Printer Friendly

Embedding Tact Instruction During Play for Preschool Children with Autism Spectrum Disorder.


Children with autism spectrum disorder (ASD) often demonstrate deficits in spontaneous language during play and other social interactions. The purpose of this study was to evaluate the efficacy of an instructional procedure that could be used during play to increase tacts by three 4-year-old children with ASD. A multiple-probe across behaviors design was used to evaluate the effects of the play-based intervention on participants' acquisition of tacts. The participants demonstrated rapid acquisition of targeted tacts during the training procedure and two participants maintained responses during a two-week follow-up probe. The results demonstrate that the play-based intervention procedure can be effective for teaching verbal behavior to children with ASD.

Keywords: autism spectrum disorder, language, naturalistic developmental behavioral intervention, tact training


Skinner (1957) outlined a functional taxonomy of language wherein verbal operants were defined based on the effect they had on social partners. The tact is one of the most important verbal operants and, in lay terms, refers to labeling objects, activities, events, or properties of those stimuli (Skinner). The tact is occasioned by observing a stimulus with one of the sense modes (e.g., auditory and visual) and maintained by various forms of social attention (e.g., praise, acknowledgment, ongoing interaction). For example, a child at the zoo in the presence of a zebra might say "Zebra!" and her father might say, "Yes sweetie, that's a zebra."

When children engage in early play interactions with adults, they frequently tact to share information with others and learn about the environment (Bates, Bretherton, Beeghly-Smith, & McKnew, 1982). Children tact when playing, going for a walk, reading a book, or when something out of the ordinary occurs (e.g., a bird flies into the classroom). The natural maintaining consequences are social praise from an adult or peer (e.g., "Yeah, that is a bird") or nonverbal attention (e.g., looking at the object, smiling, eye contact). However, the developmental trajectory of language in young children with autism spectrum disorder (ASD) shows marked differences compared to that of typically developing children, especially within social-communicative contexts such as play (Charlop & Haymes, 1994; Kim, Junker, & Lord, 2014; Wing & Wing, 1971). Tact repertoires may therefore be deficient among children with ASD, particularly within the context of play or other natural social interactions (Paul, 2008).

Teaching tacts requires a child to attend to a specific stimulus in the environment, which may be difficult for children with ASD (Sundberg, Endicott, & Eigenheer, 2000). Additionally, the consequences for tacting are conditioned reinforcers such as social praise, which may be ineffective for some children with ASD (LeBlanc, Esch, Sidener, & Firth, 2006). Children with ASD therefore require explicit tact training, which is often conducted using a discrete trial approach (Reichow, Barton, Boyd, & Hume, 2012), where antecedents and consequences are carefully planned before teaching.

When using a discrete trial arrangement to teach tacts to children with ASD, an interventionist sits across from the child and holds up an object or picture that the child is prompted to tact (Marchese, Carr, LeBlanc, Rosati, & Conroy, 2012). For example, in Marchese et al. interventionists sat across a desk from participants and held an item in front of the participant. During the intervention, researchers either asked participants "What is it?" or simply held the item in the participant's line of sight. Researchers delivered a vocal model prompt on a 3-s delay if children did not independently tact the items. All participants acquired novel tacts.

Although discrete trial tact training procedures have extensive support for teaching children with ASD to tact items (Sautter & LeBlanc, 2006), effective language interventions that specifically address spontaneous speech during natural social interactions are needed (Chiang & Carter, 2008). One reason for the absence of generalized tact repertoires among children with ASD may be that the highly structured approach used during tact training separates the child from the naturally occurring social variables that occasion and maintain language (Reichow et al., 2012). Procedures that combine natural social interaction with structured tact training opportunities might address deficits in spontaneous language among children with ASD (Geiger, Carr, LeBlanc, Hanney, Polick, & Heinicke, 2012; Schreibman et al., 2015).

Teaching children in the natural context and using natural consequences emerged as an essential component of comprehensive early intervention programs that aimed to address generalization of skills taught in highly controlled environments (Hart & Risley, 1974, 1975, 1980). Although natural approaches to teaching language may vary, they tend to prioritize natural contexts, focus on the child's immediate interests, capitalize on naturally occurring contingencies, and involve shared control between therapist and child (Koegel, O'Dell, & Koegel, 1987). Schreibman and colleagues (2015) recently used the phrase naturalistic developmental behavioral interventions (NDBI) to describe a collection of procedures that combine a behavior analytic and developmental approach to autism intervention. The focus of NDBIs is to embed opportunities for learning in socially relevant contexts throughout the child's daily routine (Lane, Lieberman-Betz, & Gast, 2016). These interventions are hypothesized to increase generalization and spontaneity due to a child led emphasis, and natural adult-child interactions that are embedded in daily routines (Carr & Kologinsky 1983; Kaiser, Hancock, & Nietflied, 2000; Schreibman et al., 1991). These approaches are aimed to appeal to caregivers and practitioners by embedding learning opportunities to socially relevant contexts (Delprato, 2001; Lane, Lieberman-Betz & Gast, 2016). There is an extensive body of research demonstrating the efficacy of NDBIs to teach children with ASD to request, or mand for, preferred items or activities including incidental teaching (Neef, Walters, & Egel, 1984), natural language paradigm (Koegel et al., 1992), pivotal response training (Schreibman et al., 1991) and embedded discrete trial teaching (Geiger et al., 2012).

However, among the naturalistic language intervention literature there is a paucity of research examining procedures that lead to tacting in play-based contexts (Snyder, Rakap, Hemmeter, McLaughlin, Sandall, & McLean, 2015). Tacts can be difficult to teach under naturalistic teaching arrangements because the naturally occurring consequence for tacting is attention or praise, which may not function as a reinforcer for many children with ASD (LeBlanc et al., 2006) and it may be more difficult for practitioners to contrive naturally occurring opportunities for tacting. NDBIs for tacts therefore require additional considerations to contrive motivation for the child to emit verbal behavior and establish durable relations between the tact and its corresponding stimuli.

Embedding discrete trials into natural contexts (e.g., play routine) offers a systematic approach to providing repeated opportunities for language acquisition while maintaining the desirable features of NDBIs (e.g., child-led, occurring in natural contexts, and socially relevant). Furthermore, studies have demonstrated that an embedded discrete trial approach can be equally as effective and efficient to traditional DTT approaches (Geiger et al., 2012).

One example of an embedded discrete trial approach is reciprocal imitation training, which teaches children with ASD gestural and object imitation and has shown to generalize across novel settings, materials, and therapists within a play routine (Ingersoll, 2012; Ingersoll, Lewis, & Kroman, 2007). The reciprocal imitation training procedure calls for following the child's lead by imitating the child's verbal and nonverbal behavior and describing the child's actions using simplified language. In addition, the environmental arrangement is specifically designed with multiple sets of matching items that are unique and hypothesized to be of interest to the child. Such an arrangement creates several opportunities for imitation with objects while considering play routines that a child might perform by himself or with peers in the future. Teaching occurs within play-based sessions when the interventionist interacts with one of the items and delivers a verbal statement for the child to imitate (i.e., learning trial). These statements are varied each time to promote flexibility and capitalize on the child's current attention to a particular toy. Instead of immediately embedding another explicit learning opportunity, as occurs during a highly structured approach to imitation training, the interventionist returns to play and follows the child's lead.

Similar to imitation, the maintaining consequence for tacting is attention and praise. As such, it may be helpful to consider the environmental arrangements created during reciprocal imitation training, such as following the child's lead, embedding trials within a play routine, embedding several preferred stimuli, and naturally occurring consequences (e.g., social praise) when teaching tacts. However, learning to tact requires additional instructional elements because a specific response topography is required for a specific stimulus, and children may need multiple learning opportunities for each response topography in order to acquire a tact across a range of environmental arrangements. Despite a growing body of tact training literature, there is minimal research to guide tact training within a natural environment. Therefore, the purpose of the present investigation was to test the efficacy of an embedded discrete trial training intervention, based in part on reciprocal imitation training, for teaching tacts to children with ASD during a play routine.



Three children attending an Early Intensive Behavior Intervention (EIBI) program housed within a child development laboratory preschool affiliated with a midwestern university participated. The participants were admitted to the program based on an outside ASD diagnosis from a licensed psychologist. Children spent 5 hr each day in an intensive applied behavior analysis therapy room and 3 hr in an inclusive preschool classroom with typically developing four-year-old children. Participants were selected for this study based on the following criteria: (a) they had received discrete trial tact training but did not demonstrate generalization of mastered tacts to play or social contexts, (b) they scored a minimum of 25 on the Early Echoic Skills Assessment (EESA; Esch, 2008), (c) they scored at least a five on the tact subset of the Verbal Behavior Milestones Assessment and Placement Program (VB-MAPP; Sundberg, 2008), and (d) they demonstrated that they could tact a minimum of 10 items when presented with 10 common objects without prompts or reinforcement. The Preschool Language Scale, 5th Ed. (Zimmerman, Steiner, & Pond, 2011) and the VB-MAPP were administered to determine the children's language skills prior to the study.

At the start of the study, Mayra was 4 years 6 months old. According to the VB-MAPP she was observed to tact up to 25 items when asked "What is this?" but did not demonstrate generalization across pictures, different

objects, or settings. She was able to request up to 10 items without a vocal prompt and was able to perform up to five motor actions on command. Her standard score on the PLS-5 was 53 for expressive language, 50 for auditory comprehension, and 50 total, with an age equivalence of 17 months.

Earl was 4 years 8 months at the beginning of the study. According to his VB-MAPP score, he was observed to tact up to 50 items when asked "What is this?" across three different pictures but did not generalize across settings. He could also tact up to 10 different actions (e.g., jumping, sleeping). He could request up to 10 items without a vocal model and could perform up to 10 motor actions on command. In addition, he could perform up to 25 two-component noun-verb and verb-noun instructions (e.g., if instructor said "feeding baby" he would take a spoon and feed a baby doll). His standard scores on the PLS-5 were 59 for expressive language, 50 for auditory comprehension, and 51 total, with an age equivalence of 17 months.

Luke was 4 years 5 months old. According to the VB-MAPP he was observed to tact up to 25 items when asked "What is this?" but did not generalize across pictures, different objects, or settings (e.g., early childhood classroom). He could request up to 10 items without a vocal prompt and performed up to 10 motor actions when requested. His standard scores on the PLS-5 were 63 for expressive language, 50 for auditory comprehension, and 53 total, with an age equivalence of 14 months.

Materials and Setting

Stimuli were selected for training based on being: (a) three-dimensional, (b) one--to three-syllable tact, and (c) part of a play routine that was reported or observed as a potential preference for each participant. In order to teach tacts, playsets were created around a thematic play scenario (e.g., tea party), with each set consisting of three target stimuli and an additional 10 toys for a total of 13 toys in each set. Table 1 depicts target stimuli within each thematic set. Items from the playsets were placed inside a large toy-box, small plastic boxes, and various sized plastic eggs, all of which were used during experimental sessions. Paper and pencil were used to collect data and a timer application was downloaded to the interventionist's cell phone for use during experimental sessions. A video camera was used to record 50% of the sessions.

The participants' training and baseline sessions were conducted in a small room adjacent to the EIBI classroom. The room was mostly carpeted, contained two large windows, a booth with a one-way mirror and a large toy box containing toys and current target stimuli. The first author conducted all sessions and an observer (undergraduate assistant) was present.

Measurement of Dependent Variable

The dependent variable was the percentage of correct independent tacts, defined as the participant emitting the correct name of the stimulus within 5 s of the interventionist presenting the stimulus. Tacts were scored as correct if the participant emitted the full name of the object, "it is" and the object name, or the object name with minor deviations in articulation (e.g., "detergen" instead of "detergent"). If the participant emitted any vocalization that was not the name of the object, did not respond vocally within the allotted time, or required a prompt to emit the name of the stimulus, it was not scored as an independent tact. Each of the three stimuli were targeted three times within a session, for a total of nine trials. Table 1 outlines playsets and tact target stimuli for each participant.

The first author served as the primary observer and data were collected from video recordings of sessions or recorded during the sessions. Data are presented as the percentage of correct independent tacts out of a total of nine trials in every session.

Interobserver Agreement

A second observer independently scored 33% of randomly selected sessions evenly distributed across all participants, conditions, and sets to establish interobserver agreement (IOA). The second observer was trained on measurement of dependent variables using videos of intervention sessions that were not used for IOA purposes. The second observer was required to demonstrate 90% reliability with the first author on training videos prior to collecting interobserver agreement data.

Point by point agreement (Cooper, Heron, & Heward, 2007) was used to determine IOA from video. An agreement was scored when both observers scored a response as a correct independent tact. A disagreement was scored when observers differed in how they scored a specific trial. IOA was calculated by dividing the number of agreements by the number of agreements plus disagreements and multiplying by 100. The mean agreement for baseline across all sets for Mayra was 98% (range: 94-100%), for Earl was 100%, and for Luke was 90% (range: 88-93%). The mean agreement for intervention sessions across all sets for Mayra was 97% (range: 90-97%), for Earl was 96% (range: 88-100%), and for Luke was 95% (range: 85-100%).

Experimental Design

A multiple probe across behaviors design (Gast & Ledford, 2014) was used to evaluate the effectiveness of the play-based procedure on acquisition of tact targets. The experiment was then replicated across participants. The multiple probe design is ideal when behaviors are not in a participants' repertoire as it limits the number of baseline sessions with zero-rate responding while still allowing for demonstration of a stable baseline (Gast & Ledford, 2014). The first author administered baseline sessions concurrently for all target responses (i.e., organized into three sets of three responses each) until stable, then started intervention for the first set of tacts, while remaining sets were periodically probed under baseline conditions. Once a participant demonstrated three consecutive sessions with an increase in level of responding and responding remained stable in the other two sets, the first author administered intervention for a second set of tacts. This process was repeated for the third set of tacts. Mastery criteria during intervention was set at one session at 100%.


Baseline probes involved a play routine consisting of three distinct subroutines. The play routine was centered on a thematic set in order to create an environment similar to those observed when children play in early childhood environments. The purpose of the subroutines (i.e., taking toys out, playing with toys, and putting toys away) was to contrive sufficient opportunities for participants to tact each of three target stimuli (i.e., three times per session per target stimuli). See Figure 1 for a flowchart depiction of baseline and intervention procedures.

Taking Toys Out. During the first subroutine, the interventionist previously hid target stimuli within different plastic toy eggs, pulled the eggs out of a large toy box one by one and handed each egg to the participant to open. The purpose of hiding target stimuli was to contrive a surprise scenario and increase the likelihood of a participant tacting an item upon opening the egg and seeing the item (e.g., "beetle"). Upon presenting the first egg, the interventionist waited for the participant to open the egg, then provided a prompt to do so if the participant did not independently open the egg. The discriminative stimulus was the appearance of the object in the participant's line of sight. The interventionist waited 5 s for the participant to look toward and tact the object. The procedure was repeated for all three target stimuli. The presentation of target stimuli was random--that is, target stimuli were rotated among three different containers each time (i.e., small, medium, and jumbo size opaque eggs).

Playing with Toys. During the second subroutine, the interventionist inconspicuously set target stimuli aside and started a 2 min timer. During the 2 min that followed, the interventionist played with the child, following his or her lead in playing with objects and imitated the child's behavior. In addition, the interventionist described what the child was doing during play, though ensuring that she never tacted target stimuli when describing the child's behavior. Every 30 s, the interventionist conducted a tact trial by placing a target stimulus that had previously been set aside into the existing play scene. For example, if the participant was spraying the table with the spray bottle, the interventionist introduced the sponge and pretended to wipe the table. The antecedent stimuli during these trials was the presentation of the object and the interventionist, while holding the object, asking a question such as "What is it?" or "What do you call it?" The interventionist then waited up to 5 s for the participant to respond. The interventionist repeated this procedure for each of the three target stimuli, though varied the question she asked across trials, thereby contriving one trial per target stimulus during the second subroutine.

Cleaning Up Toys. During the third subroutine, the interventionist provided a verbal directive to clean up toys, such as, "Ok, let's clean up our toys." The interventionist then modeled cleaning up with non-target stimuli, putting toys back in the box while tacting them. For example, the experimenter said "Bye, Mop" while placing the mop in the box. A trial during the third subroutine consisted of the interventionist placing the item in the participant's hand or allowing the participant to pick up the item and waiting 5 s for a response while blocking the participant from placing the target stimulus in the box. No vocal model prompts nor specific consequences for a correct or incorrect response were provided during these probes.


The interventionist administered procedures as described in the baseline probes and also delivered a vocal model prompt following a 5-s constant time delay, delivered differential reinforcement following correct responses, and provided an error correction sequence following an incorrect response during intervention. If participants responded to the vocal model correctly during the 5-s delay, the interventionist provided social reinforcement in the form of praise and physical contact while interacting with the child. The target stimulus was always withheld from the participant following a correct response to increase the likelihood of training the response as a tact and not as a mand.

If the child did not respond within 5 s or emitted an incorrect tact, the interventionist scored the response as incorrect, then provided a vocal model and waited another 5 s for the participant to respond. Immediately after the participant echoed the model, the interventionist repeated the trial to transfer stimulus control from the presentation of the stimulus with a vocal model to presentation of the stimulus only (see Petursdottir, Carr, & Michael, 2005). The repeated trial involved the interventionist removing the target stimulus out of the participant's line of sight for 1 to 2 s and then placing the item back in the participant's line of sight and providing no vocal model. If the participant tacted the object during this repeated trial, the interventionist provided reinforcement that consisted of a praise statement, "Yes, it is a--," but withheld physical contact. The repeated trial was conducted a second time if participants failed to respond on the first occasion. If a participant had not responded to the second repeated trial, the interventionist planned to move on to the next target stimulus though this never occurred during the study. Repeated trials enabled the interventionist to quickly probe whether the participant was able to respond independent of a prompt. Repeated trials were not scored. Sessions were repeated until participants demonstrated mastery criteria of 100% once or 90% across two consecutive sessions.

Maintenance and Transfer of Training

Two weeks after a participant met mastery criteria, maintenance and transfer of training sessions were administered to Mayra and Earl. Luke had exited the clinic when these data were collected. Maintenance and transfer of training probes assessed whether participants could tact the target stimuli when presented under novel circumstances and after termination of daily instruction. Although similar to a generalization probe this condition could not experimentally assess generalization because it was not tested prior to intervention.

During maintenance and transfer of training probes, the interventionist placed all nine target stimuli in child-sized lockers in the hallway adjacent to the EIBI classroom. To mimic a similar surprise element to baseline and intervention, where objects were hidden in plastic eggs, the interventionist hid all nine target stimuli in child size adjacent lockers. To begin each trial, the interventionist brought the child to the hallway and said, "Let's open the locker" and waited 5 s for the child to respond. The interventionist did not prompt or deliver reinforcement for correct responding but simply acknowledged the response by saying, "Ok." This novel situation was contrived to test whether children would tact without being explicitly asked to tact. A session consisted of nine trials that were contrived during transitions to and from the EIBI therapy room.

Procedural Integrity

Procedural integrity was assessed during 33% of sessions, evenly distributed across baseline and intervention sessions, as well as across sets and participants. A procedural integrity checklist was created by the first author and included all implementation steps for each of the subroutines during baseline and intervention. Procedural integrity checklists for baseline and intervention were the same, except that intervention included vocal models and reinforcement. An independent observer was trained by the first author to code the checklist from videos of experimental sessions until the observer reached 90% reliability with the first author on videos for baseline and intervention. The training videos were not included in the evaluation of procedural integrity. The mean percentage of steps implemented with fidelity for baseline was 95% (range: 88-100%). The mean percentage of steps implemented with fidelity for intervention was 96% (range: 85-100%).

Social Validity

Early childhood educators unfamiliar with the purpose of the study were recruited from midwestern preschool programs to respond to a six-item online questionnaire. Educators taught in typical preschool settings that employed a center-based, student-directed approach to early childhood education. The questionnaire required respondents to view two types of videos (e.g., a video of a discrete trial tact teaching session or a video of the play-based tact training procedure) and to rate both interventions on six statements.

Using a seven-point Likert-type scale (strongly agree, agree, somewhat agree, neither agree nor disagree, somewhat disagree, disagree and strongly disagree), respondents answered the following six statements: (a) this procedure or type of teaching would fit well in my classroom, (b) the student appears to be experiencing discomfort with this teaching procedure, (c) this teaching procedure is likely to result in increased language for this student, (d) overall the student seems to be having a positive reaction to this teaching procedure, (e) this seems like a good use of this student's time, and (f) I am likely to use this teaching approach with my students who do not tact objects.

Two videos from a pool of eight videos, four from intervention sessions of the current study and four of discrete trial sessions teaching tacts to the same children, were randomly assigned to the questionnaire. The order of video presentation was kept constant. Respondents first saw a video of the play-based session followed by the discrete trial session. The videos of discrete trial teaching sessions depicted the participant seated across from the instructor at a child-sized table. Objects not included in the present experiment were presented at the table one at a time. The instructor gave the student a token contingent on correct tacts, and tokens were exchanged for edible or tangible reinforcement. The videos were equal in length and had been collected 6 months prior to the videos from the current study.


Figure 2 depicts the percentage of independent correct responses for Mayra during baseline, play-based intervention, and transfer of training probes. During baseline, Mayra did not tact target stimuli across Sets A, B, or C. Upon introduction of the play-based procedure, Mayra immediately demonstrated an increased level of independent tacts requiring five, five, and three sessions to reach mastery criteria across Sets A, B, and C, respectively. In all cases, Mayra showed a rapid increase in percentage of independent responding with the only overlap occurring during the first trial of Sets A and B when responses were prompted, and no overlap for Set C. Mean responding during intervention for Sets A, B, and C, respectively, was 51% (range: 0-100%), 49% (range: 0-100%) and 55% (range: 22-100%). During transfer of training, Mayra's mean responding was 100% across sets.

Figure 3 depicts the percentage of independent correct responses for Earl during baseline, play-based intervention, and transfer of training. During baseline, Earl did not tact target stimuli across Sets A, B, or C, which demonstrates that the response topographies were not in his repertoire. Upon introduction of the play-based procedure, Earl demonstrated an increased level of independent tacts requiring four, eight, and eight sessions to reach mastery criteria across Sets A, B, and C, respectively. In all cases, Earl showed a rapid increase in percentage of independent responding with the only overlap occurring during the first session of Set B, when responses were prompted, and no overlap for Sets A and B. Earl's mean responding during intervention for Sets A, B, and C respectively was 73% (range: 33-100%), 63% (range: 0-100%), and 48% (range: 22-100%). During transfer of training, Earl's mean responding was 100% across sets.

Figure 4 depicts the percentage of independent correct responses for Luke during baseline and the play-based intervention. Luke transitioned out of the EIBI program shortly after training responding of the last set of target stimuli, not providing sufficient time for transfer of training probes. During baseline, Luke did not tact target stimuli across Sets A, B, or C, which demonstrates the response topographies were not in his repertoire. Upon introduction of the play-based procedure, Luke demonstrated an increased level of independent tacts for each set, and required ten, nine, and ten sessions to reach mastery criteria across Sets A, B, and C, respectively. In all cases, Luke showed a rapid increase in percentage of independent responding with the only overlap occurring during the first one to three intervention sessions, when all responses were prompted. Luke's mean responding during intervention for Set A was 55% (range: 0-100%), for Set B was 48% (range: 0-100%) and for Set C was 55% (range: 22-100%).

Social Validity Survey Results

A total of nine early childhood educators completed the online survey. Respondents rated the extent to which they agreed or disagreed with statements after watching videos of intervention sessions. See Table 2 for a summary of means and ranges for all survey responses. Overall, early childhood educators rated the play-based procedure more favorably than the discrete trial procedure. Educators reported that the procedure would fit well in their classroom, was likely to result in increased language, seemed like a good use of the students' time, and that they were likely to use the play-based approach with students in their classroom.


All participants in the present investigation acquired each set of tacts relatively quickly, every time the intervention was applied. The two participants for whom we administered transfer of training probes both tacted objects in novel contexts. And early childhood providers rated the procedures as highly acceptable. The current study demonstrates that the instructional variables (e.g., prompts, rein-forcers, error correction) that teach children with ASD to tact during highly structured teaching arrangements can also be applied to teach children with ASD to tact during play-based arrangements. Additionally, multiple trials for each item can be incorporated into the play sequence, which is likely an important component for acquisition. The procedures and outcomes of the present investigation have several implications for research and practice pertaining to naturalistic interventions for children with ASD.

The primary aim of this study was to extend procedures similar to those used during embedded discrete trial (Geiger et al., 2012) to teach tacts. Preferred stimuli were incorporated into the play sequence for each participant and children could freely access those stimuli for most of the session. Additionally, the interventionist spent the majority of the session following the child's lead, with explicit trials delivered every 2 min during the second subroutine. This procedure aligns with the NDBIs described by Schriebman and colleagues (2015), wherein behavior analytic practices are embedded in an interactive play-based context that replicates the natural antecedents and consequences a child may experience outside the clinical setting. The participants in the present study reliably tacted objects during the play routine. Although many NDBIs that teach language target mands, the procedures in the present investigation demonstrate a way to apply the principles to teaching tacts.

Overall, the participants learned tacts relatively quickly. Mayra demonstrated a faster rate of acquisition over time; acquiring the first two tact sets in five sessions and the last set in three sessions. Luke acquired the first tact set in ten sessions and subsequent sets in nine sessions. These outcomes suggest there is likely to be heterogeneity in rate of responding and service providers may need to give some children with ASD more time to demonstrate independent responses than others. Interestingly, Earl acquired the first set of tacts in five sessions and subsequent sets in eight sessions. This is somewhat surprising as we anticipated the first set, when participants were introduced to the structure of sessions, would take the most time for participants to learn. Earl's pattern might speak to the benefit of including highly preferred items in the instructional session, as he appeared to be more motivated to play with stimuli in the first playset compared to the latter two.

A challenging aspect of teaching tacts to children with ASD using a naturalistic approach is that children often require several trials with the same target in a given session in order to reach mastery criteria in a reasonable period of time (e.g., Marchese et al., 2012). When preparing for the present investigation, we initially created a play sequence based on reciprocal imitation training (Ingersoll & Schreibman, 2006) whereby the experimenter presented objects while following the child's lead during play, but this would only allow for a single trial for each target stimulus. We therefore designed additional subroutines around getting toys out and putting them away in order to increase trials for individual target stimuli while maintaining a naturalistic approach. For the participants in this study, three trials were sufficient to promote rapid learning for each target stimulus.

Another potential advantage of using the three subroutines to vary context across trials is that participants learn a specific response is appropriate across a range of antecedent conditions, which may promote generalization of acquired targets for children with ASD (Stokes & Baer, 1977). Mayra and Earl demonstrated a high level of tacts during the transfer of training trials that followed intervention. Although responding to stimuli in lockers wasn't officially assessed during baseline, it is likely participants could not emit the target responses as we expect they would have done so during baseline sessions in the research room. It is more likely that participants learned to tact the items upon seeing them and generalized responding to a novel context. This further speaks to the potential of the play-based procedure for teaching language that is not dependent on extraneous stimuli, such as prompts or highly controlled teaching arrangements.

Many children with ASD may become dependent on vocal prompts used during instructional sessions, which is a known barrier to generalization (LeBlanc et al., 2006). And in the case of tacts, some children may be dependent on adults asking them "What is it?" prior to tacting an object (Marchese et al., 2012). Therefore, in the present study, we extended the procedure in Marchese and colleagues (2012) by varying the antecedents from presenting a question (e.g., "What is it?") and object, and the object only. It is possible that the routine-based sequence of each session helped bring responding under control of the presentation of the item and when asked to tact, thereby facilitating transfer of stimulus control. These findings are preliminary yet might speak to the benefits of varied trial types (e.g., taking items out of box versus contriving presentations during play) on the acquisition of tacts.

Another important outcome of the present investigation is the apparent social validity of the intervention. First, providers rated the intervention as highly acceptable, particularly when compared to a more structured approach to language training. This is an important empirical contribution as it confirms the likelihood that early childhood practitioners will use a naturalistic intervention and that naturalistic interventions have high social validity. Second, play-based instructional arrangements align with how services in early childhood settings are often delivered (Schreibman et al., 2015). Third, early childhood practitioners reported that the instructional procedure was easy to implement. Many early intervention providers require simple interventions, yet children with ASD often require a high number of learning opportunities with explicit presentation of antecedents and consequences (Reichow et al., 2012). The present procedures fulfill priorities of early intervention providers while also ensuring that children had frequent learning opportunities with consistent presentation of instructional arrangements.

Several limitations must be considered in light of the outcomes. First, participants' vocalizations were not always only under the control of a nonverbal stimulus, but may also have been controlled by an establishing operation, particularly during the first two subroutines, when items were hidden in eggs or in the possession of the experimenter. In such scenarios, it is possible the response was multiply controlled and functioned partially as a mand. However, if responses were initially under primary control of an establishing operation, we would expect responding to decrease over time as children had non-contingent access to the stimuli in the hidden eggs during the first subroutine and did not receive access following correct tacts to the items held by the experimenter during the second subroutine. The sustained performance across sessions suggests the nonverbal stimuli evoked responding. In addition, the aim of the procedure was not to teach pure tacts (i.e., controlled only by nonverbal stimuli) but in offering a naturalistic approach to tact training that is socially valid and may be utilized in early childhood inclusive settings.

Additionally, we evaluated the play-based procedure in an environment outside the children's typical environment (e.g., classroom or home, see Hepting & Goldstein, 1996 for a continuum of natural language interventions). Our rationale for using a separate room was to ensure participants did not observe one another's sessions, which could have interfered with experimental control in the multiple probe across participants design. Our results suggest the naturalistic intervention can be effective for teaching tacts, though an important next step for future research is to evaluate the effects of the play-based procedure in natural environments. Doing so could establish the effectiveness of the procedure, as well as its feasibility and acceptability in natural environments.

Also, during the playing with toys subroutine, participants were asked "What do you call this?" which is an antecedent condition that they may be less likely to encounter during play. Future investigations may contrive more natural antecedent conditions during play, such as asking "What's in the oven?" when pretending to bake. Future research may contrive less discriminable antecedent conditions to increase the likelihood that participants will generalize acquired responses during play with caregivers and peers.

In addition, participants were highly similar in terms of their preference for social attention and pre-intervention language levels. The present procedure is unlikely to be effective for children whose behavior is not reinforced by social attention. The extent to which very early language learners would acquire tacts at similar rates to the current participants cannot be known. Future research could address this limitation by replicating the procedures with emerging speakers. Doing so may first require that social consequences be paired with primary rein-forcers and systematically faded. Researchers could then test whether social praise or attention functions as reinforcement for tacts prior to using the play-based procedure (see LeBlanc, & Dillon, 2009 for this recommendation). Meanwhile, practitioners should carefully consider whether this intervention is appropriate for children who do not show a preference for social attention or who have very limited language.

Another limitation worth mentioning is the absence of generalization data prior to intervention. Although we believe it is unlikely that Mayra and Earl would have tacted target stimuli without instruction, we are not able to claim that the intervention was completely responsible for the emission of target responses under novel contexts. Doing so would require testing for tacts in all contexts prior to instruction. In addition, other tests for generalization (e.g., with other adults, with peers) would provide meaningful information about the generalizability of the play-based procedure.

One final limitation that requires future investigation is that although we varied the types of trials children were exposed to, we used the same stimuli from one trial to the next. That is, we did not incorporate multiple exemplars of each target stimulus within the play-based sequence. Multiple exemplar instruction is recommended as a part of natural environment training as it can enhance generalization (LeBlanc, 2006). It may be helpful to evaluate whether the varied trial types with static stimuli, as used in the present investigation, are sufficient to promote generalization or if children would benefit from being exposed to multiple examples of a stimulus during instructional arrangements.

The present investigation documents the efficacy of an embedded tact training procedure during a play routine for children with ASD. The results show that the same instructional mechanisms (e.g., prompts, reinforcement, error correction) used in prior tact training research (e.g., Marchese et al., 2012) can also be effective during play sequences that might be more similar to how early childhood educators interact with children. As such, the play-based procedure might have broader applications to early intervention personnel, who may not have expertise and training in behavioral interventions for individuals with ASD.

Ana D. Duenas

Joshua B. Plavnick

Courtney E. Maher

Michigan State University

Address correspondence to: Ana D. Duenas, 620 Farm Lane, #341, Michigan State University, East Lansing, MI 48824. E-mail:


Bates, E., Bretherton, I., Beeghly-Smith, M., & McKnew, S. (1982). Social bases of language development: A reassessment. In H. W. Reese & L. R Lipsitt (Eds.). Advances in child development and behavior (pp.7-15). New York, NY: Academic. doi:10.1016/s006 5-2407(08)60067-1

Carr, E. G., & Kologinsky, E. (1983). Acquisition of sign language by autistic children: Spontaneity and generalization effects. Journal of Applied Behavior Analysis, 16, 297-314. doi:10.1901/jaba.1983.16-297

Charlop, M. H., & Haymes, L. K. (1994). Speech and language acquisition and intervention: Behavioral approaches. In J. L. Matson (Ed.). Autism in children and adults: Etiology, assessment, and intervention (pp. 213-240). Pacific Grove, CA: Thomson Brooks.

Chiang, H., & Carter, M. (2008). Spontaneity of communication in individuals with autism. Journal of Autism and Developmental Disorders, 38,693-705. doi:10.1007/s10803-007-0436-7

Cooper, J. O., Heron, T. E., & Heward, W. L. (2007). Applied behavior analysis (2nd ed.). Upper Saddle River, NJ: Pearson/Merrill-Prentice Hall.

Delprato, D. J. (2001). Comparisons of discrete-trial and normalized behavioral language intervention for young children with autism. Journal of Autism and Developmental Disorders, 31, 315-325. doi:10.1023/A:1010747303957

Esch, B. E. (2008). Early echoic skills assessment. The verbal behavior milestones assessment and placement program. Concord, CA: AVB Press.

Gast, D. L., & Ledford, J. R. (2014). Single case research methodology: Applications in special education and behavioral sciences. New York, NY: Routledge.

Geiger, K. B., Carr, J. E., LeBlanc, L. A., Hanney, N. M., Polick, A. S., & Heinicke, M. R. (2012). Teaching receptive discriminations to children with autism: A comparison of traditional and embedded discrete trial teaching. Behavior Analysis in Practice, 5, 49-59. doi:10.1007/BF03391823

Hart, B., & Risley, T. R. (1980). In vivo language intervention: Unanticipated general effects. Journal of Applied Behavior Analysis, 13, 407-432. doi:10.1901/jaba.1980.13-407

Hart, B., & Risley, T. R. (1975). Incidental teaching of language in the preschool. Journal of Applied Behavior Analysis, 8, 411-420. doi:10.1901/jaba.1975.8-411

Hart, B., & Risley, T. R. (1974). Using preschool materials to modify the language of disadvantaged children. Journal of Applied Behavior Analysis, 7,243-256. doi:10.1901/jaba.1974.7-243

Hepting, N. H., & Goldstein, H. (1996). What's natural about naturalistic language intervention? Journal of Early Intervention, 20, 249-264. doi:10.1177/105381519602000308

Ingersoll, B. (2012). Effect of a focused imitation intervention on social functioning in children with autism. Journal of Autism and Developmental Disorders, 42, 1768-1773. doi:10.1007/ S10803-011-1423-6

Ingersoll, B., Lewis, E., & Kroman, E. (2007). Teaching the imitation and spontaneous use of descriptive gestures in young children with autism using a naturalistic behavioral intervention. Journal of Autism and Developmental Disorders, 37, 1446-1456. doi:0.1007/s10803-006-0221-z

Ingersoll, B., & Schreibman, L. (2006). Teaching reciprocal imitation skills to young children with autism using a naturalistic behavioral approach: Effects on language pretend play, and joint attention, journal of Autism and Developmental Disorders, 36, 487-505. doi:0.1007/s10803-006-0089-y

Kaiser, A. P., Hancock, T. B., & Nietfeld, J. P. (2000). The effects of parent-implemented Enhanced Milieu Teaching on the social communication of children who have autism. Early Education and Development, 22,423-446. doi:10.1207/s15566935eed1104_4

Kim, S. H., Junker, D., & Lord, C. (2014). Observation of Spontaneous Expressive Language (OSEL): A new measure for spontaneous and expressive language of children with autism spectrum disorders and other communication disorders. Journal of Autism and Developmental Disorders,44, 3230-3244. doi:10.1007/s10803-014-2180-0

Koegel, L. K., Carter, C. M., & Koegel, R. L. (2003). Teaching children with autism self-initiations as a pivotal response. Topics in Language Disorders, 23, 134-145. doi:10.1097/00011363-200304000-00006

Koegel, R. L., O'Dell, M. C., & Koegel, L. K. (1987). A natural language teaching paradigm for nonverbal autistic children. Journal of Autism and Developmental Disorders, 17, 187-200. doi:10.1007/BF01495055

Lane, J. D., Lieberman-Betz, R., & Gast, D. L. (2016). An analysis of naturalistic interventions for increasing spontaneous expressive language in children with autism spectrum disorder. The Journal of Special Education, 50, 49-61. doi:10.1177/0022466915614837

LeBlanc, L. A., Esch, J., Sidener, T. M., & Firth, A. M. (2006). Behavioral language interventions for children with autism: Comparing applied verbal behavior and naturalistic teaching approaches. The Analysis of Verbal Behavior, 22, 49-60. doi:10.1007/bf03393026

Marchese, N. V., Carr, J. E., LeBlanc, L. A., Rosati, T. C., & Conroy, S. A. (2012). The effects of the question "What is this?" on tact-training outcomes of children with autism. Journal of Applied Behavior Analysis, 45, 539-547. doi:10.1901/jaba.2012.45-539

Neef, N. A., Walters, J., & Egel, A. L. (1984). Establishing generative yes/no responses in developmentally disabled children. Journal of Applied Behavior Analysis, 17, 453-460. doi:10.1901/jaba.1984.17-453

Paul, R. (2008). Interventions to improve communication. Child and Adolescent Psychiatric Clinics of North America, 17, 835-856. doi:10.1016/j.chc.2008.06.011

Petursdottir, A. I., Carr, J. E., & Michael, J. (2005). Emergence of tacts and mands of novel objects among preschool children. The Analysis of Verbal Behavior, 21, 59-74. doi:10.1007/BF03393010

Reichow, B., Barton, E. E., Boyd, B. A., & Hume, K. (2012). Early intensive behavioral intervention (EIBI) for young children with autism spectrum disorders (ASD): A systematic review. The Cochrane Database of Systematic Reviews, 10, 1-63. doi:10.4073/csr.2014.9

Sautter, R., & LeBlanc, L. A. (2006). Empirical applications of Skinner's analysis of verbal behavior with humans. The Analysis of Verbal Behavior, 22, 35-48. doi:10.1007/bf03393025

Schreibman, L., Dawson, G., Stahmer, A. C., Landa, R., Rogers, S. J., McGee, G. G., ... Halladay, A. (2015). Naturalistic developmental behavioral interventions: Empirically validated treatments for autism spectrum disorder. Journal of Autism and Developmental Disorders, 45,2411-2428. doi:10.1007/s10803-015-2407-8

Schreibman, L., Kaneko, W. M., & Koegel, R. L. (1991). Positive affect of parents of autistic children: A comparison across two teaching techniques. Behavior Therapy, 22,479-490. doi:10.1016/S0005-7894(05)80340-5

Skinner, B. F. (1957). Verbal behavior. Englewood Cliffs, NJ: Prentice Hall.

Snyder, P. A., Rakap, S., Hemmeter, M. L., McLaughlin, T. W., Sandall, S., & McLean, M. E. (2015). Naturalistic instructional approaches in early learning: A systematic review. Journal of Early Intervention, 37, 69-97, doi:10.1177/1053815115595461

Sundberg, M. L. (2008) VB-MAPP Verbal Behavior Milestones Assessment and Placement Program: A language and social skills assessment program for children with autism or other developmental disabilities: guide. Concord, CA: AVB Press.

Sundberg, M. L., Endicott, K, & Eigenheer, P. (2000). Using intraverbal prompts to establish tacts for children with autism. The Analysis of Verbal Behavior, 17, 89-104. doi:10.1007/BF03392958

Wing, L., & Wing, J. K. (1971). Multiple impairments in early childhood autism. Journal of Autism and Childhood Schizophrenia, 1, 256-266. doi:10.1007/ BF01557347

Zimmerman, L., Steiner, V. G., & Pond, R. E. (2011). Preschool Language Scale-5. San Antonio, TX: Psychological Corporation.

Caption: Figure 1. Flowchart of baseline and intervention procedures.

Caption: Figure 2. Percentage of independent responding during baseline, intervention, and transfer of training sessions across sets for Mayra.

Caption: Figure 3. Percentage of independent responding during baseline, intervention, and transfer of training sessions across sets for Earl.

Caption: Figure 4. Percentage of independent responding during baseline and intervention sessions across sets for Luke.
Table 1
Targets Selected by Thematic Play Set

              Set A           Set B           Set C
Participant   Theme/stimuli   Theme/stimuli   Theme/stimuli

Mayra         Bug Hunt:       Tea Party:      Cleanup:
              beetle          toaster         detergent
              vest            tea             dustpan
              grasshopper     honey           squeegee
Earl          Bug Hunt:       Tea Party:      Cleanup:
              beetle          toaster         spray bottle
              vest            tea             dustpan
              grasshopper     honey           squeegee
Luke          Food Truck:     Office:         Cleanup:
              burger          stapler         broom
              ketchup         ruler           sponge
              tray            tape            glove

Table 2 Mean Responses and Ranges of Social Validity Questionnaire

Survey Statement                                       Discrete Trial
                                                       M (range)

This procedure or type of teaching would fit well in   3.75 (2-6)
my classroom.
The student appears to be experiencing discomfort      5.5 (3-7)
with this teaching procedure.
This teaching procedure is likely to result in         3.1 (1-5)
increased language for this student.
Overall the student seems to be having a positive      2.75 (1-4)
reaction to this teaching procedure.
This seems like a good use of this student's time.     3.25 (1-5)
I am likely to use this teaching approach with my      4 (3-6)
students who do not label objects.

Survey Statement                                       Play-based
                                                       M (range)

This procedure or type of teaching would fit well in   2.2 (1-2)
my classroom.
The student appears to be experiencing discomfort      6.4 (5-7)
with this teaching procedure.
This teaching procedure is likely to result in         2.2 (1-3)
increased language for this student.
Overall the student seems to be having a positive      1.6 (1-2)
reaction to this teaching procedure.
This seems like a good use of this student's time.     2.2 (1-3)
I am likely to use this teaching approach with my      2.5 (1-4)
students who do not label objects.

Note. 1 = Strongly Agree, 2 = Agree, 3 = Somewhat Agree, 4 = Neither
Agree nor Disagree, 5 = Somewhat Disagree, 6 = Disagree, 7 =
Strongly Disagree; n = 9
COPYRIGHT 2019 West Virginia University Press, University of West Virginia
No portion of this article can be reproduced without the express written permission from the copyright holder.
Copyright 2019 Gale, Cengage Learning. All rights reserved.

Article Details
Printer friendly Cite/link Email Feedback
Author:Duenas, Ana D.; Plavnick, Joshua B.; Maher, Courtney E.
Publication:Education & Treatment of Children
Article Type:Report
Date:Aug 1, 2019
Previous Article:Assessment and Treatment of Aggression During Public Outings.
Next Article:Effects of Behavioral Skills Training on Teacher Implementation of a Reading Racetrack Intervention.

Terms of use | Privacy policy | Copyright © 2019 Farlex, Inc. | Feedback | For webmasters