Materials and methods

Participants

A total of 34 Brazilian families were invited to participate in this study. Seventeen mothers and ten fathers agreed and responded to the pre-test instruments. However, the total sample that participated in all stages of the study consisted of 12 mothers and five fathers and their children, who were randomly assigned to the experimental group (six mothers and three fathers) and the control group (six mothers and two fathers). Furthermore, among the participating parents, five were couples, of whom three were in the Experimental Group (EG) and two in the Control Group (CG). Among the mothers, seven participated individually, three in the EG and four in the CG. All participants were parents of children diagnosed or undergoing diagnostic evaluation for autism spectrum disorder, aged between one and four years.

Inclusion criteria: 1) Parents had to be at least 18 years of age, and 2) Children had to have a diagnosis of ASD through a report issued by a pediatric neurologist or child psychiatrist or be undergoing diagnostic evaluation. Exclusion criteria: 1) Parents with a history of or undergoing treatment for psychotic disorders, and 2) Children and their parents who had previously received ABA intervention. Criterion 2 was established as a means of controlling for possible effects of a previous intervention.

The sociodemographic characteristics of the participants are presented in Table 1. In general, mothers and fathers in both the experimental and control groups had similar profiles in terms of age range, marital status, paid employment, and number of children. Greater variability between the groups was observed in terms of educational level and socioeconomic status, particularly among the fathers.

Table 1: Sociodemographic characteristics of the mothers and fathers in the CG and EG 

^Note: EG: experimental group; CG: control group.

Regarding the sociodemographic characteristics of the children who comprised the sample, as shown in Table 2, six were part of the EG and six of the CG. Overall, the groups exhibited a similar distribution in terms of age and school attendance. Most of the children were male and were the firstborn in their families. In the experimental group, all had a diagnosis of ASD, while in the control group, some were still undergoing diagnostic evaluation. A diversity of communication modes was observed, with a greater predominance of both vocal and gestural difficulties in the control group.

Table 2: Characterization of the children in the EG and CG 

^Note: EG: experimental group; CG: control group.

Procedures

The parents of children with ASD were identified at a private clinic in the city of Londrina/PR, Brazil, where they had sought care for their children, and were invited to participate in the present study. In cases of acceptance, participants were allocated to the EG and CG in order of arrival. The method of simple randomization by fixed blocks of two participants was used as a way to reduce bias and allocate participants into the groups in a balanced manner (^{Efird, 2010}). An individual meeting was scheduled to present the objectives and phases of the research. After clarifying any doubts, the parents signed the consent form. On a previously scheduled date and time, the children were assessed using the VB-MAPP.

Regardless of the group randomization, based on the VB-MAPP data, a quarterly IEP was developed for each child. It included short-term, medium-term, and long-term goals aimed at developing the child’s skills through the application of ABA teaching programs. These programs were applied and measured in each session conducted with the child, and were therefore continuously reviewed and systematically adjusted based on the child’s performance throughout the 12 weeks of intervention. Consequently, as the intervention was conducted with children from both groups, it was treated as a constant variable in the study design.

Simultaneously with the ABA intervention for the children, the parents in the GE participated in the Parent Orientation and Teaching Program (Programa de Orientação e Ensino para Pais - PROEP), developed for this study, based on Analytic-Behavioral (ABA) principles and implemented by the first author. The guidance was provided individually and in person. The PROEP was not monitored by an external judge, but weekly external supervision sessions were conducted. The PROEP consisted of 12 weekly meetings, each lasting approximately 90 minutes, and addressed topics such as welcoming the parents’ concerns regarding their child’s diagnosis. The description of the PROEP is presented in Table 3 .

Table 3: Description of the PROEP 

In each meeting, the content was described and addressed in an explanatory and dialogical manner with the parents, including scheduled observations during the child’s sessions and small homework tasks (reports, records, or videos on topics relevant to the theme worked on in the meeting and related to the items in the child’s IEP). Doubts regarding the challenges experienced during the week with the child and the tasks requested in the previous meeting were also resolved (^{Ferreira, 2023}).

Since it is an ABA program involving mothers, fathers, and their children with ASD, in some clinical sessions, behavioral management videos that they could practice at home (video modeling) were recorded, and in others, they were able to participate in the session in person, receiving guidance and feedback. The type of conduct adopted (video modeling or in-person feedback) was determined by the short-term, medium-term, and long-term goals described in the child’s IEP and the results presented by the child in each session. At the end of the trimester, the children were reassessed using the VB-MAPP.

Results

Table 4 presents the comparisons of the verbal repertoire of the children in the EG in relation to those in the CG based on the VB-MAPP in the pre- and post-test phases. In the pre-test, the children in the EG presented higher means than those in the CG in nine of the 15 evaluated dimensions, namely: mand, tact, listener, play, imitation, echoic, LRFFC, intraverbal, and language; however, there were no statistically significant differences between the groups. The group dimension was not evaluated in either group due to precautions related to the COVID-19 pandemic, which was ongoing at the time of data collection.

In the post-test, the comparisons of the verbal repertoire of the children in the EG in relation to those in the CG based on the VB-MAPP showed that the children in the EG performed better in all evaluated repertoires. However, the differences were not significant (Table 4).

Table 4: Comparisons between the verbal repertoires of the children in the EG and CG in the pre- and post-test 

^Notes:M: mean; SD: standard deviation; EG: experimental group; CG: control group; Pre: pre-test; Post: post-test; VB-MAPP: Verbal Behavior Milestones Assessment and Placement Program; VB-MAPP*: Only the categories that were scored in developmental milestones in the instrument were presented; Significance level: p < .05; Mann-Whitney test.

Table 5 presents the intragroup comparisons of the verbal repertoire of the children in the EG and CG based on the VB-MAPP in the pre-test and post-test. It was observed that in all dimensions evaluated in the EG, the means were higher in the post-test, with five dimensions showing statistically significant differences: listener (p = .041), VP/MTS (p = .042), play (p = .027), social (p = .042), and imitation (p = .027). In the CG, the difference was significant for play (p = .026). For the CG, higher means were also observed in the post-test for the evaluated dimensions, except for the vocal repertoire.

Comparisons were made between the learning barriers of the children in the EG in relation to the CG at the pre-test moment based on the VB-MAPP. The EG presented higher scores in seven of the 18 evaluated barriers, and the CG in seven of them. In five of the barriers, performance was equal for both groups. Significant differences were observed only in eye contact, which was higher for the CG (p= .003), as shown in Table 6.

Considering the post-test of the comparisons between the learning barriers of the children in the EG in relation to the CG based on the VB-MAPP, it can be seen that the EG presented higher means in seven barriers, while the CG presented higher means in nine of them. In only two, the result was the same for both groups. However, the difference was significant only for eye contact, with lower performance for the CG (p = .013). The data are presented in Table 6.

Table 5: Intragroup comparisons between the verbal repertoires of the children in the EG and CG in the pre-test and post-test 

^Notes:M: mean; SD: standard deviation; EG: experimental group; CG: control group; VB-MAPP: Verbal Behavior Milestones Assessment and Placement Program; VB-MAPP*: Only the categories that were scored in developmental milestones in the instrument were presented; Significance level: p < .05; Wilcoxon test.

Table 6: Comparisons between the learning barriers of the children in the EG and CG in the pre- and post-test 

^Notes:M: mean; SD: standard deviation; EG: experimental group; CG: control group; Pre: pre-test; Post: post-test; VB-MAPP: Verbal Behavior Milestones Assessment and Placement Program; VB-MAPP*: Only the categories that were scored in developmental learning barriers in the instrument were presented; Significance level: p < .05; Mann-Whitney test.

Table 7 demonstrates the comparisons between the learning barriers of the children in the EG at the pre- and post-test moments based on the VB-MAPP. It can be observed that the children in the EG presented a reduction in the mean in thirteen of the eighteen evaluated learning barriers in the post-test; however, this difference was not significant.

Table 7 also presents the pre- and post-test data of the CG in the comparison of learning barriers. It can be seen that there was a decrease in the mean in sixteen of the evaluated dimensions, with this difference being significant in difficulties making eye contact (p = .046).

Table 7: Intragroup comparisons between the learning barriers of the children in the EG and CG in the pre- and post-test 

^Note:M: mean; SD: standard deviation; EG: experimental group; CG: control group; Pre: pre-test; Post: post-test; VB-MAPP: Verbal Behavior Milestones Assessment and Placement Program; VB-MAPP*: Only the categories that were scored in developmental learning barriers in the instrument were presented; Significance level: p < .05; Wilcoxon test.

Discussion

The application of ABA science for children with ASD involves direct work with the child, beginning with an assessment of the child’s behavioral repertoire. Based on this data, an individualized teaching curriculum is developed with the objective of promoting the development of motor, social, communicational, academic, and daily living skills, in addition to reducing behaviors considered dysfunctional (^{Farias & Elias, 2020}). This process must be monitored through continuous evaluations regarding the effectiveness of the teaching procedures employed, so that, if necessary, the proposed teaching contingencies can be adjusted and rearranged (^{Barcelos et al., 2020}). Being considered an intensive therapy, intervention periods may vary from four to 40 hours weekly, defined according to the child’s needs and the availability, including financial, of the parents. Accordingly, due to the high cost generated by the intensity required in ABA interventions and the scarcity of qualified professionals to apply them, parental training has been considered a viable alternative to intensify children’s learning and to generalize the repertoires developed in therapy to family life (^{Romano & Bagaiolo, 2022}).

Parental guidance that provides information about ASD, reinforcing the importance of parental involvement in child development, can contribute to a more consistent support network, resulting in parents who are more committed to the interventions (^{Ang & Loh, 2019}; ^{Crowell et al., 2019}). The present study evaluated the effects of a parent guidance program, based on ABA, on the verbal behavior of children with ASD. For this purpose, comparisons were made between the EG and CG. The initial hypothesis was that the parent program, combined with ABA interventions conducted with the child, would result in superior verbal behavior performance for the EG and a decrease in learning barriers evaluated by the VB-MAPP.

Comparisons regarding children’s verbal repertoire showed that, at pre-test, the groups were equivalent, although the EG presented higher means in a larger number of developmental milestones assessed; however, the differences were not significant. Comparing the two groups at post-test, superior performance was observed in the EG across all constructs investigated, suggesting a possible relation to the parent program, although the differences were not significant. In the systematic review and meta-analysis conducted by ^{Deb et al. (2020}), all studies included showed a positive effect of interventions performed with parents on behaviors of children with ASD, although not all studies found significant differences. The lack of statistical significance in between-group comparisons may be related to the small sample size and the individual ABA interventions that children in both groups were undergoing, which may also have limited the identification of differences produced by the parent intervention. Among the VB-MAPP dimensions showing increase at post-test for the EG compared to the CG, notable were the verbal operants: tact, echoic, and intraverbal, as well as listener skills and LRFFC. Although without statistical differences, these results highlight the importance of parental involvement in interventions, which can increase the likelihood of successful outcomes (^{Burrell & Borrego Jr, 2012}).

Considering the intragroup comparison results, this hypothesis can be reconsidered, since the five dimensions in which the EG showed statistically significant results (listener; VP/MTS; play; social; and imitation) are considered prerequisites for vocal verbal operants. Given that the fourth session of the PROEP addressed teaching skills to the child, parents participating in the PROEP began to encourage generalization of the child’s learning at home, which may suggest the program’s influence on optimizing the target skills mentioned. Furthermore, it is known that with the acquisition of new skills, interfering behaviors tend to decrease in frequency and intensity, which may also have contributed to the result, since in the sixth session parents were taught to functionally analyze their children’s behavioral responses. Studies such as ^{Andrade et al. (2024}) and ^{Bradshaw et al. (2022}) reinforce the importance of parental involvement to promote development, social skills, and communication in children with ASD.

^{Andrade et al. (2024}) reviewed studies on distance educational programs for parents of children with ASD and found positive results in daily routines, behavioral flexibility, and children’s communication. In the present study, the gains observed in the behavioral repertoire of the children in the EG in intragroup comparisons may indicate greater comprehension of instructions, improved participation in group and school activities, and expanded interactions with parents, siblings, teachers, peers, and other significant adults. These improvements contribute to the children’s development, support their school inclusion, and enhance their daily lives and those of their families. Although the results did not achieve statistical significance between groups, the advances obtained by the children in the EG may contribute to family relationships and quality of life for children and their families, presenting a clinically significant impact.

^{O’Donovan et al. (2019)} highlighted that individual parental training interventions improve the child’s repertoire and management of dysfunctional behaviors; however, for these to be effective, they should last an average of 10 to 12 weeks. Although time-consuming and demanding for both healthcare providers and parents, the PROEP adhered to this recommendation by offering 12 weekly sessions conducted with parents.

^{Anjos and Morais (2021}) noted that establishing a bond between the therapeutic team, the child, and the parents can positively influence challenges faced from the diagnosis onward. This relationship allows professionals to understand how family interactions occur in response to the new condition and to develop strategies that may alleviate difficulties parents face during this process. According to the authors, the problem-focused strategy is the most adopted by parents of children with ASD, which consists of sharing fears and doubts related to the diagnosis with healthcare providers and people in their environment, exchanging experiences, and combining efforts. The authors also highlighted that parents with higher educational and socioeconomic levels tend to hold more favorable beliefs about developmental difficulties based on therapeutic possibilities available to them, even acknowledging that ASD is a lifelong condition. Considering that families in this study’s sample were attended in a private clinic and predominantly belonged to socioeconomic class AB among mothers and fathers, this may have influenced the results obtained.

The intragroup comparison results indicated higher means in all developmental milestones at post-test for both groups, with statistical significance in five dimensions for the EG and one for the CG. This result suggests greater effectiveness for an ABA science application associating child intervention and parental training in acquiring new skills, especially those considered prerequisites for vocal behaviors, compared to an ABA intervention focused only on the child. ^{Cavalcante et al. (2023}), in their literature review, presented several studies demonstrating benefits of early intervention based on Applied Behavior Analysis principles for ASD. More specifically, ^{Farias and Elias (2020}) reported the effectiveness of an ABA intervention, using the VB-MAPP, for teaching multiple verbal operants to triplets, three years and six months of age, with ASD. However, in the present study, the superior performances of the children in the EG indicate the importance of systematic intervention with parents.

Evaluating learning barriers allows identification of behavioral repertoires that may negatively influence or even prevent children’s learning (^{Montallana et al., 2019}). Considering the 24 learning barriers investigated, 18 were presented by children in this study.

In a comparison between the two groups at pre-test, equivalence was observed in the barriers with higher means, seven for each group, with difficulty making eye contact being significantly greater in the CG. This significance remained unchanged at post-test; however, there was an increase in the dimensions where higher means were observed in the CG and a reduction in the number of barriers with higher means in the EG.

Comparing the mean score of each group before and after intervention, the EG showed lower means in more constructs assessed, although this difference was not significant. Although the CG’s decrease occurred in fewer learning barriers, there was a significant reduction in difficulty making eye contact. This finding suggests that the PROEP carried out with parents may have facilitated the management of challenging behaviors that impaired the children’s learning. ^{Bagaiolo et al. (2019}) assessed the impact of parent training on the management of challenging behaviors of their children with ASD and found a reduction of more than 70 % of this repertoire in the final two weeks of the intervention.

Final Considerations

This study made it possible to evaluate the effects of an ABA program combined with a specific parental guidance program on the verbal repertoire of their children with ASD. The results were consistent with the literature regarding the effectiveness of evidence-based practices, particularly ABA, for the development of the verbal repertoire of children with ASD. The gains of the EG were greater than those of the CG in verbal performance, as assessed by the VB-MAPP. This result suggests a positive impact on 31% of the behaviors evaluated for the EG. For the CG, improvement occurred in 6% of the behaviors evaluated by the VB-MAPP. However, the intervention with ABA and parents did not result in significant improvement in learning barriers. Although no statistically significant differences were observed between the groups, the advances obtained by the children in the EG suggest clinically relevant gains, especially in skills related to functional communication and social interaction. Such advances may produce improvements in family routines and the inclusion of children with ASD.

Some limitations of the study may have influenced these findings, among them, the difficulty of adherence to the program by the parents, especially due to excessive absences from the sessions, the failure to complete the proposed tasks, the difficulty in engaging in the suggested activities with the child, and the work overload to cover the costs of the child’s interventions, which resulted in a limited sample. The intervention having sessions related to the characteristics of ASD, ABA concepts, and procedures exclusively focused on child development may have contributed to parental overload, highlighting the need for a space for more directed listening to the parents.

The fact that the sample was predominantly composed of parents with socioeconomic classes A and B may also be considered a limitation for the study, making the generalization of the results unfeasible, since it is not a representative sample of the Brazilian population and consists mainly, especially in the EG, of individuals with greater access to information, therapies, and support networks. Furthermore, the children in both groups were in intervention, and the objectives were individualized and established based on a prior evaluation that considered their behavioral reserves, excesses, and deficits. Consequently, it is also possible that this aspect interfered with the results obtained from the intervention with the parents.

Other limitations found in the study refer to one of the researchers having conducted the interventions and the intervention fidelity not being monitored by external judges. Future studies could include distinct and independent implementers unrelated to the research objectives, instruments for monitoring fidelity, as well as strategies to increase parental adherence. Additionally, there are no widely recognized studies published in high-impact scientific journals presenting psychometric properties of the Portuguese version of the VB-MAPP (translated by ^{Martone, 2017}). The original instrument by ^{Sundberg (2008}) was also not developed based on traditional psychometric criteria but rather as a functional tool from Behavior Analysis for the assessment of verbal repertoires. Despite its widespread clinical use, the VB-MAPP lacks formal evidence of validity or reliability in the classical psychometric sense, which may constitute a limitation of the present study.

Although the results were promising, there is a clear need for more robust investigations with larger populations, follow-up assessments, and programs exclusively focused on child care, with evidence-based practices that also include experiences related to the recovery of other roles these individuals have in the world beyond parenting a child with ASD. Despite the limitations, this study contributes evidence regarding interventions performed with parents and advances in the verbal repertoire of their children with ASD, reinforcing the importance of including parents in their children’s therapeutic process.