A Protocol for Assessing Early Communication of

4 downloads 0 Views 1MB Size Report
th-old boy, might with autism have no speech) to delayed development of btain it. Locution functional ... listener on the arm, clapping, and screaming may be functionally .... In the morning, a parent hands the child a long- sleeve shirt made of a ...
49

TECSE 27:1 49-61 (2007)

A Protocol for Assessing Early Communication of Young Children With Autism and Other Developmental Disabilities

Jim Halle Universily of Illinois Hedda Meadan Illinois State Universily

T

he purposes of this article are (a) to describe a structured protocol for assessing request, reject, and repair behavior of young children with autism spectrum dis-

orders and other developmental disabilities and (b)to highlight the importance

of assessing these communicative behaviors. These three communicative functions are basic building blocks for social encounters, permitting the "speaker" to influence others in desired ways. Concepts of operant psychology, such as response class, functional equivalence, response competition, and response efficiency, are invoked to provide a conceptual framework for the assessment. To illustrate how the structured protocol can be implemented and the types of findings resulting from the assessment, an actual case study is presented with the data on requesting and rejecting for five young children with autism.

Communication occurs in the context of an interaction between a speaker and a listener. The success of a communication act depends on whether tIie listener appropriately intekprets the speaker's inten tions. Therefore, effective communication occurs when the communicative intention (i.e., the goal) of the spea ker is the same as the communicative function interpret ed by the listener (Wetherby, Warren, & Reichle, 1998). The development of intentional ity, according to Bates, Camaioni, and Volterra (1975), consists of three stages. They assumed developmental continuity from preintentional through intentional preve'rbal (around the age of 9-10 months) to intentional verb al communication (around the age of 12-13 months). Perlocution is the first stage and involves a signal issued by orne person that has an effect on the listener. For example, the cry of babies when they are hungry might unintenttionally influence mothers to provide them with milk. IllIocution is the second stage. An illocution requires the intentional use of a conventional signal to carry out some scocially recognized function. For example, John, a 10-mon th-old boy, might intentionally point to a cup of milk to obtain it. Locution is the last stage and includes all of th.e acts associated with speech. A locution requires the onse t of verbal speech. For example, a child might say, "Want milk," to obtain a cup of milk.

Bates and her colleagues (Bates et al., 1975) demonstrated that young children communicate intentionally even before they develop language. Ogletree, Wetherby, and Westling (1992) described three primary intentional communicative functions: (a) bebavior regulation,to request or reject an object/action (e.g., a young child can request a cookie by pointing to the container of cookies and vocalizing or reject an un-preferred toy by pushing it away); (b) social interaction,to request a social routine, show off, or greet (e.g., a child might wave good-bye when his father leaves for work or when his father returns home, -he might run to him and pull his coat as an invitation to play); and (c) joint attention, to request information or comment (e.g., a child might point to the sky to direct his mother's attention to a helicopter). Young children with an autism spectrum disorder (ASD) or other developmental disabilities (DD) often experience challenges in the acquisition of communication and language. These challenges range from a failure to develop any functional speech (33%-S0% of individuals with autism have no speech) to delayed development of functional speech and language (Wetherby, Prizant, & Schuler, 2000). Two major areas of communication deficits are the development of joint attention and symbol use; both are important ingredients in effective social interaction (Wetherby et al., 2000).

Address: Jim Halle, Departmentof Special Education,University of Illinois, 1310 S. Sixth St., Champaign,IL61820; V e-mall: [email protected]

Spro.ed

Topics In Early Childhood Special Education 27:1

so

Toplc In Early Childhood Special Education 27:1

50 Children with ASD and DD who have limited expressive language often communicate for behavior regulation (i.e., requesting and rejecting) using unconventional and idiosyncratic nonverbal behavior. Compared to typically developing children, their messages are more difficult to understand and frequently produce communication breakdowns (Halle, Brady, & Drasgow, 2004; Keen, 2003). When faced with a breakdown in communication, repair strategies, such as repeating or modifying the initial communication act, are used to produce the desired outcome, and they represent a level of persistence (Wetherby & Prizant, 1993). Communication repairs may be considered an early indicator of intentional communication (Wetherby, Alexander, & Prizant, 1998) and are important from theoretical and clinical perspectives (Meadan, Halle, Watkins, & Chadsey, 2006). Their theoretical relevance emanates from their relationship to intentionality; evidence for the transition from pre-intentional to intentional communication is unequivocal when children acquire repairs. Clinically, if children with autism communicate prelinguistically and their communicative efforts are often misunderstood due to their unconventional or idiosyncratic nature, then repairs become the fundamental mechanism for self-efficacy and selfdetermination. That is, communicative repairs become the primary way in which many children influence people in their environment.

CONCEPTUAL FRAMEWORK Prelinguistic communicative behavior, whether it be conventional (e.g., pointing, head nodding) or unconventional (e.g., leading, screaming, hitting), can be organized into response classes. According to Carr (1988), a response classes is a "group of two or more topographically different behaviors, all of which have the same effect on the environment" (p. 222). Producing the same effect on the environment means that the varying behaviors are maintained by the same consequence or reinforcer. If two or more behaviors have the same effect on the environment, they can be said to be functionally equivalent (Carr, 1988). Thus, for the children who are the focus of this communication assessment protocol, tapping a potential listener on the arm, clapping, and screaming may be functionally equivalent communicative topographies that produce a common outcome: listener attention. Likewise, when a child is brought to the large-group circle time activity, he or she may scream, fall to the ground, or run away from the group. Although each of these behaviors differs in topography, they share the same communicative function of escape; therefore, they constitute a response class in which each member is functionally equivalent. The assumption is that each of these members has been reinforced in the past by producing escape from circle time.

Two additional concepts that are heuristic in any discussion of response classes are response competition and response efficiency. If members of a response class share the same function and thereby produce the same outcome, how can we predict which member will occur on any particular occasion? In a sense, they are in competition with one another, and depending on the conditions operating in a particular situation, one of the members will prevail (and be expressed). According to this framework, the specific member expressed will be the most efficient one for that situation, with every other member being arranged in a hierarchy of probability in reference to that particular situation. Recently, researchers have begun to identify factors that determine the efficiency of a response: (a) response effort, (b) schedule of reinforcement, (c) immediacy of reinforcement, (d) quality of reinforcement (Homer & Day, 1991; Mace & Roberts, 1993; Neef, Mace, & Shade, 1993; Neef, Mace, Shea, & Shade, 1992), and (e) history of punishment for the response (Halle & Drasgow, 2003). Briefly defined, response effort pertains to the physical (measured in calories) and/or emotional effort expended in producing the response. Schedule of reinforcement refers to the number of times the response must be produced before the outcome is achieved. Immediacy of reinforcement is defined by the latency between the response and obtaininig the outcome. Quality is determined by the reinforcing value of the outcome. And history of punishinent for the responserefers to prior experiences in which a class member has been punished. It is noteworthy that only the first factor, response effort, is entirely determined by the child; the other four factors are heavily influenced by listeners or social partners. How is this conceptual framework relevant to our assessment protocol? The purpose of any assessment methodology is to gather information that will inform intervention design and development. If a child requests materials using a particular topography (e.g., vocalizing) and then is tAught a new request topography (e.g., handing a listener a picture of the desired item), the two topographies form a response class and potentially compete with one another for expression. Thus, in any situation in which a new communicative topography is taught (i.e., intervention) to replace an existing topography, concepts of response class, response competition, and response efficiency are relevant. Of course, the response classes formed in the lives of real children will not be as simple as one topography for one function; therefore, the analysis of class membership and the efficiency of members in a particular context can be quite daunting.

ASSESSMENT OF COMMUNICATION BEHAVIOR The primary goal of assessing the communication of nonverbal children or children who have very limited ex-

51

Communication of Young Children With Autism

pressive language is to generate a clear, reliable, and representative description of their communication behavior (i.e., topographies) that could be useful for developing appropriate treatments (Ogletree, Fischer, & Turowski, 1996). Assessing communicative behavior of young children with ASD or DD is challenging due to the unique and varied social and behavioral features associated with these disabilities. Ogletree, Pierce, Ham, and Fischer (2002) recommended five strategies for addressing unique features of ASD during assessment: 1. modify procedures and practices according to the child's unique characteristics; dynamic assessment might permit such an individualized approach to identifying a child's communication abilities; 2. collect data external to the traditional assessment process; assessment needs to involve not only the child with disability, but also communication partners and contexts; 3. incorporate informant-based procedures; a child with ASD or other DD might not exhibit optimal communication abilities in unfamiliar settings or with unfamiliar partners; using informant-based assessment procedures (e.g., questionnaires) could provide a more complete picture of the communication abilities; 4. conduct multiple observations, if possible, in different settings; 5. use functional activities; assessment activities should be natural and engaging, with clear expectations. Five different approaches have been used to assess communication behavior of individuals with disabilities and limited expressive language: caregiver interviews, observations of naturally occurring caregiver-child interaction, structured sampling, formal assessment measures, and observations across settings with different communication partners (Ogletree et al., 1996). All five of these have been described at length elsewhere, with one possible exception. The focus of this article is on structured protocols for assessing request, reject, and repair behavior of young children with ASD and other DD who have limited expressive language. Our rationale for targeting these three communicative functions in this population is that, in our experience, these are the pragmatic functions most often used by young prelinguistic children with autism. These children are communicating intentionally but without a symbol system, and they communicate to influence the behavior of others to obtain preferred (and avoid unpreferred) encounters, materials, and activities. The assessment is sensitive to the child's current developmental level; few of these children possess more social functions, such as joint attention, commenting, or information

seeking. A number of researchers (e.g., Casby & Cumpata, 1986; Cirrin & Rowland, 1985; Coggins, Olswang, & Guthrie, 1987; Iacono, Carter, & Hook, 1998; Lichtert, 2003; Ogletree et al., 1992) have described protocols for assessing the intentional communication of children with disabilities. These protocols include "routines involving highly motivating stimuli (e.g., food items, toys that moved or that made noises) and activities (e.g., making Kool-Aid, tickling) guided by a familiar adult" (Cirrin & Rowland, 1985, p. 53). The structured protocol presented here is unique in at least three respects: (a) the highly motivating stimuli are individually, rather than generically, determined; (b) the assessment focuses on individual children instead of large numbers of children; and (c) repeated trials are presented over time to each child, rather than presenting only a few trials to many children. Although prescriptive, this approach to assessment requires individualizing the protocol for each child. Unlike most structured protocols that are designed for pre-post assessment, the feature that entails repeated trials delivered over time permits continuous testing to monitor children's performance intermittently and longitudinally. Although the protocol has not been used for this purpose, it would seem to accommodate evaluations of effectiveness for programs that focus on early communication of children with autism and other developmental disabilities. In addition to the multiple-trial sessions (described later), structured probes strategically embedded in natural routines could be a fundamental component of a program-evaluation process. These structured probes would be administered intermittently to assess any changes in request, reject, and repair topographies, as well as changes in repair persistence under challenge conditions (i.e., intentional breakdowns delivered by social partners). Figure 1 describes specific examples of structured probes that could be repeatedly but intermittently administered for purposes of program evaluation. METHOD

Assessment Protocol The assessment protocol consists of three steps: 1. assessingpreferences by interviewing parents and teachers and observing the child in natural settings; 2. building rapportby interacting with the child in preferred routines, following the child's lead, and being responsive to the child's communication attempts; and 3. assessingin the context of a structuredprotocol by intentionally programming opportunities for requests, rejects, and repairs. These steps in the assessment protocol represent an effort to optimize the assessment context to produce the

Topics In Early Childhood Special Education 27:1 Topic; In Early Childhood Special Education 27:1

52

52 best communicative performance possible in as efficient a manner as possible. Thus, the outcome of this assessment process would not represent how the child would perform when (a) the environment lacks opportunities or reasons to communicate or (b) only unfamiliar social part-

1. During the naturally occurring routine of toothbrushing, the toothpaste dispenser that is made In the shape of a dinosaur Is placed on a shelf out of reach of the child, who must emit a request topography to obtain access to the dinosaur. 2. During the naturally occurring routine of dressing In the morning, a parent hands the child a longsleeve shirt made of a material that Is known to be uncomfortable for the child. The child must emit a reject topography to avoid touching the shirt. 3. Exploiting either of the first two scenarios above, the social partner Introduces a breakdown by feigning a lack of understanding. For example, If the child looks at the parent and reaches for the dinosaur, the parent might say, "Here, let me turn on the water for you." In this case, a repair topography Would be assessed. Similarly, Ifthe child rejected the shirt in the scenario above by knocking It out of the parent's hand, the parent might feign misunderstanding by saying, "Oops, the shirt fell. Here, let me help you" and representing It to assess how the child would repair the breakdown. Multiple opportunities could be embedded Into the many ongoing routines that occur on a daily basis at home and school. In cases where routines are difficult to Identify, they can easily be developed. In order to examine progress or change In communicative topographies (and to assess whether those targeted by the Intervention are used In everyday routines), the naturalistic probe occasions must be controlled as much as possible; otherwise, changes In topographies may be a function of the different conditions that are present when the opportunity Is Introduced. For example, If Mom conducts a toothpasterequesting probe on one occasion and Dad does It on another, these two opportunities are not equivalent, and any difference In topography could be attributed to who Is conducting the probe, Instead of to the Intervention program Itself. Of course, If the Intervention program Issuccessful, then we would hope to see the newly taught topography used with both Mom and Dad. FIGURE 1. Examples of structured probes for evaluating an Intervention program.

ners are available for interaction. It would be possible, and probably informative, to arrange occasional probing under these less-than-optimal conditions to examine how the child's communication attempts vary as a result of these varying conditions. However, this is not the purpose of the approach to assessment described herein. We implemented the protocol described herein with seven young children who have developmental disabilities. The assessment protocol was individualized for each child (i.e., materials and activities chosen for the sessions varied by child). In addition to generating operational definitions for each child's behavior, we also developed a set of recording rules for each child such that on any trial, requesting and rejecting were mutually exclusive (lists of recording rules are available from the authors upon request). Table 1 illustrates the results from multiple sessions conducted with five of the children. (Detailed information on the assessment of the two additional children can be found in Meadan et al., 2006.) Multiple topographies of requesting and rejecting were displayed within and across children; however, a high degree of consistency across repeated trials within a child was found (e.g., Ben pointed on 79% of the request trials and gave back oin 61% of the reject trials). In addition, although children occasionally emitted the same topography on requesting and rejecting occasions (e.g., Ben vocalized and tapped), they overwhelmingly used different topographies, reflecting their discriminated responding to these occasions. Reliability of the recording was conducted on approximately 25% of the sessions, chosen at random, for each participant. A second observer watched the videotape independently and recorded the occurrence of type of response (request/reject/repair) and topography of response (one or more of the topographies identified for request and reject). Prior to the reliability checks, the second observer was trained by the second author until high reliability was achieved. An agreement (i.e., point-by-point) was defined by occasions when both observers (a) recorded the same code for type of response and (b) recorded at least one common topography for topography of response. Overall, for the five children whose data are displayed in Table 1 (reliability data for the other two children are reported in Meadan et al., 2006), the range of agreement for type of response was 76% to 100%, with a mean of 94%, and the range of agreement for topography of response was 64% to 100%, with a mean of 90%.

BEN:

A CASE ILLUSTRATION

To illustrate the steps of the assessment protocol, we describe the case of Ben, a 3-year-old boy with autism. Ben's case is invoiced not only to exemplify the assessmient protocol but also to demonstrate the variety of out-

53

Communication of Young Children With Autism

TABLE 1. Topographies of Request and Reject for Five Young Children With Autism or Other Developmental Disabilities Topographies

Type of response Child Alan

Justin

Reject

Request

15% (23)

85% (127)

28% (45)

72% (119)

More (64%) Reach (55%)

Throw (48%)

Bite (4%) Other (3%) Grab (1%)

More (44%) More (17%) Other (Other = push away [2], cry [1]; 9%) Grab partner (4%)

Reach (78%) Vocalize (33%)

Vocalize (55%) Head nod (no; 43%) Sign (all done; no; 43%) Give back (32%) Reach (5%) Point (2%)

Throw +

Point (31%) Sign (12%; yes = 11, more 2) Other (11%) Milo

5% (8)

95% (144)

Reject trials

Request trials

Reach (68%)

Word ("I want") (68%) Open palm (43%) Point (21%) Vocalize (4%) Other (4%)

Push away (50%) Vocalize (50%) Word (no; 38%) Bang head (13%) Head nod (no; 13%) Reach (13%)

Ben

14% (38)

86% (230)

Point (79%) Head nod (yes; 43%) Vocalize (25%) More (21%) Tap (10%) Reach (11%)

Give back (61%) Point (13%) No (10%) Other (10%) Vocalize (10%) Tap (3%) Head nod (3%) More (3%)

Judy

13% (17)

87% (112)

Point (94%) Vocalize (90%) Reach (6%)

Give back (88%) Sign (all done; 35%) Vocalize (11%)

15%

85%

Overall

Note. Numbers in parentheses next to "Type of response" refer to the number of trials the child requested or rejected; percentages next to topographies refer to the percentage of request or reject trials in which the child used each topography.

comes or results that can be derived from the assessment. Ben lives at home with his adoptive parents and is an only child. Both his parents work outside the home, and he attends an early childhood program for children with disabilities and children at risk for behavior and academic difficulties. Ben has very limited expressive language (i.e., less than 10 functional words and signs) and receives speech and occupational therapy. Ben's parents and teachers reported that Ben has difficulties with transitions and expression of his wants and needs. He often screams, bites, and kicks when he wants something that

is not delivered immediately or when he is required to engage in an activity that he doesn't like. As a result of his behavior, adults who work with him are in a constant state of stress and often avoid introducing tasks that Ben does not appear to enjoy.

Preference Assessment Various indirect procedures (i.e., care providers and client interviews) and direct procedures (i.e., systematically exposing participants to stimuli) of preference as-

54

Topics In Early Childhood Special Education 27:1

sessment for individuals with developmental disabilities are described in the literature (Hagopian, Long, & Rush, 2004). Indirect measures of preference are often used because they are easy to implement, efficient in time, and require limited resources (Hagopian et al., 2004). Prior to the assessment of the communication topographies of requesting, rejecting, and repairing, preferred and unpreferred foods, objects, and activities need to be identified. Ben's parents and teachers were asked to provide information regarding Ben's preferences. This information is summarized in Table 2. In addition, Ben was observed in the early childhood program, while playing and engaging with various objects and activities, to validate the information gathered from the parents and teachers. The information from the preference assessment was embedded in the structured protocol, a unique feature of this assessment approach. RapportBuilding A few researchers have suggested that the quality of the relationship (i.e., rapport) between individuals with disabilities and others may influence the communicative behavior of individuals with disabilities (Carr, Levin, McConnachie, Carlson, Kemp, & Smith, 1994; McLaughlin & Carr, 2005). Ogletree et al. (2002) argued, "Individuals with autism frequently will not exhibit optimal communicative or other abilities in unfamiliar settings or

in the presence of unfamiliar partners" (p. 64). Guidelines for building rapport are presented in Figure 2. Prior to the structured assessment of communication topographies, the assessor spent time in Ben's early childhood program. To become more familiar, the assessor helped the teacher with regular classroom activities and observed Ben's interactions with other children. Following a few hours of naturalistic observation, the assessor interacted with Ben, delivering preferred items noncontingently (e.g., more food or drink during snack time) and playing with his most preferred activities and toys (e.g., books and pop-up toys). When Ben showed Interest in a different activity than the one with which he was currently engaged (e.g., pointed to another toy or moved to a different area of the classroom), the assessor followed his lead. After a number of hours of one-onone interaction with Ben, the assessor was well acquainted with him, had developed a positive relationship (iLe., rapport), and had begun to establish a routine that approximated the structured protocol. After a few semistructured one-on-one sessions in Ben's classroom, the assessor gradually transitioned Ben to a separate room and conducted a few sessions in the new environment that gradually introduced more and more structure. Following this rapport-building effort, Ben showed excitement when he saw the assessor enter the classroom and immediately approached her, smiled, and took her hand. He was ready to go to the other room and "play."

TABLE 2. Summary of Likes and Dislikes for Ben From Reports by Parents and Teachers Preference categories Food/drink

Like Apple Animal crackers Banana Milk

Dislike Oranges Macaroni and cheese Rice Most meat

CheeriosTMI

Olives Cookies Oranges Object

Books Sensory toys (vibrate and music) Sesame StreetT1M toys

Blocks Cars

Activities

Playing with pop-up toys Playing with water Blowing bubbles Working a foam puzzle Climbing Hammering things

Working on wooden puzzles Drawing (holding the color) Watching TV/movies/cartoons Riding a tricycle Shooting baskets

Unique likes and dislikes

Swinging

Putting on shoes and socks Traositioning from (gym to classroom)

55

Communication of Young Children With Autism

1. Be familiar. Spend time in the child's natural, everyday settings. 2. Be associated with positive outcomes. Deliver preferred items and activities noncontingently to the child (i.e., the child is not required to perform any particular action in order to receive the preferred event). The child will attend to your presence and approach you. 3. Engage In positive, fun, unstructured interactions with the child: a. Play with the child in the context of the most preferred objects and activities that were identified In the preference assessment. b. Follow the child's lead during play. The child's interests at any moment in time should determine the flow of activities. c. Be responsive to the child's communication attempts by acknowledging them and either honoring them or attempting to discover their purpose. 4. Build a routine similar to the routine that will constitute the structured protocol: a. Create opportunities for communication by playing with the child with preferred items. b. Program opportunities for requests, rejects, and repairs (see structured protocol, Figure 3).

FIGURE 2. Guidelines for building rapport.

StructuredAssessment The purpose of the structured assessment is to identify the behavior topographies that convey each of the three functions: request, reject, and repair. During structured assessment sessions, the assessor followed a scripted protocol that included creating opportunities for requesting, rejecting, and repairing (see protocol in Figure 3). On each trial the assessor presented one item from a bag that included preferred and unpreferred food, objects, and activities that were identified in the preference assessment. To ensure positive and enjoyable sessions, most of the child's conventional topographies for requests and rejects were honored immediately, and only a few breakdowns (i.e., the assessor pretended, to misunderstand the child's

communication attempt and asked, "What?") were programmed during the session. A repair was recorded if the child responded within 3 s of a breakdown. Each session was 15 to 20 min long and included about 30 trials (see Table 3 for model session).

1. Establish a routine and develop a positive relationship with the child that will motivate him or her to stay and play. 2. Encourage child to sit down on a chair or on a carpet. Having the child stationary is helpful but not an absolute necessity when delivering opportunities. 3. Present (take out of the bag and show out of reach) a preferred object/activity and vocalize about the object/activity (i.e., "Lets play with..." "Ihave a ..."), and wait 5 seconds for a response (i.e., request opportunity). 4. Ifthe child does not respond, model by playing the activity or with the object and wait 10 seconds for a response. 5. For #3 through #4: a. If the child rejects the item with behavioral indication (e.g., pushes it away), remove it from child's sight. b. If the child requests the item with behavioral Indication (e.g., reaches), give item to the child. c. If the child does not respond after the 15-second sequence (#3 + #4), offer a new item or activity. 6. After playing with one item (e.g., finished the puzzle, rejected the object) or with the same item for 20 to 30 seconds, offer a new item. 7. Present (take out of the bag and show but retain out of reach) unpreferred object/activity and wait 3 seconds for a response (i.e., reject opportunity). If the child does not respond after 3 seconds, give the item to the child or place it on table/ground within reach. If child rejects, remove object/activity immediately. If reject topography is a problem, prompt an alternative replacement (e.g., touch symbol, says/signs "stop"), ensure it is used, then remove object/ activity. If child requests, treat as a request trial. Present unpreferred item on half the number of trials that are breakdowns (e.g., 30 trial sessions with 10 breakdowns = 5 reject trials). 8. If the child requests to leave the session, attempt to redirect with a preferred item or activity to maintain child in session. If child continues to request terminating the session or does not want to return to the table, honor by giving the child a 2-minute break and then start a new trial. Programa Single IntentionalBreakdown * On an average of every third request trial (i.e., honoring 2 of 3 requests), look at the child and say, "What?" If the child does not produce a repair after 3 seconds, begin a new trial by presenting a new object. * Once or twice during a session of 30 trials, on reject trial, look at the child and say, "What?" FIGURE 3. Protocol for the structured assessment.

Topics In Early Childhood Special Education 27:1

56

Topics In Early Childhood Special Educalion 27:1

56 TABLE 3. Model Session With Request, Reject, and Repair Opportunities Trial number

Type of trial

1 2 3 4

Request Request Request + breakdown Request Reject Request + breakdown Request Request + breakdown Reject Request Request Request + breakdown Request Request Reject + breakdown Request Request + breakdown Request Request Request + breakdown Request Reject Request Request Request + breakdown Reject + breakdown Request Request Request + breakdown Request

5 6 7 8 9 10 11 12

13 14 15 16

17 18

19 20 21 22 23 24 25 26 27 28 29 30

Data Collection and Coding During structured assessment sessions, the assessor collected data on the type of trial conducted (e.g., request or reject opportunity with or without breakdown) and the child's behavior (e.g., request, reject, or repair). This process helped the assessor to monitor the type of trials to be conducted by following the scripted protocol. All of the sessions were videotaped (with parents' permission). The assessor and an additional person watched the video clips and developed behavioral definitions (see descriptions of Ben's communication topographies in Tables 4 and 5). Following the development of behavioral definitions, Ben's communication topographies were recorded using the coding sheet in Figure 4. The results of the structured protocol are summarized in Ben's communication profile (see Figure 5).

Outcomes and Interpretationsof Ben's Assessment Results Among the many potential issues that could be gleaned from Ben's results, five are briefly highlighted here as prompts for future efforts to enhance early communication of children with autism and developmental disabilities. The data appear to validate the preference assessment process in that the percentage of requests expressed on requesting trials (94%) and on rejecting trials (68%) was well above chance performance. Relative- to requesting, Ben used few topographies In his rejecting response class. For example, give back occurred on more than 60% of the rejection trials, with pointing the next most frequent at 13%. This is noteworthy because of the potential implications for the development of problem behavior. That is, if a child has only one primary, socially adaptive rejecting topography In his or her response class and it is not honored, the probability of problem topographies may increase. Recently, many researchers and practitioners have recognized the fundamental role of social partner responsiveness in determining the communicative topographies used by young children with autism and other developmental disabilities (Halle et al., 2004; Warren & Yoder, 1998; Wilcox, 1992). Social partner responsiveness, or lack thereof, has been implicated in the etiology of problem behavior as well as the enhancement of socially adaptive, communicative topographies. Frequent and substantial problem behavior when engaging in everyday routines was reported by Ben's teachers and parents and clearly was a significant concern for them. It probably was not accidental or serendipitous that problem behavior was almost entirely absent from the assessment sessions. From our perspective, we arranged an ideal context: Highly preferred materials and activities were introduced by a fully engaged and responsive adult who had intentionally developed rapport with Ben and who on the vast majority of trials honored immediately Ben's first topography, which minimized the likelihood of problem behavior. Parents and teachers could capitalize on this ideal programming by arranging similar conditions at home and at school in an effort to eliminate coercive communicative topographies. A third noteworthy issue gleaned from Ben's results pertains to the ambiguity of his signaling. Some members of Ben's requesting response class were not specific and few were symbolic, leading to difficulty in listener comprehension and probable breakdowns. Examples of topographiei that were imprecise are pointing in the direction of toys and materials without identifying any specific item or vocalizing and signing "more" without specifying what more he wanted. Fourth, the structured protocol designed for assessment could easily be adapted for intervention; it provides many opportunities for teaching new topographies in an efficient manner (Meadan et

Communication of Young Children With Autism

57

TABLE 4. Ben's Behavioral Deflnltions-Requestlng Topographies of request

Description

Example and nonexample

Vocalize

In the presence of an adult, Ben emits a vocalization that does not approximate a word. (The vocalization is in a context that makes clear its requesting function.)

Example: produces the sounds while looking at the adult or an object Nonexample: laughs, giggles, or makes sounds without referencing the adult or an object

Point

Ben touches an object with any one or two fingers (i.e., contact pointing) or extends a finger toward an object (i.e., distal pointing),

Example: touches a toy with one finger; extends index finger toward a toy in the bag; touches the place of a missing piece in a puzzle; puts finger on table in reference to an object Nonexample: extends one finger with no reference to an activity or an object

Reach

In the presence of an adult, Ben moves one or both hands toward an object but does not make contact with the object. Arm must be extended in the air with no part of the arm resting on the table.

Example: extends arm toward a toy (arm is out, but no finger is extended) Nonexample: waves good-bye, claps (Grabbing an object will not be recorded as a request)

Sign "more"

Ben taps hands together at midline (one or multiple times) while looking at object or communication partner.

Example: claps hands together in reference to an object Nonexample: Waves, hits the table, or claps hands in excitement with no reference to an object

Lead

Ben takes an adult by the hand or wrist and guides the adult's hand or wrist to an object.

Example: holds adult's hand and leads hand to the tape recorder's On button Nonexample: pushes the adult's hand away (rejecting)

Verbalize

Ben produces an intelligible word or a meaningful sound that approximates a word.

Example: says "baw" for ball and "more" for more Nonexample: unintelligible word or a cry

Tap the table

Ben hits (audible hit) with one or both hands, once or twice, with an open palm or using the wrist or entire fist on the table in reference to an object.

Example: hits the table while looking at object Nonexample: taps the table without the presence of an item

Head nod

In the presence of an object and in the context of a request, Ben shakes head up and down while making eye contact with the adult.

Example: when presented a toy, nods head up and down Nonexample: nods head without the presence of an item

Note. Requests were defined as in the context provided, such as presenting an item or playing with the item, the child responded with one or more of the topographies described.

TABLE 5. Ben's Behavioral Definition-Rejecting Topographies of rejecting

Description

Example and nonexample

Give back

Holds an object or an activity and hands it to the partner or puts it on the table in front of the partner.

Example: extends arm out toward partner with the object in his or her hand Nonexample: pushes toy away or puts the toy on the ground

Shake head no

Shakes his head from left to right only in the context of rejecting, when unpreferred item is presented.

Example: shakes head "no" in reference to a car Nonexample: shakes head "no" between activities when an item is not presented

Place item aside

Picks up object and drops it to the side or off the surface in front of him.

Example: puts toy next to him on the ground off the table Nonexample: hits the toy on the mirror behind him

Note. Rejects were defined as inthe context provided, such as presenting an item or playing with the item, the child responded with one or more of the topographies described.

58

r

-

0

0

0

0

0

0

O

0

o

0

z

z

z

z

z

z

zU

z

z

•'.2

tU

U9 C.

r

T

S.

H

o, 0, *13

z9

oN .g

H9

H9

S.

-4 0)

IxI

H

z

S.

S.

'(1)

>.

H

H

H

22 0.

,-4

S.

-4

I4

0. 02

22o-.2 o 0 o

0

V

Z

>.. 0

0

-

-

z

-z

X

Z

z

z

-

-.

.

z

z

z

z

z

z

z

0

0

0

0

0

0

0

z

Z

z

z

z

z

o .. 0 2I U.

-

-

S..5 ,,oIz e4-

06

0 II >2 c 0

.2>

U, U,

U..5

So

-.

o

o

.-.

o

.o

0.U

-

o

U.

U.

S.

S.

6

.

ur.

cit R-A it2 '2 >u

"

U.

0 .

S.

.

r-L S.

S

.

S

S

.

S

.-...

S

a) 0

r-

-

.

CDS

-

.

w.

0

o

0

0 c

-

S

H o

S

.

.

0

7

o

a.

U.

-

0

.

.

-

.

S

, 0o

S

a.U

U

•. 0o

.

S

.

S

H

o

C.

.7

.

S

.

RE o

CL

0.

r-, r_0

r_-o

0)

U

t.

r-

L)

-

.

•7 0

-'

0 Nu00

S

.

Ei =.K6-6

E... 'o r. =

0

o

P4S. o

m, 1 UO

0c

l1 0 -- u .

a

S

.

S

E.

S

.

.

u

. U0

.:

CD

E

2

> o

0 0

a) 0a

0.

u

E

It

itJ ,-0-O

o>

0.

L

~r.

U) a)

4-'n

0) CD

_c 0 0)

u 0

00.

.C~

-6 N

-4 I

I

L

I

.1.

u

0

1-

tn

m

a

00 .5-

!!Z

u

E

0 cn LU

cz

59

Communication of Young Children With Autism

al., 2006). A fifth and final issue pertinent to Ben's resuits was the development of a CD with video clips showing the topographies Ben used in context and identifying his current repertoire of communicative topographies and their functions. The CD was shared with teachers, therapists, caregivers, and other new adults who worked with Ben. The goal of the CD was twofold: (a) to familiarize caregivers with Ben's repertoire of requesting and rejecting by showing actual examples in context and (b) to enhance caregivers' responsiveness to those topographies that are more socially adaptive. IMPLICATIONS FOR RESEARCH AND PRACTICE The structured assessment protocol described in this article has multiple potential implications for research and practice. A few that seem most compelling are delineated here. We have endeavored to cite other work that provides greater details about each implication. First, by generating current repertoires or response classes of requesting and rejecting, this protocol permits the identification of those extant topographies that are effective, socially adaptive, and understood by a wide audience of listeners (if any exist). Such topographies should become the target of responsiveness interventions to ensure their maintained use (i.e., they should be reinforced and become the most efficient member of the response class; Reichle, Halle, & Drasgow, 1998). The sessions could be brief and used initially to assess communicative forms and functions. If implemented consistently and intermittently over time, they could provide data on progress (e.g., the appearance of new forms that are more conventional or easily comprehended). Practitioners can draw on those parts of the protocol they find more relevant or compelling for children with whom they work, without replicating the more comprehensive format that is described above. A second implication pertains to those cases for which no effective, socially adaptive, and broadly understood topographies currently exist. In such cases, we cannot capitalize on an existing topography and therefore must teach a new one that meets these three criteria (Halle et al., 2004). A third implication takes us to the next level of intervention. Assume that a child possesses a topography (either already exists or is newly taught) that meets the three criteria. This topography will not be effective or successful in all contexts, so building tolerance (Carr et al., 1994) or resilience (Halle et al., 2004) in the face of failure or breakdowns must become a consideration. Halle et al. (2004) described an intervention that encompasses the teaching of multiple adaptive topographies to build resilience such that when the child encounters breakdowns in an interaction, multiple repair topographies are

Functions of Communication Ben requests on 94% and rejects on 6% of the trials when items identified as preferred are presented. Ben requests on 32% and rejects on 68% of the trials when items identified as unpreferred are presented.

Topographies of Requesting Point (79%), head nod "yes" (43%), vocdlize (25%), "more" sign (21%), reach (11%), tap on table (10%); other topographies with frequencies lower than 10%: lead and word approximations (i.e., more, bubble, car,apple, duck, and ball).

Topographies of Rejecting Gives item back or places item aside (61%), points to something different (13%), verbalizes "no" (10%), vocalizes (10%); other topographies with frequencies lower than 10%: throw away, push away, cry, scream, pinch, bite, drop to the floor (e.g., when walking from the gym back to the classroom), and shakes head "no."

Repair Behavior Ben persisted in communicating on most trials when his original message was not understood (i.e., repaired on 98% of the breakdown trials). His repair topographies were similar to the topographies for requesting and rejecting. FIGURE 5. Ben's communication profile.

available. We know of no empirical demonstrations of this intervention with young, prelinguistic children, but we currently are conducting a study to examine the strategy. , A next-step, future implication that builds on this conceptualization of intervention is to teach conditional or discriminated use of response class members to build adaptability to differing contextual conditions. In lay terms, this means not only teaching children multiple topographies for expressing a particular communicative function (e.g., requesting) but also teaching them in a way that the repair topography is sensitive to or takes into account the context and the partner's perspective. This broaches the issue of perspective taking (BaronCohen, 1989; Bretherton, 1992) but does not require the level of inference often imputed by these authors. For example, assume that Ben and his father are in the kitchen; Ben is playing with blocks on the kitchen table and his father is preparing dinner while listening to the radio. Ben tires of block play and wants a snack

60

Topics In Early Childhood Special Education 27:1

60

TopiCs in Early Childhood Special Education 27:1

while waiting for dinner, so he vocalizes and moves his arm in the direction of his father, who is cutting vegetables and not looking toward Ben (nor can he hear his low-volume vocalization because of the radio program). A breakdown in communication occurs-Ben's request is not honored-and Ben is about to repair the breakdown. Let us assume Ben's response class of requests includes leading, vocalizing, reaching, and screaming. The concept of discriminated repairs would entail teaching Ben to attend to the contextual conditions of (a) his father not looking at him and (b) competing noise from the radio so he would issue a repair that takes into account these conditions (e.g., leading his father by the hand to a snack item). Of course, another discriminated option might be screaming loud enough to be heard over the radio; however, Ben's father has been taught about functional connmunication training (FCT; according to the literature, once the function of a behavior is determined, a functionally equivalent and widely understood topography can be identified and taught to replace the more ambiguous, nonspecific form). He thus knows that he should not respond to screaming but should prompt a functionally equivalent topography once the screaming has stopped for a short period of time. In summary, the structured assessment protocol described in this article focuses on early communicators who have acquired basic repertoires of requesting, rejecting, and repairing-those instances when their first signals do not produce the intended outcome and a second signal is attempted. The protocol permits the identification of the differing topographies used by a child, organized by functional classes (i.e., requesting, rejecting). Furthermore, it provides a means to assess the topographies to which the child will resort if the first effort fails. Finally, the effects of an intervention program could be evaluated, at least in part, by arranging communicative opportunities or probes that are embedded into children's everyday routines to assess progress or change over time in their topographies of requesting, rejecting, and repairing. Do they use the newly taught topographies? Do they resort to functionally equivalent, socially adaptive topographies when encountering a probe in which a breakdown is programmed? What this structured assessment protocol lacks in terms of breadth of coverage, it makes up for by sampling actual performances of critical child communicative behavior. AUTHORS' NOTES i. This study was supported, in part, by the U.S. Department of Education, Office of Special Education Programs Grants H324C020098 and H325D010009. Any opinions, findings, and conclusions or recommendations expressed in this publication are those of the authors and do not necessarily reflect the views of the U.S. Department of Education or those of the University of Illinois. 2. We thank the parents and children who participated in the study.

REFERENCES Bates, E., Camaloni, L., & Volterra, V. (1975). The acquisition of performatives prior to speech. Merrill-PalmerQuarterly, 21(3), 205-226. Baron-Cohen, S. (1989). Perceptual role taking and protodeclarative

pointing in autism. The British Journal of Developmental Psychology, 7, 113-127. Bretherton, I. (1992). Social referencing, intentional communication, and the interfacing of minds in infancy. In S. Feinman (Ed.), Social

referencing qnd the social constructionof reality in infancy (pp. 5777). New York: Plenum Press. Carr, E. G. (19118). Functional equivalence as a mechanism of response generalization. In R. Horner, G. Dunlap, & R. L. Koegel (Eds.),

Generalizatlonand maintenance: Life.style changes in applied settings (pp. 221-241). Baltimore: Brookes, Carr, E. G., Levin, L., McConnachie, G., Carlson, J. I., Kemp, D. C., &

Smith, C. E. (1994). Connnunication-basedintervention for problem behavior. B:lhtimore: Brookes. Casby, lM.W., & Cumpata, J. F. (1986). A protocol for the assessment of prelingui#tic intentional communication. Journalof Communi-

cation Disorders,19, 251-260. Cirrin, F. M., 84 Rowland, C. M. (1985). Communicative assessment of nonverbil youths with severe/profound mental retardation.

Mental Retardation,23(2), 52-62. Coggins, T. E., Olswang, L. B., & Guthrie, J. (1987). Assessing communicative Intents in young children: Low structured observations

or elicitatioi tasks? The Journal of Speech and HearingDisorders, 52, 44-49. Halle, J,, Brady, N. C., & Drasgow, E. (2004). Enhancing socially adaptive communicative repairs of beginning communicators with

disabilities. American Journalof Speech-Language Pathology, 13, 43-54. Halle, J. W., & Drasgow, E. (2003). Response classes: Don Baer's contribution to understanding their structure and function. In K. S. Budd & T. Stokes (Eds.), A small matter of proof: The legacy of Donald M. lBaer (pp. 113-124). Reno, NV: Context Press. Hagopian, L P Long, E. S., & Rush, K. S. (2004). Preference assessment procedures for individuals with developmental disabilities.

BehaviorModification, 28(5), 668-677. Homer, R. H., & Day, H. M. (1991). The effects of response efficiency on functionailly equivalent competing behaviors. JournalofApplied

Behavior Analysis, 24, 719-732. lacono, T., Carter, M., & Hook, J. (1998). Identification of intentional communication in students with severe and multiple disabilities.

Augnmentatipe and Alternative Communication, 14, 102-114. Keen, D, (2003). Communication repair strategies and problem behavfours of children with autism. InternationalJournal of Disability

Development and Education, S0(1), 53-64. Lichtert, G. (2003). Assessing intentional communication in deaf tod-

dlers. Journml of Deaf Studies and Deaf Education, 8(1), 43-56. Mace, F.C., & Roberts, M. L. (1993). Factors affecting selection of behavioral interventions. In J. Reichle & D. P. Wacker (Eds.),

Connnunicationalternatives to challenging behavior: Integrating functional assessment and intervention strategies (pp. 113-133). Baltimore: Brookes. McLaughlin, D, M., & Carr, E. G. (2005). Quality of rapport as a setting event for problem behavior: Assessment and intervention.

Journalof Positive BehavioralInterventions, 7, 68-91. Meadan, H., Halle, J. W., Watkins, R. V., & Chadsey, J. G. (2006). Examining communication repairs of two young children with autism spectrum disorder: The influence of the environment. American Journalof Speech-Language Pathology, 15(1), 57-71. Neef, N. A., Mace, F. C., & Shade, D. (1993). Impulsivity in students with severe enmotional disturbance: The interactive effects of reinforcer rate, delay, and quality. Journalof Applied Behavior Analysis, 26, 37-52.

Communication of Young Children With Autism

61

Communication of Young Children With Autism

61

Neef, N. A., Mace, E C., Shea, M. C., & Shade, D. (1992). Effects of reinforcer rate and reinforcer quality on time allocation: Extensions of matching theory to educational settings. Journal of Applied BehaviorAnalysis, 25, 691-699. Ogletree, B. T., Fischer, M. A., & Turowski, M. (1996). Assessment targets and protocols for nonsymbolic communicators with profound disabilities. Focus on Autism and Other Developmental Disabilities,11 (1), 53-58. Ogletree, B. T., Pierce, K., Ham, W. E., & Fischer, M. A. (2002). Assessment of communication and language in classical autism: Issues and practices. Assessment for Effective Intervention, 21, 61-71. Olgetree, B. T., Wetherby, A. M., & Westling, D. L. (1992). Profile of the prelinguistic intentional communicative behaviors of children with profound mental retardation. American Journal of Mental Retardation,97(2), 186-196. Reichle, J., Halle, J. W., & Drasgow, E. (1998). Implementing augmentative communication systems. INA. M. Wetherby, S. E Warren, & J. Reichle (Eds.), Transitionsin prelinguisticcommunication (pp. 417436). Baltimore: Brookes. Warren, S. R, & Yoder, P. J. (1998). Facilitating the transition from prelinguistic' to linguistic communication. In A. M. Wetherby, S. E

Warren, & J. Reichle (Eds.), Transitionin prelinguisticcommunication (pp. 365-384). Baltimore: Brookes. Wetherby, A. M., Alexander, D. G., & Prizant, B. M. (1998). The ontogeny and role of repair strategies. In A. M. Wetherby, S. F. Warren, & J. Reichle (Eds.), Transitions in prelinguistic communication (pp. 135-159). Baltimore: Brookes. Wetherby, A. M., & Prizant, B. M. (1993). Communication and symbolic behaviorscale manual. Chicago: Applied Symbolix. Wetherby, A. M., Prizant, B. M., & Schuler, A. L. (2000). Understanding the nature of communication and language impairments. In A. M. Wetherby & B. M. Prizant (Eds.), Autism spectrum disorders: A transactionaldevelopmental perspective (pp. 109-141). Baltimore: Brookes. Wetherby, A. M., Warren, S. E, & Reichle, J. (1998). Introduction to transitions in prelinguistic communication. In A. M. Wetherby, S. F. Warren, & J. Reichle (Eds.), Transitions in prelinguisticcommunication (pp. 1-11). Baltimore: Brookes. Wilcox, J. M. (1992). Enhancing initial communication skills in young children with developmental disabilities through partner programming. Seminars in Speech and Language, 13(3), 194-212.

Test of Early Mathematics Ability-Third Edition, HerbertP. Ginsburgand Arthur J. Baroody The TEMA-3 measures the mathematics performance of Several important improvechildren between the ages of 3-0 and 8-11 and is also useful ments were made in the TEMA-3. with older children who have learning problems in mathe- a) A linear equating procedure is TE M matics. Used as a norm-referenced measure or a diagnostic tests so they may be used interinstrument, the test can be used to measure progress, evaluAbI1 ate programs, screen for readiness, discover the bases for changeably. b) Bias studies show 1.1t,, of $ugly ... ..... poor school performance in mathematics, identify gifted the absence of bias based on genstudents, and guide instruction and remediation. The test der and ethnicity. c) The pictures measures informal and formal (school-taught) concepts and of animals and money in the skills in the following domains: numbering skills, number- Picture Book are now in color to make them more appealing comparison facility, numeral literacy, mastery of number and more realistic in appearance. facts, calculation skills, and understanding of concepts. It has COMPLETE TEMA-3 KIT INCLUDES: Examiner's Mantwo parallel forms, each containing 72 items. The characteristics of the sample of 1,219 children ap- ual, Picture Book Form A, Picture Book Form B,25 Examproximate those in the 2001 U.S. Census. Internal consis- iner Record Booklets Form A, 25 Examiner Record Booklets tency reliabilities are all above .92; immediate and delayed Form B, 25 Worksheets Form A, 25 Worksheets Form B, Assessment Probes, 5"x 8"cards, 20 blocks, 20 tokens, and a alternative form reliabilities are in the .80s and .90s. Also provided is a book of remedial techniques (Assess- mesh bag, all in a sturdy storage box. (2003) ment Probes and Instructional Activities) for improving skills $265.00 08 TEMA-3 Complete Kit in the areas assessed by the test.

I#10880

COPYRIGHT INFORMATION

TITLE: A Protocol for Assessing Early Communication of Young Children With Autism and Other Developmental Disabilities SOURCE: Top Early Child Spec Educ 27 no1 Spr 2007 The magazine publisher is the copyright holder of this article and it is reproduced with permission. Further reproduction of this article in violation of the copyright is prohibited. To contact the publisher: http://www.proedinc.com/