior Analysis, 34, 81–84. children on the autism spectrum in peer play at home and school.
Sasso, G. M., Melloy, K. J., & Kavale, K. (1990). Generalization, Autism, 7, 437–453.
maintenance, and behavioral co-variation associated with social Zurcher, C., Hunt, P., Schuler, A., Webster, J. (2001). Increasing joint
skills training through structured learning. Behavioral Disorders, attention, play and language through peer-supported play.
16, 9–22. Autism, 5, 374–398.

123