You are on page 1of 19



Predicting Language Outcomes

for Children Learning Augmentative
and Alternative Communication:
Child and Environmental Factors
Nancy C. Brady,a Kathy Thiemann-Bourque,a
Kandace Fleming,a and Kris Matthewsa

Purpose: To investigate a model of language development for model indicated that the intrinsic child predictor construct
nonverbal preschool-age children learning to communicate predicted different words children produced. The amount of
with augmentative or alternative communication. input received at home, but not at school, was a significant
Method: Ninety-three preschool children with intellectual mediator.
disabilities were assessed at Time 1, and 82 of these children Conclusions: The hypothesized model accurately reflects a
were assessed 1 year later, at Time 2. The outcome variable latent construct of Intrinsic Symbolic Factor (ISF). Children
was the number of different words the children produced who evidenced higher initial levels of ISF and more adult
(with speech, sign, or speech-generating devices). Children’s input at home produced more words 1 year later. The
intrinsic predictor for language was modeled as a latent variable findings support the need to assess multiple child variables
consisting of cognitive development, comprehension, play, and suggest interventions directed to the indicators of ISF
and nonverbal communication complexity. Adult input at and input.
school and home, and amount of augmentative or alternative
communication instruction, were proposed mediators of
vocabulary acquisition. Key Words: children, augmentative and alternative
Results: A confirmatory factor analysis revealed that measures communication, assessment, outcomes, intellectual
converged as a coherent construct, and a structural equation disabilities

pproximately 51,046 children between the ages of about what contributes to successful outcomes for these
4 and 6 years in the United States are learning to children. Many children learn to use AAC for brief periods of
communicate with augmentative or alternative time and then transition into speech communication. Other
forms of communication (AAC), according to data from the children continue to rely on AAC as their primary form of
U.S. Department of Education (2011) and Binger and Light communication. Whether in speech or AAC, productive voca-
(2006). AAC is typically prescribed when children are strug- bulary acquisition varies considerably. The research described
gling with learning to communicate with speech or if they in this article was aimed at describing a set of variables that
are at risk for having delayed speech development because of predict the outcome of vocabulary acquisition.
a diagnosed condition such as Down syndrome or autism. We addressed three research questions. The first was
As previous misconceptions about AAC have been addressed, whether a set of variables identified a priori in accordance
it has become more accepted and common among very young with a model of early symbolic development would covary
children (Cress & Marvin, 2003). However, little is known and converge on a latent construct we refer to as Intrinsic
Symbolic Factor (ISF). Our second question was whether this
construct would significantly predict symbolic communica-
University of Kansas tion outcomes. Our third question was whether the effects
Correspondence to Nancy C. Brady: of this construct were mediated by social environmental
Editor: Janna Oetting variables for children in the beginning stages of learning to
Associate Editor: Katherine Hustad communicate. We addressed these questions by first evalu-
Received March 31, 2012 ating the psychometric properties of the construct and then,
Revision received September 4, 2012 using a structural equation model, investigating how well the
Accepted January 11, 2013 construct predicted vocabulary outcomes and whether the
DOI: 10.1044/1092-4388(2013/12-0102) effects were mediated by input and instruction.

Journal of Speech, Language, and Hearing Research • Vol. 56 • 1595–1612 • October 2013 • A American Speech-Language-Hearing Association 1595
Model of Early Symbolic Development (Bartholomew, Knott, & Moustaki, 2011). Analyzing the
relationships between latent constructs is superior to ana-
We began with an a priori model of multiple intrinsic
lyzing relationships between individual indicators because
child and environmental variables proposed to predict com-
measurement error is removed when the latent constructs are
munication outcomes for children learning AAC. The theore-
estimated. Consequently, the relationships between latent
tical basis of the proposed model most closely aligns with an
constructs provide better estimates of the true scores of each
emergentist view (Poll, 2011). According to the emergentist
construct and, therefore, more accurate estimation of the
theory of language learning, language is a product of the
causal relations over time. In addition, a multimethod ap-
interaction of intrinsic capabilities of the child and the outside
proach increases assessment validity by reducing reliance on
language environment. Although often applied to grammat-
a single measure (Bornstein & Haynes, 1998). A convergent
ical learning, emergentist theories have also been used to
approach such as latent construct analysis allows one to
explain early symbol development (Bates, Benigni, Brethern,
evaluate an individual child’s inferred predisposition for
Camaioni, & Volterra, 1979; Bates & Dick, 2002). According
learning that is not dependent on just one testing procedure.
to an emergentist view, both intrinsic child variables and
This approach is particularly helpful for children with in-
specific characteristics of social interaction, including lan-
tellectual disabilities (IDs), who may show uneven profiles
guage input, are critical for language learning. Our focus was
due to preferences or behavioral limitations. In the next
on early symbol use, which, according to this theory, emerges
section, we describe the components of ISF and introduce the
from presymbolic communication in social contexts (Bates
social interaction and instructional variables included in our
et al., 1979; Brady, Marquis, Fleming, & McLean, 2004;
model of communication development.
Iverson & Thal, 1998). However, not all children progress
in symbol learning at similar rates, and many are significantly
impaired in learning to communicate with symbols. We pro- Indicators of ISF
pose that a latent construct, ISF, reflects children’s intrinsic ISF is proposed as a latent construct with multiple
predisposition for learning to communicate with symbols. The indicators. The specific variables identified as potential
variables selected for this model include not only traditional indicators of ISF were drawn from past research on
predictors of early language growth, such as receptive lan- predictors of early communication development in both
guage, but also measures that do not rely on speech, such as typically developing children and children with IDs
play. Non–speech-related indicators were selected because we (e.g., Maatta, Laakso, Tolvanen, Ahonen, & Aro, 2012;
were modeling early symbol development that is not exclu- McCathren, Warren, & Yoder, 1996; Watt, Wetherby, &
sively reliant on speech. Each variable in the model is described Shumway, 2006). Important for the current investigation is
below, along with a rationale for its selection. that we wanted to include indicators that did not rely on
In addition to demonstrating the importance of each speech production. In essence, our aim was to measure
individual variable for predicting language outcomes, we the intrinsic resources a child brings to learning symbolic
investigated how well a group of variables converged on a relationships between referents and symbols such as visual
single latent construct: ISF. This construct is theoretical but images, signs, or spoken words. We selected indicators from
indicated by a set of real measures. To address our first the domains of nonverbal cognitive development, language
research question, we investigated the contributions of dif- comprehension, complexity of presymbolic communication
ferent variables of child language predictors using a latent (e.g., gesture use), and play. Assessment of children with
construct model. In a latent construct model, a complex IDs is often difficult, and standardized tests alone seldom
construct is measured through multiple manifest indicators provide sufficient information. Therefore, we used both
(Graham, Collins, Wugalter, Chung, & Hansen, 1991). For standardized and nonstandardized measures where appro-
example, Bornstein and Haynes (1998) investigated the ex- priate. Further discussion of the rationale for each indicator
tent to which various assessments, observations of child and its measure is provided in the following paragraphs.
speech, and maternal reports all converged on a construct Nonverbal cognitive development. The term cognitive
they referred to as vocabulary competence. This construct was development refers to development of skills or abilities that
then investigated as a predictor of vocabulary outcome contribute to a child’s aptitude for learning in general. Mea-
measured 2 years later. In a study of children with autism, sures such as IQ or other age-referenced composite scores are
Harrison and Smith (2012) used a latent modeling approach derived by averaging results from subtests such as visual
to measure the construct autism severity, indicated by com- perception or processing, auditory perception or processing,
munication, social, and stereotyped behavior and restricted- attention, memory, sequencing, fine motor skills, and lan-
interests scores from two different autism measures. The guage. For the present study, we focused on nonverbal mea-
authors then determined how the construct of autism severity sures of cognitive development for three reasons: (a) All
predicted child well-being. participating children were identified as having severely lim-
There are several advantages to a latent variable ap- ited verbal productions, (b) verbal language was one of our
proach to modeling outcomes. A latent variable approach outcome measures, and (c) nonverbal cognitive levels have
links observable data to theoretical constructs of interest. In been found to predict later language outcomes in children with
terms of measurement theory, condensing multiple measures IDs and in children with autism (Brady, Steeples, & Fleming,
into an aggregate reduces the dimensionality of the data 2005; Thurm, Lord, Lee, & Newschaffer, 2007). The Visual

1596 Journal of Speech, Language, and Hearing Research • Vol. 56 • 1595–1612 • October 2013
Reception subtest from the Mullen Scales of Early Learning research team called the Communication Complexity Scale
(MSEL; Mullen, 1995) is particularly relevant to symbol (CCS). We use the term communication complexity to refer to
learning for children who are candidates for AAC because the development of communication that typically occurs
most AAC systems rely on visual reception and discrimination. before linguistic communication. Prior to speech, children
The Fine Motor subtest provides a measure of visual–motor communicate with a variety of gestures and vocalizations, and
planning and control that is also relevant to symbol use. These these presymbolic communication acts are thought to pave the
subtests from the MSEL have been used in past studies to way for later symbolic communication (Butterworth, 2003;
measure nonverbal cognitive development in children with Tomasello, 2003). Research from typically developing children
IDs (Philofsky, Hepburn, Hayes, Hagerman, & Rogers, 2004; (Carpenter, Nagell, & Tomasello, 1998; Crais, Day Douglas,
Wetherby, Watt, Morgan, & Shumway, 2007). & Cox Campbell, 2004), as well as from children with dis-
Language comprehension. Language comprehension is abilities (Brady et al., 2004, 2011), suggests that the progression
often cited as an early predictor of later word production in in prelinguistic communication is related to the onset of
typically developing children (Bates, O’Connell, & Shore, symbolic communication. For example, children who point
1987) and in children with disabilities (Chapman, Seung, earlier tend to also speak earlier (Brooks & Meltzoff, 2008).
Schwartz, & Bird, 2000; Sevcik & Romski, 1997). Although Advancements in forms (specifically, gesture types),
many measures of language comprehension in children less functions, and rates of prelinguistic communication can be
than 2 years of age rely on parent report, findings by Watt observed in changes in the use of communicative gestures
et al. (2006) indicated that direct behavioral measures of and vocalizations and in the coordinated use of these behav-
comprehension can also be robust predictors of later expres- iors with communicative partners (Adamson & Chance,
sive language. Early language comprehension abilities indi- 1998; Legerstee & Fisher, 2008). The CCS was developed for
cate an understanding that things have names, and this and used in this study to measure these prelinguistic com-
insight appears critical to symbolic understanding. munication behaviors.
Comprehension of both verbal and nonverbal aspects Play levels. Play has been related to language develop-
of language is a core deficit for children with autism (Prizant, ment both in typical populations (Eisert & Lamorey, 1996;
Wetherby, & Rydell, 2000) and Down syndrome (Chapman, McCune-Nicolich & Carroll, 1981) and in children with
2003). The importance of comprehension for predicting disabilities (Kasari, Freeman, & Paparella, 2001; Landa,
expressive language development appears to vary by the Holman, & Garrett-Mayer, 2007). Children’s play may
child’s stage of development. For example, Vandereet and provide insight into underlying symbolic understanding that
colleagues (Vandereet, Maes, Lembrechts, & Zink, 2010) might not be obvious through other cognitive and language
found that vocabulary comprehension was the only unique assessments. In typical development, symbolic or pretend
predictor of initial expressive vocabulary for a group of play and language occur at around the same time. Researchers
36 children with IDs. However, children in Vandereet et al.’s have argued that this co-occurrence is because both
study had expressive vocabulary averages of 67 words at symbolic play and language are manifestations of the same
the start of their study. Research has demonstrated that, underlying representational ability (Piaget, 1962; Werner &
for young prelinguistic children (expressive vocabularies Kaplan, 1984). McCathren et al. (1998) reported that chil-
<20 words), receptive vocabulary is not exclusively predic- dren’s level of representational play was significantly related to
tive of expressive vocabulary in typically developing chil- later expressive language. Similarly, Laakso et al. (1999) found
dren (Laakso, Poikkeus, Eklund, & Lyytinen, 1999) or in that children’s scores on the Symbolic Play Test (Lowe &
children with developmental disabilities (McCathren, Yoder, Costello, 1976) at age 14 months significantly predicted
& Warren, 1998). language comprehension at age 18 months in typically devel-
Despite conflicting findings in past research, it is oping children. Similar findings have been reported for children
worthwhile to continue investigating the role of language with Down syndrome (Cunningham, Glenn, Wilkinson, &
comprehension, in particular for children learning AAC. Sloper, 1985) and children with autism (Stone, Lemanek,
Sevcik and Romski (2002) described the different learning Fishel, Fernandez, & Altemeier, 1990; Thiemann-Bourque,
processes for individuals learning to use AAC who have Brady, & Fleming, 2012). On the basis of research indicating
good speech comprehension skills versus those who have the importance of play as an index of child communication
poor comprehension. Their research centered on children development, we adapted a play measure from research with
learning to use symbols paired with speech-generating children with IDs to yield the play indicator for our construct.
devices (SGDs; Romski et al., 2010). Communication via an This play measure is described in the Method section.
SGD involves selection of a symbol that produces a synthe-
sized or digitized spoken utterance. Individuals who are Social Interaction Variables
skilled in comprehending speech may use their knowledge Consistent with emergentist theories of language
about the relationship between words and the referents in acquisition, we hypothesized that certain elements of the
their environment to help learn AAC selection responses. social interactions that surround communication contribute
Communication complexity. We also wanted a measure directly to child outcomes and significantly mediate con-
to reflect how children communicate with gestures, vocali- tributions by the child’s intrinsic abilities. Specifically,
zations, words, and symbols in social communicative con- communication input by adults is proposed to significantly
texts, and for this we relied on a measure developed by our affect communication development (MacWhinney, 2004).

Brady et al.: Language Predictors in Children Learning AAC 1597

Communication input. The amount of communication input to children with more severe language impairments
directed to a child in everyday interactions directly relates to compared to children with higher language ability.
the child’s communication and language attainments (Hart Amount of instruction. In addition to interventions
& Risley, 1995; Landry, Smith, Swank, & Guttentag, 2008). aimed at increasing partner input and facilitation, direct
One of the key themes of emergentist views of language is AAC intervention with a child is a likely source of variability
that language learning depends heavily on the amount and in children’s language outcomes. For example, based on
quality of language input (Poll, 2011). Emergentist research lengthy videotape analyses of teachers in classrooms, Connor
on grammar suggests that both the overall volume of lan- and colleagues found that the amounts of particular instruc-
guage input and the exposure to particular grammatical tional activities was associated with literacy outcomes in
forms relate to improved learning outcomes. Input is rec- preschool children (Connor, Morrison, & Slominski, 2006).
ognized as an important variable for vocabulary learning as Although it may seem intuitive to ask how the amount of
well. Children who experience more language input during instruction within AAC instructional activities leads to
their early years have significantly better vocabulary out- positive changes in communication, the amount of AAC
comes than children with less input (Huttenlocher, Haight, instruction as a variable or outcome measure has yet to be
Bryk, Seltzer, & Lyons, 1991). Hoff (2003) reported that documented. Given the significant variability of instructional
differences in maternal mean length of utterance and word contexts, types of systems being used, and treatment dosages
types and tokens spoken to their 2-year-old children were inherent across different preschool classrooms and popula-
associated with differences in the children’s number of dif- tions, defining and clearly measuring this variable is difficult.
ferent words produced. Children’s language comprehension For example, most preschool programs embed teaching
is also improved through greater input; hence, input may opportunities throughout the day, which is beneficial for the
facilitate word learning by increasing comprehension for a children but creates difficulty in measuring instruction time.
referent word that is being represented by one or more dif- An estimate of amount of AAC instruction time based on
ferent symbol types (Brady, 2000; Romski & Sevcik, 1993). teacher report is one method to determine whether this
Researchers have explored input involving different variable has a potential effect on language outcome.
communication modes for children with IDs. Many children To summarize, in the current study, we first determined
with IDs use modes other than speech to communicate. Input whether the set of variables proposed as indicators of ISF
to these children is typically speech, although some scholars (nonverbal cognition, language comprehension, communi-
advocate using other modes in addition to speech. For cation complexity, and play) did, in fact, significantly con-
example, in a review of 33 studies involving a total of 216 tribute to this latent construct. Second, we determined whether
children with various disabilities, Dunst and colleagues this construct predicted child language outcomes 1 year later,
(Dunst, Meter, & Hamby, 2011) reported that simultaneous and third, we evaluated whether language input and instruc-
input that combined speech with sign language had positive tion time interacted with ISF to affect language outcome.
effects on child oral language output (the average effect
size was 0.72 for group studies and 1.48 for single-subject
studies). The review focused on intervention studies, how- Method
ever, and did not examine the effect of different amounts of The research reported in this study was approved by
input. For children learning to use graphic forms of AAC, the human subjects committee of the internal review board of
including SGDs, various interventions have been developed the University of Kansas.
that rely on providing augmented input (e.g., Harris &
Reichle, 2004; Romski et al., 2010). Under these approaches,
communicative partners are taught to speak while touching Participants
symbols on the child’s communication device. For example, Ninety-three children participated in this study.
the partner might say, “It’s sunny outside” while pointing Seventy-three of the children were boys. Two observation
to the symbol for “sunny.” Strategies that provide aug- points, 1 year apart, were measured in this study. Eighty-
mented input (either sign or symbol) are designed to emulate two of the children were observed at the second data point.
natural language learning opportunities and capitalize on the The mean chronological age at the first observation was
learner’s abilities to associate speech, referent, and symbol. 50.3 months. All of the children had composite scores on the
Although most studies of input effects have measured MSEL that were at least 2 SDs below the norm for their age.
input by parents, a few studies have also considered input The range in standard scores was 48 through 58, based on
from teachers and other school personnel. Brady, Herynk, a normative mean of 100. Many children had a specific
and Fleming (2010) reported that input at preschools was diagnosis, such as an autism spectrum disorder (n = 44) or
relatively sparse, but there was a wide range of variability in Down syndrome (n = 15). The remaining children either
teacher initiations to children with IDs (mean initiation rate had a relatively rare disorder, such as Opitz trigonocephaly,
of 1.62 per minute, SD = 0.56). Children who experienced or did not have a diagnosis specifically associated with their
more input from teachers also communicated more during intellectual and language disability. Descriptive information
these observations. Similar findings were reported by for the participants, including age equivalent test score
Rowland (1990) in classrooms serving children with multiple information, is presented in Table 1. The information is
disabilities. Hadley and Rice (1991) reported less teacher broken down according to prominent diagnostic groups

1598 Journal of Speech, Language, and Hearing Research • Vol. 56 • 1595–1612 • October 2013
Table 1. Descriptive statistics of participant demographics: Medians (and ranges) of cognitive and language assessments
at Time 1 for participants with autism spectrum disorders (ASDs), Down syndrome, and all other diagnoses.

ASDs Down syndrome Other

Variable (n = 45) (n = 15) (n = 33)

Age in months (range) 50 (37–71) 52 (36–68) 46 (37–66)

Percentage male 87% 60% 76%
MSEL age equivalents
Receptive Language 13 (2–34) 23 (11–34) 18 (5–34)
Expressive Language 14 (5–24) 16 (10–21) 12 (0–26)
Visual Reception 20 (1–29) 23 (8–31) 19 (0–37)
Fine motor 21 (11–33) 22 (10–34) 18 (2–34)
PLS–4 age equivalents
Auditory Comprehension 15 (10–28) 21 (10–33) 18 (7–45)
Expressive Communication 20 (11–27) 20 (13–27) 17 (7–25)
Total age equivalent 17 (10–27) 18 (16–30) 18 (7–30)

Note. MSEL = Mullen Scales of Early Learning; PLS–4 = Preschool Language Scale—4.

(autism spectrum disorder, Down syndrome, other), purely (without prompting), intentionally (directed to another person),
for descriptive purposes. These groups were not compared or and intelligibly. If there were inconsistent reports between
contrasted in our analyses. teachers and parents for individual children, we moved
Children were recruited by contacting school districts forward with the first visits and planned to screen the child
in and near the Kansas City metropolitan area, specifically, on the basis of further discussions and observations of the child
in Topeka, Kansas, and Wichita, Kansas. Teachers and at school and home. Any children whom we observed to pro-
speech-language pathologists were asked to nominate any duce more than 20 different words during our initial observa-
children meeting the criteria outlined below. Once the tions at school and home were not included in the study.
children had been nominated, their parents were contacted to The types of AAC used by participants at the first
gain informed consent and complete the screening and as- observation, according to teacher report, are reported in
sessment process. At intake, each participant met the fol- Table 2. These types of AAC were selected by teachers from
lowing eight criteria, verified through either teacher report a list of options, and teachers were asked to record all types
or direct observation: (a) chronological age between 3 and of AAC in use for a given child. Many children were being
5 years; (b) enrollment in a preschool program; (c) vision taught to use multiple forms (e.g., sign plus a graphic form
reported as 20/80 or better in at least one eye (with or without of AAC); however, teachers also frequently reported that
correction); (d) hearing reported as 25 dB HL or better in they were “trying out different systems” for a particular
at least one ear (with or without amplification); (e) upper child, which is common, particularly for this age group.
body motor skills sufficient to directly select symbols with Many of the children added or changed systems over the
fingers, hands, or arms; (f ) English as the primary language course of this study (hence, analyses related to specific types
spoken at home; (g) current teaching plans that included of AAC were not possible).
AAC (graphic symbols, sign language, and/or an SGD); To estimate socioeconomic status, we asked families
and (h) vocabularies of fewer than 20 different words said, to indicate the highest level of maternal education. Seventy-
signed, or selected. For this last criterion, parents and eight families responded to this question; of these, 15 had
teachers were asked to list words produced spontaneously graduate degrees, 22 had 4-year college degrees, 17 had

Table 2. Percentages (and numbers) of participants reported to use a particular type of augmentative and alternative
communication (AAC) at first observation.

ASDs Down syndrome Other

AAC system(s) used (n = 45) (n = 15) (n = 33)

PECS only 56% (25) 27% (4) 21% (7)

Sign only 0 13% (2) 6% (2)
SGD only 2% (1) 0 30% (10)
PECS and sign 18% (8) 40% (6) 24% (8)
PECS and SGD 20% (9) 7% (1) 3% (1)
Sign and SGD 4% (2) 7% (1) 6% (2)
PECS, sign, and SGD 0 7% (1) 9% (3)

Note. PECS = Picture Exchange Communication System; SGD = speech-generating device.

Brady et al.: Language Predictors in Children Learning AAC 1599

2 years of college; 22 had a high school education or equiv- activities with an RA as the examiner. Five different sets of
alent, and two had less than a high school education. The toys were used. The examiner modeled one symbolic play
mean income level was between $50,000 and $59,000, with a act, and then the child played with each set of toys for 3 min.
range of <$9,999 to $100,000+. In terms of ethnic diversity, Examiners were trained to only imitate the child’s play
60 children were White, 16 were African American (not actions, to not introduce any new play actions on the toys,
Hispanic), six were Hispanic, three were Asian, four were and to limit use of language-facilitating strategies. Play skills
another race (“Other”), and four were multiracial. were coded using an online play coder program (adapted
from Tapp & Yoder, 2000) that contained 11 of 15 categories
Procedure from the Developmental Play Assessment instrument (Lifter,
Sulzer-Azaroff, Anderson, & Cowdery, 1993). A complete
Once children were recruited and informed consent definition for each play behavior coded is provided in
had been obtained, the research staff contacted the parents Thiemann-Bourque et al. (2012). To summarize, the 11 cate-
and school staff to schedule the following three assessments, gories comprised four play levels: Level 1—Indiscriminate
which were completed at home: (a) the MSEL, (b) the Pre- Actions, Level 2—Functional Play/Object Use, Level 3—
school Language Scale—4 (PLS–4; Zimmerman, Steiner, Functional Combinatorial, and Level 4—Symbolic Play. For
& Evatt Pond, 2003), and (c) a structured play assessment. the purposes of this article, play analyses did not include
Each of these assessments is described below. During the indiscriminate actions on objects (i.e., mouthing, shaking,
home visit, research staff also collected demographic and banging, and inspecting), because these acts were considered
background history information. Once a child was enrolled undifferentiated play. Each child received a weighted score
in the study, additional observations of the child communi- by assigning a score of 1, 2, or 3 to each play act: one point for
cating in his or her classroom and home environments were any Level 2 act, two points for any Level 3 act, and three
made at the time of enrollment and again approximately points for any Level 4 act. A weighted play score was calcu-
1 year later. To be exact, the mean time between the first and lated for each child by totaling these points across the five
second observations was 13.4 months, and the range was 3-min play sets. The means and standard deviations for the
10.4 to 17.6 months. Adult input was obtained from video- weighted play scores are presented in Table 3.
taped observations collected at home and school at Time 1. Two RAs independently coded 33% randomly se-
Children’s number of different words produced—our out- lected play assessments, and intraclass correlation coeffi-
come variable—was obtained in two different settings at cients (ICCs) were calculated for reliability. The ICC for the
Time 2: (a) home observation and (b) a scripted communi- overall weighted play score was quite high: .99. The ICCs for
cation sample. Each of these contexts is described below. the individual play types that were combined to form the
weighted play score also were high, ranging from .87 for
Time 1 Predictor Measures
Takes Apart Combinations to .99 for Child as Agent and
Procedures for obtaining the measures that were pro-
Specific Combinations.
posed as indicators of ISF are described in the following
Scripted communication sample: CCS. A scripted set
of activities was provided to each child in order to obtain a
Standardized assessments. Two standardized assess-
sample of communication that would be in a similar context
ments were administered at Time 1. Research assistants (RAs)
across all participants. The scripted communication sample
were taught to administer these tests to standard criteria,
was administered by one of our trained RAs in a quiet
and periodic fidelity rechecks were performed by the second
location at the child’s school. A series of 12 age-appropriate
author to ensure accurate administration. The MSEL is an
play opportunities—six to request and six to comment—
assessment of early learning designed for children from birth to
were presented by a trained RA to each child. Two examples
6 years of age. It consists of multiple scales, including Recep-
tive and Expressive Language. We used the Visual Reception
and Fine Motor subscales from the MSEL as the measure
of nonverbal learning. We computed the average of the raw Table 3. Means and standard deviations of variables used in the
confirmatory factor analysis (first four variables) and in the structural
scores from these subscales to yield an estimate of nonverbal equation model.
cognitive skills.
The Auditory Comprehension subtest of the PLS–4 Variable M SD
was administered to obtain a measure of receptive language.
In terms of early language testing, the PLS–4 has some ad- Communication complexity 8.85 1.41
vantages compared to the MSEL in that it includes more items PLS–4 Language Comprehension 22.91 5.47
that test earlier developmental skills. We used raw scores for Weighted total play 33.09 20.10
MSEL nonverbal composite raw 41.70 12.04
the PLS–4 in our analyses because most of our participants Rate of adult input at home, time 1a 4.09 1.49
scored at the lowest level possible on standard scores, leading Amount of AAC instructionb 3.13 1.07
to low variability across participants. The means and ranges Rate of different words at home, Time 2 1.30 1.40
of scores from these tests are presented in Table 1. Rate of different words (scripted assessment at Time 2) 0.71 1.08
Play assessment. Children’s play was assessed for a a
Rates are number per minute. bBased on a 1–5 scale.
total of 15 min during unstructured and semistructured

1600 Journal of Speech, Language, and Hearing Research • Vol. 56 • 1595–1612 • October 2013
are playing with a wind-up toy (one broken, one working, to Classroom observations. Each participant was ob-
elicit a request) and looking through books with disfigured served in his or her classroom over two 1-hr visits scheduled
pages (to elicit a comment). These opportunities were on different days. Our goal was to observe a total of 120 min
similar to what have been referred to in other assessments for each child over two different days at Time 1 and again
as temptations (Eadie et al., 2010; Wetherby & Prizant, 2003) 12 months later at Time 2. The mean number of minutes
in that opportunities to play with an item are provided and of actual observation at Time 1 was 127 min, with a range
the experimenter takes turns interacting with the child and of 108 through 155 min, and at Time 2 the mean duration was
the materials. Next, the experimenter waits for the child to 126 min, with a range of 60 through 180 min. The first and
request continuation of the activity (in behavior regulation second hours of observation were intentionally scheduled
tasks) or introduces a surprising element (joint attention) and during different periods of the preschooler’s day. This ensured
waits for the child to respond. Similar strategies have been the broadest sample of activities and interactions across the
used by numerous research teams and interventionists to day. Our goal was to capture a typical day in each child’s
assess intentional communication acts (e.g., Brady et al., preschool environment. The activities during the observed time
2005; Kasari, Paparella, Freeman, & Jahromi, 2008). All were not planned or adapted to our visit in any way. Thus, it
samples were videotaped and coded at a later time using the can be said that the sample of communicative behavior was
Noldus Observer XT software (Noldus Information Tech- random and represents a broad range of activities. Several
nology, 2008; see description below). different activities were observed over the course of an ob-
The scripted assessment was designed to assess pre- servation period for one child (e.g., small group, snack, free
symbolic and early symbolic communication. All tasks could play, and recess); however, because each classroom was dif-
be responded to with gestures and vocalizations, but in ferent, the activities observed also differed across school sites.
order to also allow for children to use symbols, we provided The average class size for each child was eight children and
graphic symbols representing one toy in each activity along three staff.
with general symbols for “help” and “look.” The RA dem- On a typical school visit, one or two trained research
onstrated the availability of the toy symbol by holding the assistants entered the child’s classroom to code adult–child
symbol next to the referent toy and saying “This (reference to communication interactions with hand-held personal digital
the symbol) means this (reference to the toy)” prior to each assistants. Each personal digital assistant was equipped
activity. Research staff also encouraged the use of existing with the software package Pocket Observer by Noldus. Adult
AAC systems. For example, children who used a Picture and child communication behaviors were recorded by tapping
Exchange Communication System brought their personal the screen with a stylus. Appendix B lists the communicative
symbol array to the testing. Children who were learning to use behaviors recorded. The observers positioned themselves
an SGD had the device available during the assessment; within an audible and visible distance from the child but re-
however, advanced programming of specific vocabulary was mained unobtrusive. Coding was paused for bathroom breaks,
not possible because of time constraints. The RAs were or if the communication was inaudible because of the class-
familiar with frequently used signs and responded appro- room noise level or invisible because of positioning. Also,
priately if a child produced a recognized sign during the coding was paused or stopped if the activity prevented obser-
scripted communication assessment. vation of typical communication opportunities, such as play-
The scripted communication sample was the context for ing a movie, nap time, illness, or extended disruptive behavior.
scoring the CCS. The CCS is an 11-point scale developed by For our analyses, we used the number of adult com-
the research team to describe a participant’s level of prelin- munication acts recorded during this classroom observation
guistic and symbolic communication development (Brady as an indicator of communication input. Adult communi-
et al., 2011). All potentially communicative behaviors are cation was coded whenever an educator directed communi-
recorded from the videotapes, and the potentially communi- cation to the target child. We excluded adult communication
cative behaviors that occur in response to the scripted op- directed to a group of children unless it could be determined
portunity are assigned a score based on the 11-point CCS. by the educator’s speech or gestures that the communication
Increasing scores reflect more complex prelinguistic and early was specifically intended for the focus child. Frequencies
linguistic communication. The entire CCS is presented in were converted to rates per min for total communication acts
Appendix A. To score the CCS, an RA viewed participants’ to account for differences in times. Although the types of
responses to each of the 12 communication opportunities adult communication (speech, sign, symbol, etc.) were re-
and determined the highest level response for that opportunity. corded, for the purpose of our analyses all types were com-
We treated the scale as a continuous, ordinal measure and bined into a total adult input variable. Child communication
computed mean scores, standard deviations, and correlations acts (e.g., initiations, responses) were also recorded during
from scores we obtained (Iacobucci, 2012). The average of the the classroom observations; however, they were not included
three opportunities (temptations) with highest scores was used in the current analyses.
as each participant’s CCS score in our analyses. Instruction time. Although comprehensive measures
of the amount of instruction children received were not
Social Interaction Variables obtained for this study, we were able to estimate variability
Procedures used to obtain data for social interaction in the amount of AAC instruction. In an effort to describe
variables are described in the following sections. variations in AAC instruction time, the RA asked the

Brady et al.: Language Predictors in Children Learning AAC 1601

primary educator the following question during the Time 1 scripted communication contexts because we videotaped
visit: “During a typical school week, how often does this and transcribed in these contexts. It was not possible to video-
child receive direct AAC intervention?” The RA filled in the tape during classroom observations because of privacy con-
teacher report on a form categorizing hours of instruction per cerns, and we were not confident in our ability to reliably
week (i.e., 1 = never; 2 = less than 30 min/week; 3 = 30 min/ transcribe words during live observations. The entire scripted
week or more, but less than 60 min/week; 4 = 60 min/week communication sample and approximately 1 hr of parent–
or more, but less than 90 min/week; or 5 = 90 min/week or child observation was videorecorded. The middle 20 min of
more). We explained that they should consider any AAC the home observations were later coded and transcribed.
instruction provided throughout the day, not just instruction These middle 20 min were selected because this allowed the
that was given during specified therapy sessions. However, caregiver and child the opportunity to get familiar with the
we asked them to exclude group activities, such as circle time, research staff and camera in the home (i.e., a warm-up period),
that could include some AAC but that was not directed and coding was finished before the family became fatigued.
specifically to the target child. AAC instruction survey data Transcription. All child utterances containing words
was collected for 93 (or 100%) of the participants at Time 1. said or signed or produced with an SGD were transcribed
The mean teacher report score of 3.13 was closest to the during the video-coding process using Noldus software.
category of 30 min/week or more but less than 60 min/week. These transcribed utterances were later analyzed using the
Home observations. Each participant was also ob- Systematic Analysis of Language Transcripts (SALT; Miller
served interacting with an identified primary caregiver in his & Iglesias, 2008). The number of different (unique) words
or her home for 1 hr at Time 1 and at Time 2. For 83 of the was derived from the SALT analyses of these transcripts. All
families, the primary caregiver was the mother. For five utterances were considered, including those with unintelli-
families, the identified caregiver was the father, and for an gible words, as long as at least one word could be transcribed
additional five families the primary caregiver was a grand- (see below). The number of different words was converted to
parent. The identified caregiver remained consistent for both rates of different words per minute before completing the
Times 1 and 2. The home observations were videotaped by structural equation analyses (described below).
research staff and coded at a later time using the Noldus Transcribers (trained RAs) were instructed to listen to
Observer XT software. an utterance no more than three times before transcribing
Our goal (which was also expressed to parents) was to it. If the transcriber could not identify any specific words
measure parent–child interactions during typical activities after watching the video segment three times, the utterance
at home, such as meal- or snack time, play time, or regular was coded as a vocalization rather than a verbalization. If
routines (e.g., getting dressed for school or being read to an utterance contained both intelligible and unintelligible
at night if preparing for bed). Whenever possible, RAs words, intelligible words were transcribed, and an X was
scheduled home observations when the primary caregiver used to denote an unintelligible word. For example, an
would be alone with his or her child. Parents were instructed utterance may have been transcribed as “I want X,” with
to stay in the same room with his or her child and to leave the X denoting the unintelligible word.
television off or the volume turned down. The RAs provided Spoken words were transcribed if they met the fol-
a bin of toys during each visit and informed parents that lowing four criteria: (a) the utterance contained one vowel;
playing with them was optional. Parents were encouraged to (b) the utterance contained one or more consonants match-
go about their typical daily routines (outside of watching ing the consonant placement in the target word, or a devel-
television). These were the only instructions provided, and opmentally appropriate substitute (e.g., if the child said
parents were free to engage in activities that may not be “tum” for thumb, this would be transcribed as “thumb”);
conducive to interaction, such as video games, if they chose. (c) close approximations of words needed nonlinguistic
Anecdotally, our RAs reported that solitary activities such as support, such as child attention to the referent object, person,
video games rarely occurred during our home observations. or event or repeated use across multiple utterances; and
Similar strategies have been used to record vocabulary input (d) the word was not a direct imitation of the adult’s imme-
and output in natural contexts (Hart & Risley, 1995; diately preceding word. If a child repeated part of an adult
Huttenlocher et al., 1991). For our study, the amount of phrase, additional evidence of communicative intent was
caregiver input to children was measured, using the required. For example, if the adult said “Do you want
definitions and conventions described above. more track?” and the child said “More track” while reaching
for the track, this utterance was transcribed and coded as
Child Outcome Variables two words.
The primary outcomes for the current analyses were Similarly, signs were coded if they were acceptable
the number of different words (said, signed, or selected on approximations of American Sign Language signs. Tran-
a device) produced by children at Time 2. Because of our scribers were allowed to look up signs in an American Sign
criteria for participants entering the study, there was little Language dictionary, if necessary, to determine whether
variability in Time 1 word production (mean rate of different the child’s production matched at least two of the following
words at Time 1 was 0.20/min, SD = 0.32/min); hence, four parameters: (a) hand shape, (b) movement, (c) location,
only differences at Time 2 were analyzed. We were able to and (d) orientation. If a child both spoke and signed the
obtain the number of different words within the home and same word simultaneously, the word would be transcribed

1602 Journal of Speech, Language, and Hearing Research • Vol. 56 • 1595–1612 • October 2013
once. Words that were both said and signed were counted as Before conducting the analysis, all frequency data were
only one unique word for determining the number of dif- converted into rates per minute in order to account for slight
ferent words. variations in observation times. Rate was computed as the
SGD productions were recorded when the child frequency divided by the duration of each observation.
independently selected a symbol on the device, resulting in
audible speech from the device. A symbol selection was
coded instead only if the child’s selection did not have Results
sufficient pressure to produce speech. If selecting one symbol Analysis Plan
produced a phrase, the phrase was transcribed together, and Our research questions about the predictive relation-
only one word was counted in SALT. For example, if se- ships between child and environmental variables and child
lecting a single symbol produced the phrase “I want more,” language outcomes required three steps. The first step was to
this was transcribed as “Iwantmore,” and only one word was complete a confirmatory factor analysis (CFA) to determine
counted. However, if a child selected separate symbols for whether we had selected indicators of ISF that fit the data
“I,” “want,” and “more,” three words were counted. As with well. Next, we determined whether ISF predicted the number
sign, if the child spoke and selected a symbol for the same of different words children produced 1 year later. Finally, we
referent, only one unique word was counted for determining determined whether the amount of communication input
the number of different words. provided by adult communication partners at home and at
school, and variations in the amount of AAC instruction,
Reliability were significant mediators.
We measured reliability for variables measured in both
live and video coding. For live coding, 80 randomly selected
observations (45%) were completed by two trained observers CFA
who simultaneously recorded the communication observed Recall that one of our research questions was whether
in classrooms. The two observers stood away from each the set of hypothesized child variables were indicators of a
other and independently coded behaviors. Comparisons latent construct, ISF. To address this question, we conducted
across the independent coding were completed with ICCs. a CFA using Mplus version 6.11 (Muthén & Muthén, 1998–
The ICC for adult communication acts was .99. 2011). Before conducting the CFA, the univariate distribu-
Reliability for video recording was measured by hav- tions of the indicator variables were checked for normalcy
ing two coders independently code 90 randomly selected and outliers. None of the variables required transformation.
videos (37%). After a file was coded by both independent Both standard maximum likelihood estimation and robust
coders, the results were compared to each other, first using maximum likelihood estimation was used given the limited
percentage agreement and then later with ICCs. If percent- sample size. The means and standard deviations for the four
age agreement fell below 75%, a consensus coding procedure indicators of ISF are shown in Table 3, and the correlations
was followed. The two coders conferred and came to a con- among all of the child variables and the other measures
sensus regarding all disagreements. The consensus file was included in the model, with the indicators of ISF listed first,
then used to extract the data for the communication variables. are presented in Table 4. Correlations among the ISF indi-
ICCs among primary, reliability, and consensus files ranged cators ranged from .40 to .75, with MSEL nonverbal scores
between .97 and .99. Reliability for the number of different and play scores being most strongly related. The a priori
child words was computed by having an additional coder model was that all child reflective indicators would load
transcribe 41 randomly selected files (47% files). The ICC without covarying errors on a single factor. Goodness of
between raters for the number of different words was .99. fit was evaluated using the root-mean-square error of

Table 4. Summary of intercorrelations for variables used in the confirmatory factor analysis (first four variables) and in the structural equation

Variable 1 2 3 4 5 6 7 8 9

1. Communication complexity —
2. PLS–4 Auditory Comprehension .40** —
3. Weighted total play .45** .50** —
4. MSEL nonverbal composite raw .46** .59** .75** —
5. Rate of adult input at home, Time 1a .32** .40** .34** .33** —
6. Rate of adult input at school, Time 1 .25* .07 –.06 .10 .06 —
7. Rate of different words at home, Time 2 .29** .43** .49** .34** .48** .16 —
8. Rate of different words scripted, Time 2 .38** .35** .36** .23* .49** .07 .80** —
9. Amount of AAC instructionb .02 .11 –.06 .06 –.12 .07 –.27* –.23* —
a b
Rates are number per minute. Based on a 1–5 scale.
*p < .05. **p < .01.

Brady et al.: Language Predictors in Children Learning AAC 1603

approximation (RMSEA), standardized root-mean-square initial observations. Parents at home had an average input
residual (SRMR), and the comparative fit index (CFI). The rate of 4.09 communication acts per minute (SD = 1.49),
RMSEA is a measure of lack of fit, whereby values below whereas teachers at school averaged 2.21 communication
.08 indicate acceptable fit and those below .05 indicate ex- acts per minute (SD = 0.82).
cellent fit. CFI values above .90 or .95 indicate acceptable Before evaluating the structural model, we checked the
and excellent fit, respectively. The SRMR is an absolute univariate distributions of each child communication vari-
measure of fit, and values less than .08 are considered to able for normalcy and outliers and examined the correlation
represent good fit. matrix (see Table 4). Correlations between the ISF indicator
The model fit the data well using the maximum likelihood variables and the child communication outcome variables
estimator (RMSEA = .056, CFI = .996, SRMR = .026). No and rate of different words produced at home and at school
modifications to the model were recommended. All of the child were all significant, ranging from .26 to .48. Play scores and
indicators loaded on the latent variable (ISF), and the loadings rate of different words used at home were the most highly
were significant at the .001 level. Loadings ranged in magnitude correlated. The correlations between the adult input variables
from .59 for CCS to .94 for MSEL nonverbal composite raw and child language outcomes were not as expected: Whereas
score. PLS–4 Auditory Comprehension had a loading of .67, correlations between adult input at home and child outcomes
and weighted total play had a loading of .76. We ran the same were moderate and significant, ranging from .31 to .39, cor-
model again using the robust maximum likelihood estimator relations between adult input at school and most of the other
given the relatively small sample size. The data fit the model variables in the model were near zero, with the exception of
even better using the robust estimator (RMSEA = .001, CFI = CCS score (r = .25, p = .02). Given the pattern of correlations,
1.00, SRMR = .026), with no change in the parameter estimates. we dropped adult input at school from the model; input at
home was the only adult input variable used in the models
described below.
Structural Equation Models To determine whether ISF at Time 1 predicted child
Model 1. A structural equation model was developed communication at Time 2, we regressed child communica-
to explore the relationships among ISF at Time 1, adult input tion at Time 2 onto the latent variable for ISF at Time 1 and
at Time 1, and child communication at Time 2, as shown adult input at home at Time 1. This allowed analyses of the
in Figure 1. The model hypothesized that ISF at Time 1 direct effects of ISF at Time 1 and adult input at home at
(as indicated by PLS–4 Auditory Comprehension, weighted Time 1 on child communication at Time 2. In addition, we
total play, MSEL nonverbal, and CCS score) influenced regressed adult input at home onto the latent variable for
both child communication at Time 2 and adult input at ISF to analyze the direct effect between adult input at home
Time 1. Adult input at Time 1 also predicted child commu- and ISF at Time 1. The model using robust maximum like-
nication at Time 2 in the model. Two variables were chosen lihood estimation fit the data fairly well (RMSEA = .091,
as indicators for child communication: (a) rate of different CFI = .966, SRMR = .051). Examination of the residuals for
words produced at home, excluding imitated words, and covariances (i.e., the difference between the observed and the
(b) rate of different words produced in the scripted inter- model-indicated covariance) indicated that the weighted
action protocol, excluding imitated words. Two variables total play scores and the MSEL nonverbal scores were more
were initially hypothesized to be indicators of adult input: related than the model predicted. Adding a residual covar-
(a) adult input at home and (b) adult input at school at the iance between these scores accounted for the statistical

Figure 1. Summary structural equation model for child language outcome (rate of different words from scripted communication assessment and
home observation) with adult input as a mediator. PLS Auditory = Auditory Comprehension scale of the PLS–4. Completely standardized robust
maximum likelihood parameter estimates are shown. **p < .001.

1604 Journal of Speech, Language, and Hearing Research • Vol. 56 • 1595–1612 • October 2013
relationship between these two scores. Theoretically, it makes Table 5. Fit indices and parameter estimates for indirect effects
through mediators.
sense that play and nonverbal MSEL scores would be related,
because they are both nonverbal performance measures
Statistic Model 1 Model 2 Model 3
based on interactions with objects.
Model 2. We evaluated a second model, identical to CFI .96 .986* .983*
Model 1 with the addition of a residual covariance between SRMR .051 .038 .046*
Mullen nonverbal scores and weighted total play scores. This RMSEA .091 .060* .057
model, Model 2, fit the data well (RMSEA = .060, CFI = Indirect effect adult input home .21** .17**
Indirect effect AAC instruction –.05
.986, SRMR = .038). All paths were statistically significant,
indicating that both ISF and adult input at home signif- Note. CFI = comparative fit index; SMSR = standardized root-mean-
icantly influenced child communication 1 year later. There square residual; RMSEA = root-mean-square error of approximation.
was also a relationship between ISF and adult input at home, *p < .05. **p < .01.
although the path weight was somewhat smaller.
To determine whether adult input at home was a
significant mediator of the effect of ISF on child outcomes, communication through both variables. We once again used
we examined the indirect effect of ISF on child communi- bootstrapping to get the appropriate standard error and con-
cation through adult input at home. To properly test the fidence interval. The estimate of the indirect effect through
indirect effect, we used bootstrapping to get the appropriate adult input at home was .17 (SE = .08, p = .023). The 95%
standard error and confidence interval. In finite samples the confidence interval for the indirect effect was .023 to .322.
normality assumption of the indirect effect is rarely met. A Thus, adult input at home remained a significant mediator
resampling strategy such as bootstrapping, in which repeated even with the addition of AAC instruction to the model.
samples of the original data are drawn with replacements to The estimate of the indirect effect through AAC instruction
get an empirical distribution of the observed data, can be was –.05 (SE = .054, p = .342). AAC instruction was not
used to obtain a nonparametric approximation of the sam- a significant mediator in this model.
pling distribution of the indirect effect. Preacher and Hayes In summary, the final model, which included child ISF,
(2008) indicated that bootstrapping methods are preferred adult input at home, and the reported amount of AAC in-
over methods that assume symmetry and normality. The struction predicting the number of different words produced
estimate of the indirect effect was .21 (SE = .09, p = .016). (through speech, sign, or SGD) at the second time point,
The 95% confidence interval for the indirect effect was .039 fit the data well. Adult input at home remained a significant
to .371 and did not include 0. Therefore, we can conclude mediator of ISF even with the addition of amount of AAC
that the effect is statistically significantly different from zero instruction to the model.
and that adult input was a significant mediator of ISF.
Model 3. We evaluated an additional model that added
teacher-reported amount of AAC instruction per week to the Discussion
model. The mean and standard deviation of instruction time The results of this study are unique and compelling
reported are presented in Table 2. Child communication at because we were able to investigate models of early language
Time 2 was regressed onto the latent variable for ISF at Time 1, development based on a large number of preschoolers with
adult input at home at Time 1, and teacher report of AAC IDs and significant language delays observed in authentic
instruction at Time 1. This allowed analyses of the direct effects communication contexts. We evaluated observational data
of ISF at Time 1 and adult input at home at Time 1 on child from 93 preschool-age children observed at home and at
communication at Time 2. In addition, adult input at home and school, in addition to standardized test results, to address
amount of AAC instruction per week were regressed onto two hypotheses about language development in young
the latent variable for ISF. To determine whether amount of children learning AAC.
AAC instruction was a significant mediator of the effect of ISF First, we hypothesized that a set of proposed indicators—
on child outcomes, we examined the indirect effect of ISF on representing both verbal and nonverbal domains theoreti-
child communication through AAC instruction. These indirect cally linked to symbolic development—accurately reflected a
effects are presented in Table 5. latent construct of ISF. Four different measures (language
The model using robust maximum likelihood estima- comprehension, play, nonverbal cognition, and complexity
tion fit the data well (RMSEA = .057, CFI = .983, SRMR = of communication) converged as a coherent construct for
.046). As can be seen in Figure 1, all paths except the path preparedness for learning symbolic relationships. Together,
from ISF to AAC instruction were statistically significant, these measures reflect intrinsic child variables that may
indicating that ISF, adult input at home, and amount of predict successful language outcomes for children learning to
AAC instruction significantly influenced child communica- use AAC and thus should be considered for initial assess-
tion rate 1 year later. ments that drive early language intervention programming.
To determine whether AAC instruction was a signif- Our findings show that these indicators share a sig-
icant mediator of ISF, and whether adult input at home nificant amount of variance and describe a presumably stable
remained a significant mediator of the effect of ISF on child underlying construct. It is not surprising that a complex
outcomes, we examined the indirect effects of ISF on child construct such as a child’s ISF is reflected by multiple

Brady et al.: Language Predictors in Children Learning AAC 1605

measures. Basing a construct on multiple measures is more classrooms; however, a direct comparison is difficult because
robust than a single measure, which may be more susceptible Rowland reported data in terms of observation intervals
to testing artifacts associated with that single measure and not rates. Taken together, these studies suggest that
(Bornstein & Haynes, 1998). In addition, individual chil- children communicate less often at school than at home and
dren’s weaknesses in one indicator may be somewhat com- that interventions aimed at increasing rates of child com-
pensated with strengths in other indicator variables according munication in classroom contexts are still needed (Bunce,
to this model. 1995; Horn, Lieber, Sandall, Schwartz, & Shouming, 2000;
The second hypothesis of this study was that ISF would Smith, Warren, Yoder, & Feurer, 2004).
significantly predict variations in word productions 1 year
later, measured across two contexts: (a) at home with parents
and (b) in a scripted communication sample with a research Clinical Implications
staff member. However, in alignment with an emergentist Our results have implications for clinical assessments
view, our hypothesized model included the amount of adult as well as interventions aimed at improving language out-
input and the amount of AAC instruction as proposed me- comes for children with IDs learning AAC. In terms of
diators of the effects of ISF. The amount of communication assessments, we found that a cluster of child-focused asses-
input from parents and from classroom staff were considered sments, taken together, captured child characteristics asso-
as mediators. Children with higher initial levels of ISF and ciated with differential vocabulary growth. These included
more adult input at home produced more words at Time 2. familiar and common assessments used by clinicians in the
The mediating effect of adult input at home indicates that field of communication disorders (e.g., PLS–4, MSEL
increased adult input can partially overcome effects of ini- nonverbal scales) and our assessments of play and commu-
tially low levels of ISF. The negative path from amount of nication complexity. Our purpose was not to determine
AAC instruction to the amount of different words produced whether child outcomes were best predicted by this versus
suggests that less AAC instruction is associated with more another set of predictors or by some subset of our indicator
word production. However, this is explainable because measures. Each of the indicators we selected was based on
children with more severe impairments are more likely to well-developed theories of early language development.
receive more AAC instruction. The fact that adult input at Approximately 2.5 hr of assessment were required to com-
school was not related to child outcomes or other child plete all of the child assessments, and this amount of time
variables was unexpected but may reflect less variability may seem lengthy. However, this comprehensive assessment
compared to the amount of adult input measured across approach may yield valuable guidance to clinicians in the
home environments. Input from teachers and para-educators beginning stages of plotting a course of treatment. Thus, the
at school was relatively high, with little variation. In addition, amount of time may be a worthwhile investment in terms
we measured only input that was specifically directed to the of better understanding children’s initial levels of intrinsic
child. Including group-directed input might have produced aptitudes for symbol (word) learning.
different results. Our model indicates that interventions aimed at im-
The results of this study complement and extend earlier proving ISF and adult input at home should directly benefit
research in several important ways. In a previous publica- vocabulary development. Although ISF is a latent construct,
tion, Brady et al. (2011) reported that children learning AAC improving the various indicators should strengthen the con-
communicated at relatively low rates in their classrooms, struct and, hence, prepare a stronger foundation for word
most often responding to adult initiations (i.e., initiation acquisition in young children. It is thus worthwhile to con-
rate of .13 communication acts per min and response rate sider curricula that specifically address the indicators of ISF.
of .49 communication acts per min). The current findings For example, Kasari et al. (2008) showed that play levels
add information about communication at home and present can be increased through direct interventions and that en-
a model for early communication development. hanced play improves language outcomes for children with
Previous results by other research teams have provided autism. Several studies have documented that the complexity
comparable results but with smaller and younger samples. of communication can be enhanced through intervention.
For example, Romski, Sevcik, Reumann, and Pate (1989) For example, Brady and Bashinski (2008) and Fey and col-
collected extensive data during mealtimes for nine individ- leagues (2006) have shown that the frequency and diversity
uals with significant disabilities. Participants were between of prelinguistic communicative gestures and vocalizations
6 and 20 years of age, and data indicated that they com- can be increased through direct interventions. Verbal com-
municated more often at home than at school and more often prehension may be targeted directly by increasing opportuni-
to adults than to peers. When converted to rates, Romski ties for children to associate words (represented through speech,
et al.’s data showed about .98 communication acts per min signs, or graphic symbols) with their referents (Brady &
at home and about .38 communication acts per min at McLean, 1998; Nigam, Schlosser, & Lloyd, 2006) or indirectly
school. However, observations in Romski et al.’s study were through methods such as aided AAC input interventions.
limited to mealtime contexts. The school communication Intervention approaches such as aided language stimulation
rates reported by Brady and colleagues (2011) and by Romski that incorporate modeling of AAC and speech input may be
et al. appear higher than those reported by Rowland (1990) particularly effective for children with relatively good verbal
in a study of deaf–blind children communicating in their comprehension skills (Binger & Light, 2006; Harris & Reichle,

1606 Journal of Speech, Language, and Hearing Research • Vol. 56 • 1595–1612 • October 2013
2004). Finally, nonverbal cognition is a multi-indicated con- study, we followed children for 1 year. Additional observa-
struct itself that often includes visual processing skills. Recent tions would allow the modeling of effects over longer pe-
research indicates that visual processing is an important but riods of time and potentially across more complex language
often-ignored component of AAC (Thistle & Wilkinson, 2009; constructs.
Wilkinson & Light, 2011). Although this research has focused We limited our observations to communication be-
primarily on considerations for optimum display construc- tween focus children learning AAC and adult communica-
tions, results also suggest possible interventions aimed at tion partners. Thus, we didn’t include potentially important
directly improving the visual discriminations used in AAC variables of peer input and peer responsiveness. Anecdot-
communication. ally, our observers noted a dearth of peer interactions
Our finding of the importance of adult input is in classrooms, but we were not able to confirm this with
commensurate with a large literature addressing environ- the data collected. Hadley and Rice (1991) reported that
mental contributors to both speech and AAC. It has been peers largely ignored children who were speaking but
recognized that children with language delays may receive who had varying degrees of language impairments in pre-
less language input from parents and peers unless these school classrooms; hence, it would not be surprising to
potential communication partners receive appropriate find similar limitations in interactions between peers and
interventions (Poll, 2011); that is, even though input is even preschool children learning AAC. Further research is needed
more critical for children struggling to learn a symbolic to describe communication (or lack thereof ) between children
communication system, there appears to be a somewhat learning to use AAC and their classmates without dis-
natural tendency to provide less input to these children. abilities, with the goal of developing treatment strategies
Thus, home- and school-based interventions aimed at that teach joint communication and responsiveness via
overriding this tendency by providing adults with specific AAC systems.
language strategies to increase language directed to children The amount of vocabulary available to children on
learning AAC appears to be indicated. Results from our their AAC system is another variable that affects children’s
model indicate that increasing adult input at home would differential AAC vocabulary use over time. For example,
positively affect vocabulary growth. Thus, our results add to if a child has only a few symbols available to select with a
the growing evidence base supporting the importance of an graphic symbol system, the number of different words used
enriched home language environment (Fernald, Marchman, as a measure is limited to the vocabulary available during
& Hurtado, 2008; Hart & Risley, 1995; Huttenlocher et al., that observation. Unfortunately, we did not systematically
1991) and suggest that increasing the amount of communi- record the number of words available to each child via their
cation input across all environments (e.g., both home and AAC systems, and this limitation should be taken into con-
school) is likely to facilitate vocabulary uptake for children sideration in interpreting our results.
with IDs struggling to learn words. Another limitation was that our measure for differ-
A typical model of language intervention for preschool ences in the amount of instruction was limited to teacher
children is to provide intervention at school and communi- report and reflected only their perceptions of differences in
cate with families to try to promote carryover of school- AAC instruction. Other instructional variables, such as
based instruction to home. However, our results suggest that group size, types of AAC, and use of language-facilitating
a more concerted effort to directly intervene with children’s strategies in the classroom, were not analyzed. These var-
parents may produce better outcomes for children. For example, iables may significantly affect vocabulary acquisition out-
parent-based interventions that have typically been used comes. However, the negative correlations between amount
with younger children (see, e.g., Girolametto & Weitzman, of AAC instruction and rate of different words produced
2006; Kaiser, Hancock, & Nietfeld, 2000) could be adapted reported in the current study reflect the difficulty in inter-
and even intensified for use with parents of preschool-age preting longitudinal effects in instructional variables. Al-
children through a dual home-/school-based intervention though it might be assumed that more instruction would
model. One recent intervention that focused on parents of yield more word learning (a positive correlation), our find-
children with autism provided interventions over the Internet ings make sense, because children who have significant
and may provide a feasible model for busy school clinicians language deficits often receive more intervention.
and families (Venker, McDuffie, Ellis Weismer, & Abbeduto, A limitation to our construct of ISF was the focus
2011). on cognitive and not biological indicators. Romski, Sevcik,
and Adamson (1997) pointed out that, in addition to mea-
sures such as those included in our model, biological factors,
Limitations to This Study such as presence of a seizure disorder, may influence a child’s
Although our sample was quite large and relatively ability to take advantage of various language learning op-
diverse in comparison to other studies of children with IDs, portunities. Related to this, we did not examine different
a larger sample would have enhanced the data-analytic diagnoses in our model. Although our purpose was not to
possibilities. In addition, observing for longer periods of compare symbolic communication development across
time within one observation session may yield more rep- different diagnostic groups, diagnosis could be a significant
resentative samples of communication behaviors by chil- predictor of language outcomes in a study with larger
dren and their adult communication partners. In the current numbers of children with specific diagnoses.

Brady et al.: Language Predictors in Children Learning AAC 1607

Conclusions and Future Directions Brady, N., Fleming, K., Thiemann-Bourque, K., Olswang, L.,
Dowden, P., Saunders, M., & Marquis, J. (2011). Development
In this study, we developed and tested a model of early of the Communication Complexity Scale. American Journal of
symbolic development in children with IDs learning AAC, Speech-Language Pathology, 21, 16–28.
based on an emergentist theory of language. Significant var- Brady, N., Herynk, J., & Fleming, K. (2010). Communication
iance in the number of words produced in authentic contexts input matters: Lessons from prelinguistic children learning to use
was accounted for by our model. ISF, adult input at home, AAC in preschool environments. Early Childhood Services, 4,
and the amount of AAC instruction were all significant pre- 141–154.
Brady, N., Marquis, J., Fleming, K., & McLean, L. (2004). Pre-
dictors of child communication outcomes 1 year later. These
linguistic predictors of language growth in children with devel-
findings support the importance of enriching social commu- opmental disabilities. Journal of Speech, Language, and Hearing
nication input across home and school environments during Research, 47, 663–667.
this sensitive period of symbolic word learning, as well as assess- Brady, N., & McLean, L. (1998). Simultaneous and delayed matching
ments of and interventions aimed at improving comprehension, to sample in gesture users and speakers with mental retardation.
play, visual discrimination, and communication complexity. Research in Developmental Disabilities, 19, 409–421.
ISF was found to significantly predict word production Brady, N., Steeples, T., & Fleming, K. (2005). Effects of prelinguistic
in children experiencing a variety of typical early interven- communication levels on initiation and repair of communication
in children with disabilities. Journal of Speech, Language, and
tions during the preschool period. Future studies may find
Hearing Research, 48, 1098–1113.
that it is also a sensitive predictor of responsiveness to dif- Brooks, R., & Meltzoff, A. (2008). Infant gaze following and pointing
ferent types or intensities of interventions (Flippin, Reszka, & predict accelerated vocabulary growth through two years of age:
Watson, 2010). At present, there is little evidence available to A longitudinal, growth curve modeling study. Journal of Child
guide decisions about what approach to use with children Language, 35, 207–220.
learning to use AAC who have different profiles of abilities, Bunce, B. H. (1995). Building a language-focused curriculum for the
histories, and social environments. Important questions such preschool classroom: A planning guide. Baltimore, MD: Brookes.
as these await further research. Butterworth, G. (2003). Pointing is the royal road to language for
babies. In S. Kita (Ed.), Pointing: Where language, culture, and
cognition meet (pp. 9–33). Mahwah, NJ: Erlbaum.
Acknowledgments Carpenter, M., Nagell, K., & Tomasello, M. (1998). Social cognition,
joint attention, and communicative competence from 9 to
This research was supported by Grants DC007684 and
15 months of age. Monographs of the Society for Research in
HD018955 from the National Institutes of Health. We thank the
Child Development, 63(4), i–vi, 1–143.
families and teachers who participated in this research and the
Chapman, R. (2003). Language and communication in individuals
research assistants who helped with data collection.
with Down syndrome. In L. Abbeduto (Ed.), Language and
communication in mental retardation (Vol. 27, pp. 1–34). Boston,
MA: Academic Press.
References Chapman, R., Seung, H., Schwartz, S., & Bird, E. (2000). Predicting
Adamson, L., & Chance, S. (1998). Coordinating attention to people, language production in children and adolescents with Down
objects, and language. In A. Wetherby, S. Warren, & J. Reichle syndrome: The role of comprehension. Journal of Speech,
(Eds.), Transitions in prelinguistic communication (pp. 15–39). Language, and Hearing Research, 43, 340–350.
Baltimore, MD: Brookes. Connor, C. M., Morrison, F. J., & Slominski, L. (2006). Preschool
Bartholomew, D. J., Knott, M., & Moustaki, I. (2011). Latent var- instruction and children’s emergent literacy growth. Journal of
iable models and factor analysis: A unified approach (3rd ed.). Educational Psychology, 98, 665–689.
London, England: Wiley. Crais, E., Day Douglas, D., & Cox Campbell, C. (2004). The inter-
Bates, E., Benigni, L., Brethern, I., Camaioni, L., & Volterra, V. section of the development of gestures and intentionality. Journal
(1979). The emergence of symbols: Cognition and communication of Speech, Language, and Hearing Research, 47, 678–694.
in infancy. New York, NY: Academic Press. Cress, C., & Marvin, C. (2003). Common questions about AAC
Bates, E., & Dick, F. (2002). Language, gesture and the developing services in early intervention. Augmentative and Alternative
brain. Developmental Psychobiology, 4, 293–310. Communication, 19, 254–272.
Bates, E., O’Connell, B., & Shore, C. (1987). Language and com- Cunningham, C., Glenn, S., Wilkinson, P., & Sloper, P. (1985).
munication in infancy. In J. D. Osofsky (Ed.), Handbook of Mental ability, symbolic play and receptive expressive language
infant development (pp. 149–203). New York, NY: Wiley. of young children with Down syndrome. Journal of Child
Binger, C., & Light, J. (2006). Demographics of preschoolers who Psychology and Psychiatry, 26, 255–265.
require AAC. Language, Speech, and Hearing Services in Schools, Dunst, C., Meter, D., & Hamby, D. (2011). Influences of sign and
37, 200–208. oral language interventions on the speech and oral language
Bornstein, M. H., & Haynes, O. M. (1998). Vocabulary competence production of young children with disabilities. CELL Reviews,
in early childhood: Measurement, latent construct, and predic- 4(4), 1–20. Retrieved from
tive validity. Child Development, 69, 654–671. cellreviews/cellreviews_v4_n4.pdf
Brady, N. (2000). Improved comprehension of object names fol- Eadie, P., Ukoumunne, O., Skeat, J., Ruth Prior, M., Bavin, E.,
lowing voice output communication aid use: Two case studies. Bretherton, L., & Reilly, S. (2010). Assessing early communica-
Augmentative and Alternative Communication, 16, 197–204. tion behaviours: Structure and validity of the Communication
Brady, N., & Bashinski, S. (2008). Increasing communication in and Symbolic Behaviour Scales—Developmental Profile (CSBS–
children with concurrent vision and hearing loss. Research and DP) in 12-month-old infants. International Journal of Language
Practice for Persons with Severe Disabilities, 33, 59–71. and Communication Disorders, 45, 572–585.

1608 Journal of Speech, Language, and Hearing Research • Vol. 56 • 1595–1612 • October 2013
Eisert, D., & Lamorey, S. (1996). Play as a window on child de- Kasari, C., Freeman, S. F., & Paparella, T. (2001). Early intervention
velopment: The relationship between play and other develop- in autism: Joint attention and symbolic play. International
mental domains. Early Education & Development, 7, 221–235. Review of Research in Mental Retardation, 23, 207–237.
doi:10.1207/s15566935eed0703_2 Kasari, C., Paparella, T., Freeman, S., & Jahromi, L. (2008). Lan-
Fernald, A., Marchman, V. A., & Hurtado, N. (2008, August). Input guage outcome in autism: Randomized comparison of joint
affects uptake: How early language experience influences processing attention and play interventions. Journal of Consulting and
efficiency and vocabulary learning. Paper presented at ICDL Clinical Psychology, 76, 125–137.
2008, the 7th IEEE International Conference on Development Laakso, M., Poikkeus, A., Eklund, K., & Lyytinen, P. (1999). Social
and Learning, Monterey, CA. interactional behaviors and symbolic play competence as pre-
Fey, M., Warren, S., Brady, N., Finestack, L., Bredin-Oja, S., & dictors of language development and their associations with
Fairchild, M. (2006). Early effects of prelinguistic milieu teaching maternal attention-directing strategies. Infant Behavior and
and responsivity education for children with developmental Development, 22, 541–556. doi:10.1016/S0163-6383(00)0022-9
delays and their parents. Journal of Speech, Language, and Landa, R. J., Holman, K. C., & Garrett-Mayer, E. (2007). Social
Hearing Research, 49, 526–547. and communication development in toddlers with early and later
Flippin, M., Reszka, S., & Watson, L. (2010). Effectiveness of the diagnosis of autism spectrum disorders. Archives of General
Picture Exchange Communication System (PECS) on commu- Psychiatry, 64, 853–864.
nication and speech for children with autism spectrum disorders: Landry, S., Smith, K., Swank, P., & Guttentag, C. (2008). A
A meta-analysis. American Journal of Speech-Language Pathol- responsive parenting intervention: The optimal timing across
ogy, 19, 178–195. early childhood for impacting maternal behaviors and child
Girolametto, L., & Weitzman, E. (2006). It takes two to talk— outcomes. Developmental Psychology, 44, 1335–1353.
The Hanen Program for parents: Early language intervention Legerstee, M., & Fisher, T. (2008). Coordinated attention, declar-
through caregiver training. In R. McCauley & M. Fey (Eds.), ative and imperative pointing in infants with and without Down
Treatment of language disorders in children (pp. 77–104). syndrome: Sharing experiences with adults and peers. First
Baltimore, MD: Brookes. Language, 28, 211–281.
Graham, J. W., Collins, L. M., Wugalter, S. E., Chung, N. K., & Lifter, K., Sulzer-Azaroff, B., Anderson, S., & Cowdery, G. E.
Hansen, W. B. (1991). Modeling transitions in latent stage- (1993). Teaching play activities to preschool children with
sequential processes: A substance use prevention example. disabilities: The importance of developmental considerations.
Journal of Consulting and Clinical Psychology, 59, 48–57. Journal of Early Intervention, 17, 139–159.
doi:10.1037/0022-006X.59.1.48 Lowe, M., & Costello, A. J. (1976). The Symbolic Play Test.
Hadley, P. A., & Rice, M. L. (1991). Conversational responsiveness Windsor, England: Nfer Nelson.
of speech-and language-impaired preschoolers. Journal of Maatta, S., Laakso, M.-L., Tolvanen, A., Ahonen, T., & Aro, T.
Speech and Hearing Research, 34, 1308–1317. (2012). Developmental trajectories of early communication
Harris, M., & Reichle, J. (2004). The impact of aided language skills. Journal of Speech, Language, and Hearing Research, 55,
stimulation on symbol comprehension and production in 1083–1096. doi:10.1044/1092-4388(2011/10-0305)
learners with moderate cognitive disabilities. American Journal MacWhinney, B. (2004). A multiple process solution to the logical
of Speech-Language Pathology, 13, 155–167. problem of language acquisition. Journal of Child Language, 31,
Harrison, B., & Smith, T. (2012). Structural equation modeling to 883–914.
measure well-being and its association with autism symptoms. McCathren, R., Warren, S. F., & Yoder, P. J. (1996). Prelinguistic
Retrieved from predictors of later language development. In K. N. Cole, P. S.
Paper11027.html Dale, & D. J. Thal (Eds.), Assessment of communication and
Hart, B., & Risley, T. (1995). Meaningful differences in the everyday language (pp. 57–76). Baltimore, MD: Brookes.
experience of young American children. Baltimore, MD: Brookes. McCathren, R., Yoder, P., & Warren, S. (1998). Determining spoken
Hoff, E. (2003). The specificity of environmental influence: Socio- language prognosis in children with developmental disabilities.
economic status affects early vocabulary development via mater- American Journal of Speech-Language Pathology, 7, 77–87.
nal speech. Child Development, 74, 1368–1378. doi:10.1111/ McCune-Nicolich, L., & Carroll, S. (1981). Development of sym-
1467-8624.00612. bolic play: Implications for the language specialist. Topics in
Horn, E., Lieber, J., Sandall, S., Schwartz, I., & Shouming, L. (2000). Language Disorders, 2, 1–15.
Supporting young children’s IEP goals in inclusive settings Miller, J. F., & Iglesias, A. (2008). Systematic Analysis of Language
through embedded learning opportunities. Topics in Early Transcripts (Research Version 9.1) [Computer software].
Childhood Special Education, 20, 208–223. doi:10.1177/ Middleton, WI: SALT Software.
027112140002000402 Mullen, E. (1995). Mullen Scales of Early Learning: AGS edition.
Huttenlocher, J., Haight, W., Bryk, A., Seltzer, M., & Lyons, T. Circle Pines, MN: AGS.
(1991). Early vocabulary growth: Relation to language input Muthén, L. K., & Muthén, B. O. (1998–2011). Mplus user’s guide,
and gender. Developmental Psychology, 27, 236–248. sixth edition. Los Angeles, CA: Authors.
Iacobucci, D. (2012). Mediation analysis and categorical variables: Nigam, R., Schlosser, R., & Lloyd, L. L. (2006). Concomitant use of
The final frontier. Journal of Consumer Psychology, 22, 582–594. the matrix strategy and the mand-model procedure in teaching
Iverson, J. M., & Thal, D. J. (1998). Communicative transitions: graphic symbol combinations. Augmentative and Alternative
There’s more to the hand than meets the eye. In A. Wetherby, Communication, 22, 160–177.
S. Warren, & J. Reichle (Eds.), Transitions in prelinguistic com- Noldus Information Technology. (2008). The Observer XT (Version 8.0)
munication (pp. 59–87). Baltimore, MD: Brookes. [Computer software]. Wageningen, the Netherlands: Author.
Kaiser, T., Hancock, T., & Nietfeld, J. (2000). The effects of parent- Philofsky, A., Hepburn, S. L., Hayes, A., Hagerman, R., & Rogers,
implemented enhanced milieu teaching on the social commu- S. L. (2004). Linguistic and cognitive functioning and autism
nication of children who have autism. Early Education & symptoms in young children with Fragile X syndrome. American
Development, 11, 423–446. Journal on Mental Retardation, 109, 208–218.

Brady et al.: Language Predictors in Children Learning AAC 1609

Piaget, J. (1962). Play, dreams and imitation in childhood. New York, Tapp, J. T., & Yoder, P. (2000). Playcoder: Aiding the coding of
NY: W. W. Norton. play in toddlers [Computer software]. Vanderbilt University,
Poll, G. H. (2011). Increasing the odds: Applying emergentist theory Nashville, TN.
in language intervention. Language, Speech, and Hearing Ser- Thiemann-Bourque, K., Brady, N., & Fleming, K. (2012). Symbolic
vices in Schools, 42, 580–591. play of preschoolers with severe communication impairments
Preacher, K. J., & Hayes, A. F. (2008). Asymptotic and resampling with autism and other developmental delays: More similarities
strategies for assessing and comparing indirect effects in multiple than differences. Journal of Autism and Developmental Disorders,
mediator models. Behavior Research Methods, 40, 879–891. 42, 863–873. doi:10.1007/s10803-011-1317-7
Prizant, B., Wetherby, A., & Rydell, P. (2000). Communication Thistle, J. J., & Wilkinson, K. (2009). The effects of color cues on
intervention issues for children with autism spectrum disorders. typically developing preschoolers’ speed of locating a target
In A. Wetherby & B. Prizant (Eds.), Autism spectrum disorders: line drawing: Implications for augmentative and alternative
A transactional developmental perspective (Vol. 9, pp. 193–225). communication display design. American Journal of Speech-
Baltimore, MD: Brookes. Language Pathology, 18, 231–240.
Romski, M., & Sevcik, R. (1993). Language comprehension: Con- Thurm, A., Lord, C., Lee, L., & Newschaffer, C. (2007). Predictors of
siderations for augmentative and alternative communication. language acquisition in preschool children with autism spectrum
Augmentative and Alternative Communication, 9, 281–285. disorders. Journal of Autism and Developmental Disorders, 37,
Romski, M., Sevcik, R., & Adamson, L. (1997). Framework for 1721–1734.
studying how children with developmental disabilities develop Tomasello, M. (2003). Constructing a language: A usage-based theory of
language through augmented means. Augmentative and Alter- language acquisition. Cambridge, MA: Harvard University Press.
native Communication, 13, 172–185. U.S. Department of Education. (2011, January). IDEA Data, Part B,
Romski, M., Sevcik, R., Adamson, L., Cheslock, M., Smith, A., child count. Washington, DC: Office of Special Education Programs.
Barker, R. M., & Bakeman, R. (2010). Randomized comparison Vandereet, J., Maes, B., Lembrechts, D., & Zink, I. (2010).
of augmented and nonaugmented language interventions for Predicting expressive vocabulary acquisition in children with
toddlers with developmental delays and their parents. Journal of intellectual disabilities: A 2-year longitudinal study. Journal of
Speech, Language, and Hearing Research, 53, 350–364. Speech, Language, and Hearing Research, 53, 1673–1686.
Romski, M., Sevcik, R., Reumann, R., & Pate, J. (1989). Youngsters Venker, C. E., McDuffie, A. S., Ellis Weismer, S., & Abbeduto, L.
with moderate or severe mental retardation and severe spoken (2011). Increasing verbal responsiveness in parents of children
language impairments: I. Extant communicative patterns. with autism: A pilot study. Autism, 16, 568–585. doi:10.1177/
Journal of Speech and Hearing Disorders, 54, 366–373. 1362361311413396
Rowland, C. (1990). Communication in the classroom for children Watt, N., Wetherby, A., & Shumway, S. (2006). Prelinguistic
with dual sensory impairments: Studies of teacher and child behavior. predictors of language outcome at 3 years of age. Journal of
Augmentative and Alternative Communication, 6, 262–274. Speech, Language, and Hearing Research, 49, 1224–1237.
Sevcik, R., & Romski, M. (1997). Comprehension and language Werner, H., & Kaplan, B. (1984). Symbol formation. Hillsdale, NJ:
acquisition: Evidence from youth with severe cognitive dis- Erlbaum.
abilities. In L. Adamson & M. Romski (Eds.), Communication Wetherby, A., & Prizant, B. (2003). Communication and Symbolic
and language acquisition: Discoveries from atypical development Behavior Scales manual—Normed edition. Baltimore, MD: Brookes.
(pp. 184–201). Baltimore, MD: Brookes. Wetherby, A., Watt, N., Morgan, L., & Shumway, S. (2007). Social
Sevcik, R. A., & Romski, M. A. (2002). The role of language com- communication profiles of children with autism spectrum
prehension in establishing early augmented conversations. In J. disorders late in the second year of life. Journal of Autism and
Reichle, D. Beukelman, & J. Light (Eds.), Implementing an aug- Developmental Disorders, 37, 960–975.
mentative communication system: Exemplary strategies for begin- Wilkinson, K. M., & Light, J. (2011). Preliminary investigation of
ning communicators (pp. 453–474). Baltimore, MD: Brookes. visual attention to human figures in photographs: Potential
Smith, J., Warren, S. F., Yoder, P. J., & Feurer, I. (2004). Teachers’ considerations for the design of aided AAC visual scene dis-
use of naturalistic communication intervention practices. Jour- plays. Journal of Speech, Language, and Hearing Research, 54,
nal of Early Intervention, 27, 1–14. 1644–1657. doi:10.1044/1092-4388(2011/10-0098
Stone, W. L., Lemanek, K. L., Fishel, P. T., Fernandez, M. C., & Zimmerman, I. L., Steiner, V., & Evatt Pond, R. (2003). Preschool
Altemeier, W. A. (1990). Play and imitation skills in the diagnosis Language Scale—4. San Antonio, TX: The Psychological
of autism in young children. Pediatrics, 86, 267–272. Corporation.

1610 Journal of Speech, Language, and Hearing Research • Vol. 56 • 1595–1612 • October 2013
Appendix A
Communication Complexity Scale

Scale number Definition Example Communication level

0 No response An opportunity is presented but the child

looks away the entire time.
1 Alerting—a change in behavior A vibrating toy stops vibrating and the child Preintentional
stops smiling.
2 Single object/event/person orientation Child handles a toy and focuses attention on Preintentional
only this toy.
3 Single object/event/person Child handles a toy and vocalizes. Preintentional
orientation plus 1 PCBa
4 Single object/event/person Child reaches toward a toy and vocalizes. Preintentional
orientation plus more than 1 PCB
5 Scanning between objects/events The child visually scans (shifts attention) Preintentional
between two different wind-up toys.
6 Dual orientation between a person While playing with a wind-up toy, the child Preintentional
and an object or event looks up at partner.
7a Triadic eye gaze Child looks from toy to partner and back to toy Intentional nonsymbolic
within a few seconds.
7b Dual orientation plus 1 or more PCBs Child looks from toy to partner and vocalizes. Intentional nonsymbolic
8 Triadic eye gaze plus 1 PCB Child looks from toy to partner and back to toy Intentional nonsymbolic
while vocalizing.
9 Triadic eye gaze plus more than 1 PCB Child looks from toy to partner and back to toy Intentional nonsymbolic
while vocalizing and giving toy to partner.
10 One word verbalization, sign, or Child says “More” to request more Cheerios. Intentional symbolic
AAC symbol selection
11 Two or more word verbalizations, Child says “More please” to request more Cheerios. Intentional symbolic
signs, or AAC symbol selections

Note. From “Development of the Communication Complexity Scale,” by N. Brady, K. Fleming, K. Thiemann-Bourque, L. Olswang, P. Dowden,
M. Saunders, and J. Marquis, 2011, American Journal of Speech-Language Pathology, 21, p. 21. Copyright 2011 by the American Speech-
Language-Hearing Association. Adapted with permission.
Includes behaviors such as vocalizations, gestures, eye gaze, or switch closures that appear to be purposeful in response to the stimulus and that
could be viewed as communicating behavior regulation or joint attention or social interaction.

Brady et al.: Language Predictors in Children Learning AAC 1611

Appendix B
Brief definitions of communication categories and communication modes.

Communication Categories
Adult communication: includes initiations, prompts, or responses from an adult to the target child. When multiple adults
are communicating, only the last adult to communicate is coded. During group activities, only adult communication clearly
directed to the focus child is coded. Prompts are defined as specific and direct prompts to help the child use words, sign,
symbols, or an SGD. “No responses” are coded if 3 s elapse after a pause in communication.
Child communication: includes initiations, responses, or repairs from the target child to an adult. Repairs are defined as
any time a child responds to the adult’s specific request for clarification when the adult does not understand the child’s initiation or
response. “No responses” are coded if 3 s elapse after a pause in communication.

Communication Modes
Speech: an utterance containing one vowel and one or more target consonants (or childlike consonant substitutions of the
target word) in the appropriate position per syllable in the word. The word must be used in a semantically and pragmatically
appropriate manner.
Sign: a combination of hand configurations that represent an expression, a word; a letter; a number; or a combination of
words, letters, or numbers that are representative of a sign language system, including American Sign Language.
Gesture: a nonverbal communicative behavior that can be easily recognized by other members of the cultural community.
Examples of conventional gestures include a head nod or shake to indicate yes or no, shoulder shrugs, finger up to mouth in
a “shush” gesture, or a raised hand in a “high five” gesture. Examples of nonconventional gestures include a child leading an adult
by the hand, tapping someone, or giving or showing an object to another person. Includes pointing gestures.
Symbol: photographed, colored, or black-and-white line drawings or object symbols that represent an object, activity, or an
event that are pointed to, exchanged between communication partners, or held up by the adult to encourage the child to focus on
the symbol. Picture cards must be part of the child’s targeted communication system; simple flash cards used during a receptive
language tasks are not considered symbols unless they contain a letter, word, or numbers.
Vocalization: a single or repeated phoneme (sound) that is not a true word but has meaning and is directed to a person.
SGD: the act of directly selecting an option (photograph, colored, or black-and-white line drawings, or an object symbol) on
the SGD, after which a recorded message is played.

1612 Journal of Speech, Language, and Hearing Research • Vol. 56 • 1595–1612 • October 2013
Copyright of Journal of Speech, Language & Hearing Research is the property of American
Speech-Language-Hearing Association and its content may not be copied or emailed to
multiple sites or posted to a listserv without the copyright holder's express written permission.
However, users may print, download, or email articles for individual use.