Comparing the Effect of Echoic and Listener Responding in the Development of Complex Intraverbals

Abstract

This study aimed to compare the effects of echoic and listener responding in the emergence of complex intraverbal behavior in four children with autism spectrum disorder (ASD). Each participant was taught to provide an echoic response or a listener response for different discriminative stimuli for each condition. We used a nonconcurrent multiple probe design across participants and adapted an alternating treatment design to compare the effects between the two conditions. Pre- and posttests were used to evaluate the effects of the two different prompt types in the emergence of complex intraverbals. The results indicate that the echoic response was more effective than the listener response at increasing the emergence of complex intraverbal responses in three out of four participants.

Intraverbal is considered a form of verbal behavior that is preceded by a verbal antecedent stimulus that does not have a point-to-point correspondence between the response and the stimulus and the behavior follows by a generalized conditioned reinforcement (Skinner, 1957). This type of verbal behavior can be in different forms, such as social interchanges (e.g., saying “You, too” when told “Stay safe”), word associations (e.g., “Long” when told “Short”), fill-in-blanks (e.g., “Round and round” when told “The wheels on the bus go”), and answering questions (e.g., “John” when told “What is your name?”). Intraverbal behavior plays a crucial role in the development of academic, language, and social interaction skills in individuals with autism spectrum disorder (ASD) and other developmental disabilities (Sundberg & Michael, 2001; Sundberg & Sundberg, 2011). Individuals with ASD usually experience difficulties developing intraverbal skills (e.g., Miklos et al., 2010) due to the complexity of such skills (Michael et al., 2011). Furthermore, some of the goals taught in the educational programs for individuals with disabilities are related to intraverbal behavior; therefore, investigating the efficacy of different teaching procedures to improve such skills is warranted (Stauch et al., 2017).

The complexity of intraverbal behavior can range from a simple format that requires simple discrimination (e.g., “The wheels on the bus go. . . .” or “Ready, set, . . . !”) to advanced, which require verbal conditional discrimination (“What do you wear on your head” or “What’s above a house?”). The difference between simple and conditional discrimination is that, in the latter, the response is reinforced when it comes under the stimulus control and the response is under the presence of another stimulus (Catania, 1998), whereas, in simple discrimination, the response is under the control of only one stimulus. Simple discrimination that is developed early in children is considered functional for the development of complex intraverbal behavior (Sundberg & Sundberg, 2011). According to the literature, the ability to make conditional discrimination for complex intraverbals is limited in children with ASD (Loukusa et al., 2007). Auditory conditional discrimination is more difficult to learn than visual discrimination in children with ASD (Serna et al., 1992). That might be because visual stimuli are more static than auditory stimuli, which are considered transitory. For example, when instructing a child to select a specific stimulus (e.g., a picture of a dog) in an array, the visual samples are still in the child’s view. The auditory conditional discrimination, on the other hand, the auditory sample (e.g., asking the child about his favorite animal) passes before the child can respond to the stimulus. Based on Sundberg and Sundberg (2011), developing other verbal and nonverbal skills ensures the development of intraverbal repertoire that does not become a rote verbal relation. When an individual talks about an absent event (intraverbal operant), the verbal behavior is not likely to be a rote response if the individual can accurately respond as a listener and a speaker to nonverbal and verbal stimuli. For example, in order for a child to respond to a question “Can you name a yellow fruit?,” the child should tact and identify the auditory stimuli “Fruits” and “Yellow” items as well as generalize and discriminate them from other groups. Therefore, it is imperative to examine various teaching strategies to strengthen both listener and speaker skills under the control of compound stimuli. Further, prior studies had proposed that developing listener skills may lead to the emergence of speaker skills without direct training (e.g., Ribeiro et al., 2010).

One of the main challenges involved in the acquisition of advanced intraverbals is that responses must come under the control of multiple verbal antecedent stimuli (Sundberg & Sundberg, 2011). According to Michael et al. (2011), there are two types of multiple control: divergent and convergent. For divergent multiple control, a single verbal antecedent evokes numerous verbal responses—for example, responding “Chips, candy, strawberry” to the statement “Tell me your favorite foods.” Convergent control, on the other hand, more than one verbal antecedent, evokes a single verbal response. For instance, presenting the question, “What foods do you see?” along with visual stimuli (picture cards of food), would evoke the verbal response “Apple.” Also, advanced intraverbal responses could come under divergent and convergent multiple control (Axe, 2008)—for example, responding with “Cucumber and lettuce” to the statement “Tell me some green vegetables.”

Some common problems that children with ASD exhibit related to faulty stimulus control are overselectivity/restricted stimulus control (Dube & McIlvane, 1999; Lovaas et al., 1979), which refers to attending to one feature of a stimulus and ignoring other relevant features. For example, when a child responds “Apple” to both “Tell me some fruits” and “Tell me some yellow fruits,” the verbal response is under the control of some verbal stimulus control, “Fruit” and not the other relevant verbal stimulus, “Yellow.” Therefore, several effective language interventions that focus on the development of intraverbals for children with ASD, such as transfer-of-stimulus control procedures (Braam & Polling, 1983; Luciano, 1986), video modeling (Sherer et al., 2001), peer-mediated interventions (Kamps et al., 1994), script fading (Krantz et al., 1981; Wichnick et al., 2010), and derived relational responding procedures (Grannan & Rehfeldt, 2012; May et al., 2013).

Transfer-of-stimulus control procedures have been reported to be effective in teaching intraverbal skills to individuals with language delays (Sundberg et al., 1990; Watkins et al., 1989), children with ASD (Kodak et al., 2012; Charlop-Christy & Kelso, 2003), and typically developing children (Partington & Bailey, 1993; Miguel et al., 2005). In this procedure, a response evoked by one stimulus becomes under the control of another stimulus. For example, a teacher can use visual (a picture of an “Apple”) or echoic (“Say apple”) prompts to transfer the stimulus control from the picture or the verbal prompt to the verbal antecedent (“What do you eat that’s red?”). The majority of the studies in the literature used transfer-of-stimulus control procedure to teach intraverbals that required simple discrimination (e.g., “What is your name?”; Finkel & Williams, 2001). A few other studies that implemented transfer-of-stimulus control targeted the development of complex intraverbal responses (Ingvarsson & Hollobaugh, 2011; Ingvarsson et al., 2012). Learning simple intraverbals can contribute to the development of more complex verbal behaviors (Sundberg, & Sundberg, 2011). Hence, there is a need to determine whether the transfer-of-stimulus control procedure would also be effective in teaching complex intraverbals.

Studies have shown that different response prompts, such as tact, textual (visual), and echoic (vocal), can be used in the transfer-of-stimulus control procedure when teaching intraverbals to children with and without developmental disabilities (e.g., Goldsmith et al., 2007; Ingvarsson et al., 2007; Valentino et al., 2012). However, only few studies compared different prompt topographies and found that the most effective prompt varied across participants (Finkel & Williams, 2001; Ingvarsson & Hollobaugh, 2011; Ingvarsson & Le, 2011; Vedora & Conant, 2015; Wallace et al., 2016). The learning history of each participant may play a role in determining the most effective and efficient prompt type (Finkel & Williams, 2001; Ingvarsson & Hollobaugh, 2011). For example, the results of Finkel and Williams (2001) and Vedora et al. (2009) indicated that textual prompts were superior in efficiency than vocal prompts in the establishment of intraverbals. However, another study, Cook et al. (2010), contradicted the findings of the previously mentioned studies showing that vocal prompts were either more or equally effective compared to textual prompts in teaching intraverbals.

The comparisons of the effectiveness and the efficiency of various prompt topographies (auditory or visual) when teaching intraverbals to children with ASD have concluded idiosyncratic results. For example, the results of Ingvarsson and Hollobaugh (2011) comparison of the efficiency between echoic (auditory) and tact (visual) in teaching intraverbals to young children with ASD yielded that both prompts were effective. However, visual prompts required less time to reach mastery criterion. These results were consistent with other studies (Finkel & Williams, 2001; Vedora & Conant, 2015) showing that visual prompts were more efficient compared to auditory ones. On the other hand, another study, Ingvarsson and Le (2011), compared different prompts (echoic, tact, and textual) and found that echoic prompts were superior to other visual prompts. In another more recent study, Swerdan and Rosales (2017), the investigators compared echoic and textual prompts to teach conversation skills (asking questions related to selected topics) to three students with ASD. The results demonstrated that echoic prompts required fewer trials to criterion compared to textual prompts.

It should be noted that most studies that compared different prompt topographies had focused on simple intraverbals compared to more complex forms of intraverbal responses. In addition, most studies did not measure the effect of transfer-of-stimulus control, echoic and visual prompts, on the emergence of intraverbal responses. Therefore, the purpose of the current study was to compare the effects of echoic and listener responses in the emergence of complex intraverbal responses under the control of verbal conditional discriminative stimuli.

Method

Participants

Four children with ASD, three boys and one girl, were recruited to participate in the study. The race of all participants was white. The participants were recruited from a center where they received ABA and other related services. They received 24–35 hr of ABA per week. The participants’ skill acquisition programs targeted different skills, such as mand, tact, listener responding, imitation, vocal imitation, intraverbal that require simple discrimination (e.g., fill-in-the-blanks), and other skills. Because the Verbal Behavior-Milestones Assessment and Placement Program (VB-MAPP; Sundberg, 2008) was used to conduct an assessment and build individual programs in the center, this particular criterion-referenced assessment was select to determine the children’s current levels of verbal behaviors. The was VB-MAPP conducted a month before the study. The total VB-MAPP scores were as follows: 146 for John, 133 for Sarah, 149 for Jacob, and 138 for Aiden. The participants’ scores showed that their skills were primarily in the 30- to 48-month range (Level 3). See Table 1 for more information on the participants’ characteristics.

Table 1.

Participant characteristics

Participants	Age	Diagnosis	Tact	LR3	Intraverbal subtest4			Echoic repertoire5
					Total	Group 7	Group 8
John	5:31	ASD	12.52	13	73	5	3	73
Sarah	6:3	ASD	11	12	62	3	1	56
Jacob	6:8	ASD	12	14	67	4	1	71
Aiden	7:0	ASD	13	14	70	7	4	70

¹Scores in the Verbal Behavior-Milestones Assessment and Placement Program (VB-MAPP; Sundberg, 2008); ²years: months; ³listener responding; ⁴total scores in the intraverbal assessment subtest; ⁵Early Echoic Skills Assessment (EESA)

The study took place in a therapy center (clinical facility) where the participants received ABA services. The instructor [the first author] conducted the sessions (pre- and postprobes, and training) in a one-on-one format between the participant and the instructor at the table, in the therapy room where their programs were implemented. The sessions lasted for approximately 5 min and were conducted once a day, three times per week.

Materials

Two individualized sets of stimuli were used for each participant based on their current ABA programs. Each set of stimuli consisted of two complex intraverbal statements (see Table 2 for a breakdown of stimuli and desired responses for each participant). The targeted items were based on each participant’s treatment plan. For example, John and Jacob did not acquire tacting and selecting stimuli based on adjectives (e.g., hot, sweet, cold, and salty). On the other hand, Sarah and Aiden did not acquire tacting and selecting stimuli based on adjectives related to shape and color. All participants had acquired the tact of each targeted verbal antecedent stimulus based on their individual ABA programs.

Table 2.

Targeted responses for each condition

Participants	Set 1 (Echoic Responses)	Set 2 (Listener Responses)
John	Q1: What are some hot drinks? Coffee, tea, hot chocolate Q2: What are some cold drinks? Milk shake, juice, soda	Q1: What are some sweet foods? Chocolate, candy, donuts Q2: What are some salty foods? Crackers, popcorn, spaghetti
Sarah	Q1: What are some green vegetables? Spinach, lettuce, cucumber Q2: What are some red vegetables? Beets, tomatoes, radish	Q1: What are some round foods? Apple, donuts, orange Q2: What are some long foods? Banana, cucumber, zucchini
Jacob	Q1: What are some cold drinks? Juice, soda, milk Q2: What are some hot drinks? Coffee, tea, hot chocolate	Q1: What are some salty foods? Chips, pizza, French fries Q2: What are some sweet foods? Jelly bean, cake, ice cream
Aiden	Q1: What are some red fruits? Apple, pomegranate, strawberries Q2: What are some yellow fruits? Banana, lemon, melon	Q1: What are some round foods? Cake, donuts, orange Q2: What are some long foods? Banana, cucumber, zucchini

Experimental Design

A nonconcurrent multiple probe design across participants (Watson & Workman, 1981) and an alternating treatment design (Sindelar et al., 1985) were used to compare the effectiveness between two different transfer stimulus control (echoic and listener response) in the emergence of complex intraverbal responses. Pre- and posttest data were collected to detect the emergence of complex intraverbal responses. We conducted an initial pretest for each condition for all participants. After the pretest, we conducted the training sessions for both experimental conditions simultaneously. One six-trial block was conducted for each condition in a counterbalanced order in each session. We conducted a posttest following the training sessions. We implemented intraverbal training for participants whose responses to both conditions were less than 50% over four consecutive sessions. The acquisition criterion was set for both conditions as responding at least five out of six correct responses across three consecutive sessions. The mastery criterion was set for intraverbal emergence as responding at least five out of six correct responses across three consecutive sessions for either condition.

Measurement

We collected data on the emission of emergent intraverbals as the primary dependent variable. Emergent intraverbal responses were defined as vocal responses to the presented question within 5 s. For example, when the instructor presented the question “What are some hot drinks?,” the correct response was “Coffee, tea, hot chocolate.” The participant had to emit at least three members of a specific class, with no particular order, to be considered a correct emergent intraverbal response. Repeated intraverbal responses were not included as correct responses. Also, data were collected on echoic and listener responses as ancillary measures. The echoic response was defined as repeating the relevant antecedent verbal stimuli (e.g., “Hot drinks”) within 5 s of the presentation of the instruction (e.g., “Say, hot drink”). Listener responding was defined as selecting the correct item when presented with an array of three stimuli and within 5 s after receiving the verbal instruction “Which one is cold?” The array contained the correct intraverbal response besides similar stimuli. For example, if the target listener response was to collect (a cold drink), the array included different types of drinks (e.g., hot, cold).

Procedure

The instructor conducted two sessions a day (one session for each condition) for 3–5 min each, 2–3 days per week during a 4–6-week period. During the session, the instructor presented each question under a condition three times in a counterbalanced order (a total of six trials in a session under each condition). For example, the instructor started the echoic response session with John by presenting the verbal instruction “What are some hot drinks?” and in another session selected to use the other question “What are some cold drinks?”

During pre- and postprobes, the instructor provided verbal praise and social reinforcement for appropriate session behavior (e.g., sitting appropriately, putting hands in lap), which were frequently used in their ABA programs. For training, however, the instructor used a token economy system based on each participant's individual schedule of reinforcement. At the end of the session, the participants were allowed to exchange these tokens for preferred items that were frequently used as reinforcers for each participant’s program as the backup reinforcers. Examples of these reinforcers were the following: cars, books, toys, and edibles.

Intraverbal Preprobes

The instructor presented the verbal antecedent stimulus “What are some farm animals?,” then waited for 5 s for the participant to respond. If no or incorrect responses occurred, the instructor moved to the next question. The instructor conducted three trials for each question in the set in a counterbalanced order under each condition. No feedback was provided for correct or incorrect responses. The reason for not providing feedback was to eliminate the effect of learning.

Echoic Preprobes

The instructor presented the verbal antecedent stimulus “Say [a] breakfast food,” then waited for 5 s for the participant to respond. If no or incorrect responses occurred, the instructor moved to the next verbal antecedent stimulus. The instructor conducted three trials for each question in the set in a counterbalanced order under each condition. No feedback was provided for correct or incorrect responses. The reason for not providing feedback was to eliminate the effect of learning.

Listener Responding Preprobes

The instructor presented a sample of a three-stimulus array and then the verbal antecedent stimulus (e.g., “Give me [a] living room furniture”). Afterward, the instructor waited for 5 s for the participant to respond. If no or incorrect responses occurred, the instructor moved to the next verbal antecedent stimulus. The instructor conducted three trials for each question in the set in a counterbalanced order under each condition. No feedback was provided for correct or incorrect responses. The reason for not providing feedback was to eliminate the effect of learning.

Echoic and Listener Response Training

For echoic response training, the instructor presented the verbal antecedent stimulus “Say [a] zoo animal,” then waited for 5 s for the participant to respond. If the participant did not respond, the instructor then provided a verbal prompt (e.g., saying “zoo animal”) and waited for 5 s for the participant to respond. If the participant did not respond to the verbal prompt, the instructor moved to the next trial. For correct and independent responses, the instructor provided tokens (number of tokens provided based on each participant’s individual program) and verbal praise. For prompted responses (responding after repeating the instruction), the instructor provided verbal praise only.

For the listener response training, the instructor presented a sample of a three-stimulus array and then the verbal antecedent stimulus (e.g., “Give me [a] hot breakfast foods”). If the participant did not respond to the verbal antecedent stimulus within 5 s, the instructor repeated the instruction and gave a gestural prompt (e.g., pointing to the correct visual stimulus). Then, the instructor waited for 5 s for the participant to respond. If the participant did not respond to the gestural prompt, the instructor moved to the next trial. For correct and independent responses, the instructor provided tokens and verbal praise. For prompted responses, the instructor provided verbal praise only.

For incorrect echoic and listener responses, the instructor followed an error-correction procedure by implementing the following steps: (1) repeat the antecedent stimulus; (2) provide immediate prompts; (3) provide a distractor (mastered skill); (4) provide a probe (repeated the instructions without prompts); and (5) reinforce the correct response to the probe with verbal praise. The instructor implemented the error correction procedure only once after the incorrect response. If the participant responded incorrectly during the probe within the error correction procedure (fourth step), the instructor repeated the prompt without requiring the participant to respond.

Postprobes

The instructor presented the verbal antecedent stimulus “What are some brown animals?,” then waited for 5 s for the participant to respond. If no or incorrect response occurred, the instructor moved to the next question. No feedback was provided for correct or incorrect responses. The instructor conducted three trials for each question in the set in a counterbalanced order under each condition. Postprobes tested the emergence of novel complex intraverbal responses.

Intraverbal Training

When the participants did not meet the mastery criterion for the emergence of complex intraverbal responses under each condition during postprobes, the instructor implemented intraverbal training. If the participant did not respond to the verbal antecedent stimulus “What are some kitchen appliances?” within 5 s, the instructor gave a verbal prompt, “Toaster, blender, and microwave.” If the participant did not respond to the verbal prompt, the instructor provided a second verbal prompt and waited for 5 s. If no response occurred, the instructor moved to the next question. For correct and independent responses, the instructor provided tokens and verbal praise. For incorrect responses, the instructor followed the error correction steps previously described. For prompted responses, the instructor provided verbal praise only.

Interobserver Agreement and Procedure Integrity

Interobserver agreement data were collected by an independent observer (a graduate student with a background in ABA), in vivo, during 100% of the preprobes and 80% of the training and postprobes sessions. The first author used the behavior skills training (BST; Reid & Parsons, 1995) protocol to train the independent observer in measuring the occurrence of the dependent variable and the treatment integrity. The training consisted of the following: (1) providing verbal instructions; (2) modeling; (3) role-playing; and (4) providing feedback. The duration of the training was 4 h, and it was completed across 2 consecutive days. The independent observer had to reach a mastery criterion (i.e., 100% agreement on the occurrence of the dependent measure and the treatment integrity checklist across four consecutive training sessions) before starting on IOA data collection. The agreement was calculated using a trial-by-trial method in each phase by dividing the number of agreements by the total number of agreements and disagreements multiplied by 100. The agreement for all participants for the pre- and postprobes was 100%. For the training sessions, the agreement averaged 96% for John (range: 90–100), 94% for Jacob (range: 90–100), 98% for Aiden (range: 98–100), and 97% for Sarah (range: 94–100). Treatment integrity was calculated by the independent observer using a checklist during 100% of the pre- and postprobes and training sessions. The treatment integrity was calculated by dividing the number of completed steps by the total number of steps multiplied by 100. The treatment integrity of the pre- and postprobes for all participants was 100% and averaged 99% for John, 98% for Jacob, 100% for Aiden, and 97% for Sarah for the training sessions.

Results

John did not emit any correct responses during the preprobes (Fig. 1). During training for the echoic condition, John’s responses averaged 4 and ranged between 1 and 6 correct responses. For the listener response condition, he averaged 5 correct responses and ranged between 2 and 6. Likewise, he reached mastery in the same number of trials (30) and sessions (six) to reach the acquisition criterion for both conditions (echoic and listener responding). During postprobes, John’s correct intraverbal responses averaged 5 and ranged between 4 and 6 for the set under both conditions. He also acquired mastery in the same number of trials and sessions to show emergence across both conditions.

Fig. 1.

The number of correct independent responses across participants

Figure 1 shows that Jacob did not emit any correct responses during the preprobes. Jacob’s echoic response averaged 4 and ranged between 1 and 6 correct responses. As for the listener response condition, Jacob’s correct responses averaged four and ranged between 2 and 6. Jacob acquired mastery after receiving more trials and sessions (48 trials and eight sessions) for the listener response condition compared to the echoic response condition (36 trials and six sessions). During postprobes, Jacob’s intraverbal responses for the echoic condition averaged 5 and ranged between 4 and 6. He showed emergence after receiving 24 trials and four sessions for the echoic condition. For the listener response condition, Jacob averaged 0.75 and ranged between 0 and 1 correct intraverbal responses.

Aiden did not emit any intraverbal correct responses (Fig. 1). During the echoic condition training, Aiden’s correct responses averaged 5 and ranged between 3 and 6. For the listener response condition, Aiden’s response averaged 4 and ranged between 1 and 6. Aiden achieved the acquisition criterion for the echoic response condition after receiving 30 trials and five sessions and 42 trials and six sessions for the listener response condition. During the postprobe sessions, Aiden’s correct intraverbal responses averaged 6 and ranged between 5 and 6 for the echoic condition set. He showed emergent intraverbal responses after receiving 24 trials and four sessions. For the listener response condition set, Aiden’s correct intraverbal responses averaged 2 and ranged between 1 and 3.

Sara did not emit any correct intraverbal responses during preprobe sessions (Fig. 1). During training for both conditions, Sara’s response averaged four and ranged between two and six. Sarah acquired mastery after receiving more trials and sessions (42 trials and seven sessions) for the echoic response condition compared to the listener response condition (66 trials and 11 sessions). During postprobes, Sara’s intraverbal responses averaged 2 correct responses and ranged between 1 and 4 for the echoic condition set. For the listener response condition, Sara’s response averaged 0.25 and ranged between 0 and 1. Because all participants showed emergent intraverbal responses under either condition, Sara needed direct intraverbal training. During the intraverbal training, Sara’s correct responses averaged 5 and ranged between 4 and 6 for the echoic condition set. For the listener response condition set, Sara’s response averaged 5 and ranged between 5 and 6. After receiving the direct intraverbal training, Sarah reached the mastery criterion after 18 trials and 3 sessions for both conditions.

Discussion

The current study compared the effects of transferring echoic and listener responding-to-intraverbal in the emergence of complex intraverbal responses. The overall results demonstrated that echoic-to-intraverbal training resulted in more emergence for three out of four participants, whereas the listener response-to-intraverbal resulted in the emergence of complex intraverbals for John only. Based on John postprobe results, he demonstrated an emergence of complex intraverbal responses without direct intraverbal training across all sets for both conditions (echoic and listener response-to-intraverbal). Jacob and Aiden, on the other hand, showed emergence for the set under the echoic-to-intraverbal condition only. Sarah did not reach the mastery criterion for the emergence of complex intraverbal responses under either condition. Therefore, it was necessary to implement intraverbal training for Sarah so that she could emit correct complex intraverbal responses. Three out of four participants demonstrated an emergence of complex intraverbal skills after the implementation of echoic-to-intraverbal training. The findings of the present study are consistent with previous research showing that echoic training was affect in the emergence of simple (e.g., Ingvarsson & Le, 2011; Kodak et al., 2012; Roncati et al., 2019) as well as complex intraverbal responses (e.g., Carroll & Kodak, 2015; Swerdan & Rosales, 2017). Future research is encouraged to replicate the results of the study with more participants to demonstrate which type of transfer-of-stimulus control is more effective in the emergence of complex intraverbal skills.

Even though three out of four participants showed an emergence of complex intraverbal responses, they required a different number of trials and sessions to meet the mastery criterion for the emergence under both sets. For example, John and Jacob received a similar number of trials and sessions to meet the mastery criterion, whereas Aiden required fewer trials and sessions to demonstrate an emergence of complex intraverbals. The different levels of complex intraverbal skills that the participants demonstrated might be the factor that contributed to the differentiated results. For example, John scored the highest in the intraverbal subtest (Sundberg & Sundberg, 2011) under the questions that required verbal conditional discrimination (e.g., “What makes you sad?” and “Tell me something that is not a food”), which might be the reason for reacting mastery criterion for the emergence of complex intraverbals.

Echoic-to-intraverbal training demonstrated emergence of complex intraverbal skills compared to listener response-to-intraverbal for Jacob and Aiden. The main reason for this result could be that Jacob and Aiden had a strong echoic repertoire as compared to Sarah’s level in the EESA. The participants’ unique characteristics, especially their verbal behavior repertories, had a great influence on evaluating the emergence of intraverbal behavior (Ingvarsson & Le, 2011; Miguel et al., 2008). For instance, Sarah’s scores for the tact (11), listener responding (12), echoic (56), and intraverbal subtest (62; Sundberg & Sundberg, 2011) were lower than the other participants, which might be one reason for her failing to demonstrate the emergence of complex intraverbal responses. Indeed, other previous studies have reported that participants with limited vocal repertoires did not show an emergence of verbal behavior (e.g., Delfs et al., 2014; Horne et al., 2004; Lee et al., 2015; Sprinkle & Miguel, 2012). Given the result of the current study, the echoic-to-intraverbal procedure might be most effective for children with ASD with stronger speaker and listener repertories (scoring Level 3 in the VB-MAPP, which demonstrates verbal skills of typically developing children within the 30- to 48-month range).

The similarity in topography between echoic prompt and intraverbal response (both requiring the participant to engage in a listener and speaker behavior) might have contributed to the results (Finkel & Williams, 2001). During the echoic-to-intraverbal condition, the participants were required to pay attention to the verbal stimuli and overtly engage in a verbal response (repeating the category’s name). In contrast, the participants, during the listener response-to-intraverbal, had to emit a selection-based response (giving the correct visual stimulus). Engaging in both listener and speaker behavior might have prompted the development of intraverbal skills (Schlinger, 2008).

Participants’ prior experience with vocal prompts might play a role in the positive results of the echoic condition. According to the participants’ skill acquisition programs, vocal prompts were mainly used to teach most verbal behaviors. Prior studies have supported the notion that learning histories have a role in determining the effectiveness of specific teaching strategies in the emergence of verbal skills (Cook et al., 2010; Coon & Miguel, 2012; Roncati et al., 2019).

Also, the presence of prerequisite skills in individuals prior to the start of intraverbal training might facilitate the development of complex intraverbal skills (Sundberg & Partington, 1998). In particular, John, Jacob, and Aiden had acquired the ability to emit spontaneous mands for preferred items and information as well as tact propositions, pronouns, and some adjectives and adverbs. According to Goldsmith et al. (2007), prior research had reported that the participants who mastered the ability to emit between 50 and 200 tacts and mands were also able to acquirer more functional intraverbal skills. As for listener responding skills, even though all participants demonstrated the ability to follow instructions that require verbal conditional discrimination, only John showed an emergence for the listener response condition. Future research should investigate whether listener responding plays a role in the development of complex intraverbals.

The results of the current study should be considered in light of several limitations. The first limitation was related to the lack of generalization data. Given that previous studies did not include data regarding generalization, future research is encouraged to evaluate the effects of the intervention on the generalization of the emergence of complex intraverbals. Another limitation stems from the lack of data regarding the measurement of novel intraverbal responses. Therefore, further studies should collect data on the emergence of novel complex intraverbal responses. Because no social validity data were collected, further studies are encouraged to include an evaluation of the social validity of the intervention and the social significance of the results. Also, some items in sets were similar across conditions (e.g., banana as the correct response for the echoic and listener responding sets), which could have affected participants’ differentiation between conditions. Thus, further studies should investigate whether including similar and different items in sets across conditions would affect the results. Finally, because we used a nonconcurrent multiple probe design instead of concurrent, we could not control for history effects (Carr, 2005).

The findings of the study might support the use of echoic-to-intraverbal training in the emergence of intraverbal responses in some cases of children with ASD. Given that there is a correlation between the existence of speaker and listener behaviors and the emergence of novel verbal responses (e.g., Lee et al., 2015; Ribeiro et al., 2010; Sprinkle & Miguel, 2012), clinicians are encouraged to teach both skills simultaneously and test for novel advanced verbal responses. The child’s learning history should also be considered when developing individualized programs due to the possible effects of learning histories during the progress of skill acquisition (Kay et al., 2020). In case the child’s learning history is unknown, clinicians can conduct a prompt-type assessment (Kodak & Halbur, 2021; Seaver & Bourret, 2014) to identify the most efficient prompt type. Further, clinicians should consider adding echoic-to-intraverbal training in their teaching procedures to improve the salience of each relevant verbal stimulus to avoid faulty stimulus control.

Data Availability

Data sharing not applicable to this article as no datasets were generated or analyzed during the current study.

Declarations

Ethical Approval

All procedures performed in studies involving human participants were in accordance with the ethical standards of the institutional committee (Research Ethics Committee, King Saud University, KSU-HE-18-110) and with the 1964 Helsinki declaration and its later amendments or comparable ethical standards.

Informed Consent

Informed consent was obtained from all individual participants included in the study.

Conflict of Interest

The authors declare that they have no conflict of interest.