Digital Intervention Training for Intraverbal Behavior in Children with Autism Spectrum Disorder

Digital Intervention Training for Intraverbal Behavior in Children with Autism Spectrum Disorder | Research Square window.SnipcartSettings = { analytics: { enabled: false } }; (function() { var accessVector = localStorage.getItem('access_vector') || ''; window.dataLayer = window.dataLayer || []; if (accessVector) { window.dataLayer.push({ user: { profile: { profileInfo: { snid: accessVector } } } }); } })(); (function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src='https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);})(window,document,'script','dataLayer','GTM-K279D39R'); Browse Preprints In Review Journals COVID-19 Preprints AJE Video Bytes Research Tools Research Promotion AJE Professional Editing AJE Rubriq About Preprint Platform In Review Editorial Policies Our Team Advisory Board Help Center Sign In Submit a Preprint Cite Share Download PDF Research Article Digital Intervention Training for Intraverbal Behavior in Children with Autism Spectrum Disorder Yingjian Zhang, Huixin Lu, Zhiguo Hu This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-8504009/v1 This work is licensed under a CC BY 4.0 License Status: Posted Version 1 posted You are reading this latest preprint version Abstract Intraverbal deficits represent a core characteristic of children with Autism Spectrum Disorder(ASD), significantly impairing their social communication abilities. Traditional face-to-face interventions are often constrained by limited accessibility and insufficient generalization of skills, while current digital approaches rarely focus specifically on intraverbal behavior. This study aimed to develop a gamified digital intervention grounded in the "Conversation Train" framework and to conduct a systematic evaluation of its effectiveness. A total of 40 children with ASD, aged 5 to 10 years, were randomly assigned to either an experimental group or a control group. The experimental group received standard training supplemented with 30 sessions of gamified digital intervention(15 minutes per day, five days per week), whereas the control group received standard training only. Outcome assessments included standardized instruments—the Autism Treatment Evaluation Checklist(ATEC) and the Communication Behavior Checklist (CB)—as well as a researcher-developed intraverbal behavior scenario test comprising two subtests: dialogue selection and real-time dialogue performance. No significant differences were observed between groups in baseline demographic or clinical characteristics (all p>0.05). Following the intervention, the experimental group showed significantly lower ATEC scores (Z=-3.551,p<0.001), higher CB scores (Z=-2.943,p=0.003), and improved performance on both subtests of the scenario test (dialogue selection: Z=-4.097, p<0.001; real-time dialogue: Z=-3.758,p<0.001), with moderate to large effect sizes (r = 0.49–0.97) across all measures. In contrast, the control group exhibited no significant changes. These findings indicate that the proposed gamified digital intervention effectively enhances intraverbal behavior and reduces core symptoms of ASD in children, offering a feasible, accessible, and scalable alternative to conventional face-to-face interventions. Autism spectrum disorder ASD intraverbal behavior digital intervention digital games Figures Figure 1 Figure 2 Figure 3 Figure 4 Highlights • Developed Conversation Train-based digital game for ASD children (n=40) • Improved dialogue selection (+67%) and real conversation skills (+85%) • Reduced ATEC scores (Z=-3.551, p<0.001) vs control group • B/S architecture enables multi-device accessibility 1 Introduction Autism spectrum disorder (ASD) is a neurodevelopmental disorder characterized by persistent deficits in social communication, repetitive and stereotyped behaviors, and restricted interests (American Psychiatric Association, 2013 ). In recent years, the global prevalence of ASD has increased significantly (Zeidan et al., 2022 ; Ameis, 2018). The latest epidemiological surveys showed that its prevalence has reached 1%–2%, severely affecting children’s social adaptation and qualities of life (Shaw et al., 2025). Among the core symptoms of ASD, intraverbal behavior impairments are particularly prominent, which are manifested by difficulties in initiating conversations, insufficient topic maintenance, and deficits in nonverbal communication.These impairments directly hinder social connections with others and have become key factors affecting the functional prognosis of children with ASD (Cui & Wang, 2023; Moraleda-Sepulveda, 2025). The term "intraverbal behavior" first appeared in Skinner’s Verbal Behavior (Skinner, 1957), which was classified into six main functional categories: mand, tact, echoic, intraverbal, textual, and transcription. Children with ASD exhibit various degrees of deficits in intraverbal behavior, including inability to understand turn-taking rules in conversations, difficulty in adjusting language content according to context, and lack of ability to actively maintain topics (ALMåS et al., 2022 ; Chen & Yang, 2014 ). Notably, as the core of social communication, the lack of such ability further leads to secondary problems such as peer interaction difficulties and family interaction pressure for children with ASD (Callanan, 2021). For a long time, traditional offline interventions, such as applied behavior analysis and speech-language therapy, have been the mainstream of language intervention for ASD, with their effectiveness partially verified (Brignell et al., 2018 ; Smith, 2001 ; Virues-Ortega et al., 2022 ). However, these intervention strategies have the following limitations: (1) the shortage of professional therapists make such intervention services less accessible ;(2) high costs of such face-to-face interventions are great financial challenges most families face (Bekele et al., 2013 ); (3) the standardization of offline interventions is insufficient, and differences in therapists’ skills and operations may affect the stability of intervention effects (Dawson, 2010); (4) traditional interventions mostly adopt structured tabletop training, lacking skill transferring to natural scenarios, which fails to meet the needs in real interactive contexts for children with ASD (Panceri, 2021). With the development of digital technology, digital interventions based on serious game and visualization have gradually become a research hotspot in the field of ASD (Zhang et al., 2022 ). Digital interventions, with their advantages of repeatability, standardized operation, and cross-scenario application, provide new ideas for addressing the pain points of traditional interventions (LaLonde et al., 2020 ). Systematic reviews have shown that digital interventions have significant effects on improving social attention (Xu et al., 2024 ) and IQ (Wang et al., 2024 ) in children with ASD, and their gamification design can effectively trigger children’s intrinsic motivation and extend intervention participation time (Rezayi et al., 2025 ). However, existing studies utilizing digital interventions mostly focus on training of the general social skills such as cognition and emotion recognition in children with ASD. To our knowledge, no study to date has used digital game to intervene in the intraverbal behavior in children with ASD. To address this research gap, the present study aimed to develop a digital game-based intervention program grounded in dialogue structure modeling and conduct a randomized controlled trial. The game was designed based on the idea of Learning Conversations with Trains ( The Conversation Train )(Shaul, 2014 ), utilizing the potential interest in trains of children with ASD. Specifically, the locomotive, power carriage, passenger carriage, track converter, and train tail are analogized to greetings, dialogue-promoting utterances, conversational turns, topic shifts, and conversation endings in a dialogue, respectively. Through analogical reasoning methods, dialogue rules were structured and visualized, in order to improve the abilities of intraverbal behavior in children with ASD.We hypothesize that compared with conventional intervention, the intervention model incorporating a digital game could be more effective in improving the intraverbal behaviors abilities in children with ASD. 2 Methods 2.1 Participants A total of 40 children with ASD aged 5–10 years were recruited and randomly assigned to the experimental group and the control group. The inclusion criteria were as follows: (1) Children aged 5–10 years; (2) A diagnosis of ASD by a professional clinician; (3) Having intraverbal behavior deficits reported by rehabilitation therapists; (4) Having moderate-to-high functioning as evaluated by the therapist.The exclusion criteria included: (1) children with severe ASD; (2) those with other neurodevelopmental disorders or physical disabilities or chronic conditions; (3) children with a family history of other genetic disorders; (4) those receiving medication currently. All the parents or legal guardians of the participants have signed the written informed consent form. The participants in the control group received only daily training, while the participants in the experimental group received both daily training and digital intervention. In the experimental group, two participants withdrew due to illness, remaining 18 participants who finally completed the post-test. In the control group, four participants did not complete the routine intervention and dropped out.Another two participants in the control group finished the daily training but did not take part in the post-test; however, their rehabilitation therapists completed the assessment scales of these two participants. Eventually, fourteen participants’ full data in the control group were collected. To compensate for the relatively small sample size of the control group, 16 data (including the data from interactive dialogue with the two children who did not complete the post-test) were analyzed for the assessments of scales. 2.2 Ethical Approval and accordance This study was approved by the Ethics Review Committee of Hangzhou Normal University Affiliated Hospital. The ethics committee that reviewed this study conducted an examination of the submitted documents based on relevant laws and regulations, the Helsinki Declaration and other ethical principles. After voting, they agreed to the relevant contents of the study design and approved the conduct of this research. The informed consent forms for publication have been obtained from all participants. 2.3 Experimental Materials This study adopted commonly used scales for assessing ASD symptoms. In addition, we also developed a scenario test to evaluate the intraverbal behavior in children with ASD. 2.3.1 Measures The Childhood Autism Rating Scale (CARS) is used to assess changes in children’s autism symptoms (Schopler et al., 1980 ), covering core domains such as interpersonal relationships, adaptability, sensory responses, and emotional-behavioral responses. It includes 15items, each rated in a 4-point Likert scale (1 = normal, 4 = severely abnormal). A higher total score indicates more severe autism symptoms. The Cronbach’s α coefficient for the scale in the present study was 0.823. The Autism Treatment Evaluation Checklist (ATEC) was developed by Rimland et al. (Rimland & Edelson, 1999 ), consisting of 77 items which were divided into 4 subscales: expressive/language communication (14 items), social ability (20 items), perceptual/cognitive ability (18 items), and health/physical/behavior (25 items). The total score ranges from 0 to 179, with higher scores indicating more severe ASD symptoms (Fang et al., 2019 ). The Cronbach’s α coefficient in this study was 0.927. The Basic Communication Behavior Assessment Scale for Children with Autism developed by Zhang and Wang ( 2005 ), was adopted to assess the development of basic communication behaviors in autistic children with no or limited verbal ability. It classifies communication behaviors into six dimensions: requesting, refusing, attracting attention, answering questions, requesting explanations, and social phrases, with a total of 90 observable behaviors.Only the 'answering questions' dimension was assessed in this study, which refers to the ability to provide verbal responses to causal questions (e.g., those beginning with 'why').A child’s language ability was evaluated based on a 4-point Likert scale (0 = never, 1 = occasionally, 2 = frequently, 3 = always). The Cronbach’s α coefficient in the present study was 0.743. The Parenting Stress Index-Short Form (PSI-SF) is a simplified version of the Parenting Stress Index developed through a series of exploratory factor analyses (Luo et al., 2021 ). It includes 36 items, grouped into three 12-item factors: Parental Distress (PD), Parent-Child Dysfunctional Interaction (PCDI), and Difficult Child (DC). Each item is rated on a 5-point Likert scale (1 = strongly disagree, 5 = strongly agree), with higher scores indicating more parenting stress. The Cronbach’s α coefficient in this study was 0.908. The Self-Rating Anxiety Scale (SAS) was employed to measure current anxiety symptoms (Zung, 1971 ). The scale consisted of 20 items.For each item, individuals were required to indicate the frequency with which they experience the symptom on a 4-point Likert scale (1 = none or a little of the time; 2 = some of the time; 3 = good part of the time; 4 = most or all of the time). The raw t otal score ranges from 20 to 80, which are then transformed to the standard score when multiplied by 1.25. Higher scores indicate higher anxiety levels. The Cronbach’s α coefficient in this study was 0.948. 2.3.2 Digital Game Materials The digital game consisted of three stages: understanding stage, learning stage, and practice stage. In the three stages,dialogue materials in the digital program were compiled, evaluated, and standardized by the researchers. Based on the analogy between trains and dialogues, different dialogue contents were designed, involving different life scenarios for children, such as school, family, and supermarket. Each scenario included 6 dialogues for training interactive conversations in children with ASD. The dialogue contents (e.g., “school”, “family”) were close to daily life, making it easy for children to understand. The alternative dialogue materials used in the three phases of this intervention study were all systematically compiled by the researchers based on common interaction scenarios in daily social settings. After compiling the materials, a standardized evaluation questionnaire was created using the Wenjuanxing platform. Ten postgraduate students majoring in psychology were recruited to form an evaluation team to assess the content. Following consistency testing, dialogue materials with an evaluation consistency coefficient of 80% or higher were finally selected and included in the formal intervention system. The learning videos in Phase 1 were recorded using train toys as the main experimental props to capture original images, which were then processed through standardized post-editing. The train image materials used in the connection training game were collected by photographing train toys to obtain original images, followed by removing irrelevant backgrounds and applying standardized cropping.The dialogue learning materials in Phase 2 were improved and rebuilt based on previously evaluated and approved basic materials. A multiple-choice question format was used for the standardized design of dialogue scenarios to help children learn and practice social dialogue strategies through selective learning and training. The real-scenario dialogue materials in Phase 3 also came from the previously evaluated and approved material library and were chosen based on the main interaction needs of real social situations. The intervention materials in the three stages presented a hierarchical characteristic of stepwise progression: Phase 1 focused on two core scenarios, namely schools and intervention institutions, including 12 segments of standardized dialogue content; Phase 2 expanded to family and supermarket scenarios on the basis of Phase 1, adding 12 new segments of dialogue materials (totaling 24 segments); Phase 3 further included hospital and amusement park scenarios, adding another 12 segments of dialogue materials (totaling 36 segments). All dialogue materials were targetedly designed around key interaction behaviors in social scenarios to ensure a high degree of adaptability between the content and children's daily life experiences. 2.3.3 Materials for the Scenario Test In addition, the researchers developed a scenario test to evaluate the children’s intraverbal behavior ability based on different dialogue scenarios (Watkins et al., 1989 ; Sundberg et al., 1990 ). The test included two parts: dialogue selection and real dialogue. In dialogue selection part, children with ASD were asked to choose the correct dialogue response from given options (e.g., Hello or Goodbye). In real dialogue, they were required to verbally respond to a daily dialogue. Each dialogue included five parts: greeting, mutual greeting, situational dialogue, topic shifting, and farewell, see Table 1 for an example of a dialogue. The dialogue materials were close to daily life, including scenarios such as school, family, and supermarket, with simple questions and answers suitable for children’s cognition. Table 1 Example of Intraverbal Behavior test materials Analogy: Different components of trains Dialogue phase Program dialogue Child’s response Locomotive greeting Program: “Hello!” C: _____ Power carriage mutual greeting P: “Did you have a good rest last night?” C: _____ Passenger carriage situational dialogue P: “I feel pretty good.” P: “Do you want to draw?” C: _____ The track converter topic shifting P: “That's great.” P: “By the way, what would you like for lunch? C: _____ The train tail farewell P: “OK. See you later!” C: _____ 2.4 Experimental Design and Procedures 2.4.1 Design and Implementation of the Digital Game Based on the idea of the book The Conversation Train ( learning conversations with trains )(Shaul, 2014 ), this study designed a digital game program to intervene in the intraverbal behavior of children with ASD. The digital program included three stages: (1) Understanding stage: Children with ASD learned the specific components of a complete dialogue and understood dialogue rules through the correspondence between different train carriages and different dialogue parts. That is, the locomotive represents the start of a dialogue (greetings, e.g., “Hello”); the power carriage promotes the continuation of the dialogue (mutual greetings, e.g., “How are you feeling today?”); the passenger carriage represents daily conversations (situational dialogue, e.g.,“What would you like for dinner?”); the track converter represents topic shifts (topic shifting, e.g.,“By the way, is there a painting class today?”) ; and the train tail represents the end of a dialogue (mutual farewells, e.g,“Goodbye”). (2) Learning stage: Children learned correct dialogue content through dialogue selection. (3) Practice stage: The ASD children’s real intraverbal behavior ability was trained. See Fig. 1 for the illustration. The intervention game was developed by professional programmers utilizing a Browser/Server (B/S) architecture. The front-end implementation employed HTML5, CSS3, and JavaScript for web development, while the back-end system used Spring Boot to develop RESTful APIs. The platform was hosted on a cloud server to ensure accessibility, with load balancing and HTTPS security encryption implemented through Nginx, and cross-environment consistency ensured by Docker containerization technology. The overall architecture supports horizontal expansion to meet performance requirements under high concurrency scenarios, while providing complete permission control and audit log functions, complying with the Level 2.0 / Level 3.0 security standards. The user interface of the digital game was shown in Fig. 1 . 2.4.2 Design and Process of the Intervention Experiment This study adopted a pre-test-post-test randomized controlled experimental design, with a 2 (group: experimental group, control group) × 2 (time: T0, T1) mixed factorial design. The dependent variables were scores of the five scales and the assessment score of intraverbal behavior ability of the participants. The research process included three stages: pre-test assessment, intervention, and post-test assessment. The pre-test assessment was conducted before the interventions (T0), including filling in the questionnaires by the children’ therapists and caregivers and completing the intraverbal behavior test by the children themselves. Rehabilitation therapists completed the following three scales: Childhood Autism Rating Scale (CARS) , Autism Treatment Evaluation Checklist (ATEC) , and Basic Communication Behavior Assessment Scale for Children with Autism . The primary caregivers (typically their mothers) of children with ASD completed the Parenting Stress Index-Short Form (PSI-SF) and Self-Rating Anxiety Scale (SAS) . In the pre-test, all ASD children underwent a test of the intraverbal behavior ability in quiet rooms of the ASD rehabilitation institutions.The test was administered by a trained researcher to a child, with the assistance of the child’s therapist. During the test, the researcher presented the test materials to the children on a tablet, including 5 dialogue selections (20 dialogues in total) and 6 real dialogues (30 dialogues in total). In the dialogue selection stage, children with ASD were required to select the correct response from options under the dialogue. Each dialogue selection was displayed for 10 seconds. After the dialogue was presented, the researcher asked the child to choose the response; if the child failed to choose within 10 seconds, it was considered an incorrect selection. For illiterate children, the researcher first read the dialogue content and options to them, then asked the child to choose and started timing. In the real dialogue stage, the researcher interacted with the child strictly according to the experimental materials. After the researcher spoke the dialogue, the child needed to respond within 5 seconds; if no response was made within 5 seconds, the researcher repeated the dialogue and waited another 5 seconds. A correct response within the given time was scored 1 point; incorrect or no response scored 0 point. During the test, the therapist also recorded the responses of the children, in order to measure inter-evaluator reliability. After the test, the records completed by the researchers and the therapists were compared to determine the consistency. Consistency was calculated by the following formula: (number of consistent records / total number of consistent and inconsistent records) × 100%. The inter-evaluator agreement for all participants in the study was over 90%. After the pre-test, the participants in the experimental group and the control group received different interventions. The participants in the experimental group received the digital intervention training in addition to the daily training in the ASD rehabilitation institutions. The digital game lasted approximately 15 minutes per day, with 5 days per week. In total, there were 30 times of digital game sessions, lasting about one and a half months. the participants in the control group received only daily training as usual in the institutions. After the participants in the experimental group completed the digital intervention training (T1), all the participants were evaluated again. The procedure of the post-test was the same as that in the pre-test. 2.5 Data analysis Due to the limited sample size and the non-normal distribution of the results obtained in this study, parametric analyses such as analysis of variance (ANOVA) were not appropriate. Therefore, a non-parametric approach was employed using a 2 (group: experimental, control) × 2 (time: T0, T1) mixed design, with the Mann-Whitney U test applied to compare differences in scores between groups across time points. Data are presented as median (Mdn), interquartile range (IQR), test statistic (U), and significance level (p). Analyses were conducted for scales、dialogue selection and authentic dialogue scores to confirm comparable baseline levels of verbal behavioral ability in children with autism spectrum disorder (ASD), thereby establishing a foundation for subsequent intervention. 3 Results 3.1 Results of Scales The mean total scale scores at the pre-test and post-test in the two groups were shown in Table 2. EXP CTL T0 T1 T0 T1 PSI 99.17(18.57) 101.00(17.48) 104.17(20.12) 106.19(24.22) SAS 40.50(9.38) 40.43(6.56) 43.42(8.26) 44.23(6.26) ATEC 100.3(29.08) 76.78(28.08) 104.56(23.27) 98.88(42.64) CARS 28.28(6.99) 30.22(9.64) 30.28(7.47) 32.31(11.35) CB 11.06(7.27) 15.67(6.32) 10.72(7.84) 10.81(5.59) (Note: EXP=experimental group;CTL=control group;CARS=childhood autism rating scale;ATEC=autism treatment evaluation checklist;PSI=parenting stress index;SAS=self-rating anxiety scale;CB=communication behavior assessment scale for children with autism.) 3.1.1 Baseline comparison between the two groups Mann-Whitney U tests on pre-test scale data in the two groups showed no significant differences on scores of PSI (U = 134.5, p = 0.389), SAS (U = 126.0, p = 0.265), ATEC (U = 159.0, p = 0.938), CARS (U = 135.0, p = 0.406), and communication behavior scale (U = 156, p = 0.864) between the experimental group and control group, indicating that the two groups were comparable at the baseline. 3.1.2 Comparison of pre-test and post-test in the two groups For the experimental group, after receiving the digital training intervention in addition to the daily intervention, the post-test score of ATEC (Mdn = 83, IQR = 51–99) was significantly lower (Z = -3.551, p < 0.001, effect size r = 0.59) than the pre-test score (Mdn = 108, IQR = 72–127). For the control group, there was no significant difference (Z = -0.776, p = 0.438, effect size r = 0.13) between the pre-test score (Mdn = 104, IQR = 95–116) and the post-test ATEC score (Mdn = 99, IQR = 71–141). These results indicated that the introduction of the digital training significantly reduced the severity of ASD symptoms in the experimental group. See Fig. 2 for the illustration. For the communication behavior scale scores, there was a significant difference (Z = -2.943, p = 0.003, effect size r = 0.49) between the pre-test score (Mdn = 9, IQR = 8–16) and post-test score (Mdn = 16, IQR = 11–20) in the experimental group. In the control group, no significant difference (Z = -0.647, p = 0.517, effect size r = 0.01) between the pre-test score (Mdn = 10, IQR = 7–13) and the post-test score (Mdn = 9, IQR = 7–15) was observed. See Fig. 3 for the illustration. For the PSI scale scores, there was no significant difference (Z = -0.501, p = 0.616, effect size r = 0.08) between the pre-test score (Mdn = 99, IQR = 84–119) and the post-test score (Mdn = 100, IQR = 87–116) in the experimental group. In the control group, no significant difference (Z = -0.028, p = 0.977, effect size r = 0.004) was found between the pre-test score (Mdn = 108, IQR = 89–119) and the post-test score (Mdn = 112, IQR = 97–123). For the SAS scale score, no significant difference (Z = -0.207, p = 0.836, effect size r = 0.03) was found between the pre-test score (Mdn = 39, IQR = 35–44) and the post-test score (Mdn = 39, IQR = 36–45) in the experimental group. In the control group, there was also no significant difference (Z = -0.313, p = 0.754, effect size r = 0.05) between the pre-test score (Mdn = 44, IQR = 36–48) and the post-test score (Mdn = 44, IQR = 41–49) . For the CARS scale score, no significant difference (Z = -0.825, p = 0.410, effect size r = 0.14) was found between the pre-test score (Mdn = 29, IQR = 22–32) and the post-test score (Mdn = 31, IQR = 22–38) in the experimental group. In the control group, there was no significant difference (Z = -0.906, p = 0.365, effect size r = 0.16) between the pre-test score (Mdn = 31, IQR = 24–34) and the post-test score (Mdn = 30, IQR = 26–44) scores. 3.2 Results of the Intraverbal Behavior Ability Test Table 3 showed the mean scores in the intraverbal behavior ability test in the experimental group and the control group. Table 3 Mean scores of the intraverbal behavior assessment (Mean (SD)) EXP CTL T0 T1 T0 T1 Dialogue selection 10.78(5.23) 18.00(2.63) 13.28(3.31) 12.71(5.76) Real intraverbal behavior 14.33(9.71) 26.50(4.25) 21.57(6.32) 18.70(9.35) (Note: EXP=experimental group;CTL=control group.) 3.2.1 Baseline comparison between the two groups The Mann-Whitney U test showed no significant difference in pre-test scores between the experimental group and the control group. Specifically, there was no significant difference in pre-test scores of dialogue selection between the two groups (U = 90.00, p = 0.180), and no significant difference in pre-test scores of real dialogue between the two groups (U = 142.00, p = 0.276). Therefore, the two groups were comparable at the baseline. 3.2.2 Comparison of pre-test and post-test in the two groups Results showed that the post-test score of dialogue selection (Mdn = 19, IQR = 17–20) was significantly higher than the pre-test score (Mdn = 10, IQR = 6–15) in the experimental group (Z = -4.097, p = 0.001, effect size r = 0.97); the post-test score of real dialogue (Mdn = 29, IQR = 24–30) was also significantly higher than the pre-test score (Mdn = 13, IQR = 7–24) in the experimental group (Wilcoxon test: Z = -3.758, p = 0.001, effect size r = 0.89). An effect size r > 0.5 indicates that the intervention effect has moderate or higher practical significance, suggesting that the digital game significantly improved the intraverbal behavior ability in children with ASD. See Fig. 4 for the illustration. The control group received only daily intervention without the digital game training. There was no significant difference between the post-test score (Mdn = 11, IQR = 9–20) and pre-test score (Mdn = 14, IQR = 10–16) of dialogue selection (Z = -0.392, p = 0.701, effect size r = 0.10); Neither significant difference was found between the post-test score (Mdn = 17, IQR = 7–28) and pre-test score (Mdn = 22, IQR = 14–27) of real dialogue (Z = -0.991, p = 0.329, effect size r = 0.26) in the control group. 4 Discussion This study demonstrated that the digital game intervention combined with conventional training can significantly improve the intraverbal behavior of children with ASD .The results showed that the participants in the experimental group (contrary to the control group) had significantly reduced ATEC scores and increased CB scores. More importantly, The test of intraverbal behavior ability further showed that after the interventions, the correct response rates of the participants in the experimental group in dialogue selection and real dialogue increased by 67% and 85% respectively, which were significantly better than those of the control group. These results were consistent with the conclusion of the systematic review that digital game intervention can improve social skills in children with ASD (Gao et al., 2025 ), and also in line with the meta-analysis finding that structured digital intervention is more targeted than traditional therapy (Wang et al., 2025 ). Specifically, the significant reduction in ATEC scores in the experimental group suggested that digital game intervention may effectively alleviate the core symptoms of ASD through structured interaction. This effect may stem from the intervention program’s structured and visual decomposition of abstract social dialogue rules. The game in our study used different carriages to represent different components of dialogue, and designed three stages (i.e.,"understanding-learning-practice") to facilitate the neuroplasticity. In addition, the real-time feedback in the game (e.g., "on track / off track" prompts) strengthens the learning acquisition of correct dialogue patterns, which is consistent with the mechanism that social interaction training can activate the mirror neuron system in children with ASD (Dapretto et al., 2006 ). In contrast, there was no significant change in ATEC scores in the control group, which aligns with the view that conventional intervention has limited short-term improvement effects on language ability in ASD children (Reichow et al., 2018 ). The significant increase in CB scores in the experimental group indicates that digital game intervention significantly improved the actual communication behavior in children with ASD. This result echoes the reduction in the ATEC score, indicating that the digital intervention not only alleviated core symptoms but also directly promoted the functional improvement of intraverbal behavior ability. This may be attributed to the design of the real dialogue practice module in the digital program, which provides structured language application opportunities through scenario simulation andreal-time feedback —a "learning by doing" model that essentially incorporates the embodiment theory in digital intervention (Kumar, 2025 ). Mechanistically, the improvement in CB scores may be related to the cognitive scaffolding role of the train carriage metaphor in the digital game. The game decomposes dialogue structures into visualized carriage units, helping children with ASD transform abstract social rules into concrete operations. This "visual-semantic mapping" intervention strategy is similar to the principle of traditional social story therapy (Thiemann & Goldstein, 2001 ; Zimmerman et al., 2020 ). It is worth noting that, the progress of the participants in the experimental group in dialogue selection and real dialogue provided more direct and specific evidences in support of the enhanced ability of the ASD children to transfer training content to actual expression. This outcome might stem from the vivid design of the digital intervention, where train-themed visual symbols transform abstract conversational structures into tangible visual representations. Previous studies have shown that immersive visualization interfaces can reduce social cognitive load in children with ASD (Bacchetta et al., 2024). Such design enables children to rapidly internalize the logical sequence required for conversations, laying the foundation for skill transfer. In addition, the reduced standard deviation of scores in the experimental group indicates that the digital intervention has good effects on children with different ability levels, providing a basis for clinical promotion.This finding supports the view that language intervention in ASD children should incorporate repetitive reinforcement task design (Reichow et al., 2018 ) and highlights the advantages of digital game intervention in standardized training. The parental SAS scores in the experimental group showed a modest numerical change (T0 = 40.50 (9.38), T1 = 40.43 (6.56)) but this difference did not reach statistical significance (Z = -0.207, p = 0.836). Despite the lack of statistical significance, this subtle trend may reflect potential indirect benefits of the digital intervention, which could be associated with the significant improvement in children’s communication abilities (evidenced by the increased Communication Behavior Scale score). When parents observed specific and visible progress in their children’s intraverbal behavior—particularly active expression in real dialogue—it may have alleviated their concerns about their children’s social communication disorders and enhanced their confidence and self-efficacy in supporting interactive communication (Karst & Van Hecke, 2012 ), aligning with the "child behavior-parental well-being" bidirectional influence model (Callanan et al., 2021 ). The absence of statistical significance for SAS scores is consistent with the non-significant change in PSI scores, likely due to these scales measuring broader, more persistent aspects of parental stress and anxiety (e.g., long-term caregiving burdens, economic pressures) that are less responsive to short-term interventions targeting specific child communication skills. Additionally, individual differences in parental stress responses and the relatively small sample size may have contributed to the lack of measurable statistical effects, despite the observed directional trend. The results of this study highlight the great potential of digital therapy, especially gamified programs based on clear theoretical frameworks, in language intervention for children with ASD. This study provided an innovative idea: utilizing the visual learning advantages and intrinsic motivation for games common in children with ASD (Grynszpan et al., 2014 ) to transform abstract social language rules into concrete, structured, and repeatedly practicable digital tasks. This study proved that the digital gamified intervention according to the "conversation train" model, which combines dialogue decomposition, rules learning, and active verbal output in three stages, was effective and has significant advantages over traditional methods. Its browser/server architecture-based digital platform ensures the usability of the program. The high standardization of study content, the gradual difficulty setting, and the real-time feedback mechanisms, made the learning progress smoothly and reduced potential confoundings caused by trainers or the environment (Spain & Blainey, 2015 ). Moreover, the program supported multi-terminal access, allowing children to use it flexibly in different places, such as home and school, which overcomes the limitations of locations and resources of therapy, thus greatly improves the accessibility. The convenient use in family environments also enables parents to participate in or assist their children’s training, and the observable progress would eventually help them alleviate the anxiety and enhance their confidence, thus forming a more supportive intervention environment. The gamified intervention demonstrated significant improvements in intraverbal behavior among children with ASD. Clinically, this approach offers three key advantages: (1) standardized protocols reduce therapist dependency; (2) flexible implementation across home/school settings; (3) real-time feedback enhances engagement. For policy makers, these findings support integrating digital tools into standard ASD intervention frameworks, potentially reducing healthcare disparities in resource-limited regions. This study also has some limitations. First, the sample size in our study (18 ASD children in the experimental group, 16 in the control group) is relatively small. Future research enrolling more children with ASD is needed to further demonstrate our findings. (2) We only conducted the pre-test and post-test to explore the efficacy of the digital intervention. Future studies should conduct follow-up assessments (e.g., 3 months, 6 months) to investigate the long-term effect of the digital intervention.(3) The study only used scales and behavioral tests, which are subjective measurements and might be influenced by the reporters’ memory bias and raters’ evaluation bias. In the future, it is necessary to utilize multimodal objective measurement, such as eye-tracking technology and multi-channel physiological signal recording, to evaluate the effect of digital intervention objectively and also to understand the associated psychological and physiological mechanism. 5 Conclusion This study developed a digital game intervention program based on the idea of Learning Conversations with Trains , and designed a randomized controlled study to explore the effect of the digital intervention in children with ASD. The results demonstrated that the digital intervention can effectively improve the abilities of intraverbal behavior and core symptoms (as indexed by ATEC and CB scores) in ASD children and enhance the interactive and conversational skills of children with autism. The findings in this study provide empirical evidence for the efficacy of digital therapy in language intervention in children with ASD. It also have important implications for the new intervention strategies in children with ASD. Declarations Funding: Research on Key Technologies of Brain-Computer Digital Regulation for Children with ADHD and Autism under the "Sharpshooter" Research and Development Program of Zhejiang Province(2023C03002) Conflicts of Interest: All authors declare no potential conflicts of interest. Ethics Approval: The study was approved.(Approval No.: 2024(E2)-HS-013]). Consent: Written informed consent was obtained from all participants’ parents or legal guardians. Materials and/or Code Availability: The digital game intervention materials and code are available from the corresponding author upon reasonable request. Author Contribution Conceptualization: Zhiguo Hu; Methodology: Yingjian Zhang; Data Collection: Yingjian Zhang; Writing: Yingjian Zhang and Huixin Lu”] Data Availability The digital game intervention materials and code are available from the corresponding author upon reasonable request. References American Psychiatric Association. (2013). Diagnostic and statistical manual of mental disorders (DSM-5®) . Washington, DC: Author. Ameis, S. H., Kassee, C., Corbett‐Dick, P., Cole, L., Dadhwal, S., Lai, M. C., ... & Correll, C. U. (2018). Systematic review and guide to management of core and psychiatric symptoms in youth with autism. Acta Psychiatrica Scandinavica , 138(5), 379-400. ALMåS, I. K., Smith, D. P., Eldevik, S., et al. (2022). Emergent intraverbal and reverse intraverbal behavior following listener training in children with autism spectrum disorder. Analysis of Verbal Behavior, 38(1), 1–23. Bekele, E., Zheng, Z., Swanson, A., Crittendon, J., Warren, Z., & Sarkar, N. (2013). Understanding how adolescents with autism respond to facial expressions in virtual reality environments. IEEE transactions on visualization and computer graphics , 19(4), 711-720. Brignell, A., Chenausky, K. V., Song, H., Zhu, J., Suo, C., & Morgan, A. T. (2018). Communication interventions for autism spectrum disorder in minimally verbal children. Cochrane Database of Systematic Reviews , (11). Callanan, J., Signal, T., & McAdie, T. (2021). What is my child telling me? Reducing stress, increasing competence and improving psychological well-being in parents of children with a developmental disability. Research in developmental disabilities , 114, 103984. Chen, G. X., & Yang, X. J. (2014). A study on conversational ability of autistic children. Chinese Journal of Special Education , 11, 45-52. (In Chinese) Cui, M., Ni, Q., & Wang, Q. (2023). Review of intervention methods for language and communication disorders in children with autism spectrum disorders. PeerJ , 11, e15735. Dapretto, M., Davies, M. S., Pfeifer, J. H., Scott, A. A., Sigman, M., Bookheimer, S. Y., & Iacoboni, M. (2006). Understanding emotions in others: mirror neuron dysfunction in children with autism spectrum disorders. Nature neuroscience , 9(1), 28-30. Dawson, G., Rogers, S., Munson, J., Smith, M., Winter, J., Greenson, J., ... & Varley, J. (2010). Randomized, controlled trial of an intervention for toddlers with autism: the Early Start Denver Model. Pediatrics , 125(1), e17-e23. Fang, H., Yanling, R., Li, C., & Xiaoyan, K. (2019). Reliability and validity of the Chinese version of autism treatment evaluation checklist. Sichuan Mental Health , 32(6), 518. (In Chinese) Jespersen, A. E., Lumbye, A., Vinberg, M., Glenthøj, L., Nordentoft, M., Wæhrens, E. E., ... & Miskowiak, K. W. (2024). Effect of immersive virtual reality-based cognitive remediation in patients with mood or psychosis spectrum disorders: study protocol for a randomized, controlled, double-blinded trial. Trials , 25 (1), 82. Gao, J., Song, W., Huang, D., Zhang, A., & Ke, X. (2025). The effect of game-based interventions on children and adolescents with autism spectrum disorder: A systematic review and meta-analysis. Frontiers in Pediatrics , 13, 1498563. Grynszpan, O., Weiss, P. L., Perez-Diaz, F., & Gal, E. (2014). Innovative technology-based interventions for autism spectrum disorders: a meta-analysis. Autism , 18(4), 346-361. Karst, J. S., & Van Hecke, A. V. (2012). Parent and family impact of autism spectrum disorders: A review and proposed model for intervention evaluation. Clinical child and family psychology review , 15, 247-277. Kumar, A. (2025). Unique models of embodied cognition and eco-social niches proposed to validate hypothesis of social attunement and mis-attunement with a focus on autism. Frontiers in Psychiatry , 16, 1562061. LaLonde, K. B., Jones, S., West, L., & Santman, C. (2020). An evaluation of a game-based treatment package on intraverbals in young children with autism. Behavior Analysis in Practice , 13, 152-157. Luo, J., Wang, M. C., Gao, Y., Zeng, H., Yang, W., Chen, W., ... & Qi, S. (2021). Refining the parenting stress index–short form (PSI-SF) in Chinese parents. Assessment , 28(2), 551-566. Moraleda-Sepulveda, E., Pulido-García, N., Loro-Vicente, N., & Santos-Muriel, N. (2025). Effectiveness of Intensive Linguistic Intervention in Autism Spectrum Disorder: A Case Study. Children , 12(2), 182. Panceri, J. A. C., Freitas, É., de Souza, J. C., da Luz Schreider, S., Caldeira, E., & Bastos, T. F. (2021). A new socially assistive robot with integrated serious games for therapies with children with autism spectrum disorder and down syndrome: a pilot study. Sensors , 21(24), 8414. Peretti, S., Pino, M. C., Caruso, F., & Di Mascio, T. (2024). evaluating the potential of immersive virtual reality-based serious games interventions for autism: A pocket guide evaluation framework. Education Sciences , 14 (4), 377. Reichow, B., Barton, E. E., Boyd, B. A., & Hume, K. (2018). Early intensive behavioral intervention (EIBI) for young children with autism spectrum disorders (ASD). Cochrane database of systematic reviews , (10). Rezayi, S., Shahmoradi, L., & Tehrani-Doost, M. (2025). Systematic Review and Thematic Analysis of Digital Games for Cognitive Enhancement in Children with Autism Spectrum Disorder: Toward a Conceptual Framework. Cognitive Computation , 17(1), 60. Rimland, B., & Edelson, S. M. (1999). Autism treatment evaluation checklist. Journal of Intellectual Disability Research . Schopler, E., Reichler, R. J., DeVellis, R. F., & Daly, K. (1980). Toward objective classification of childhood autism: Childhood Autism Rating Scale (CARS). Journal of autism and developmental disorders . Shaul, J. (2014). The Conversation Train: A Visual Approach to Conversation for Children on the Autism Spectrum . Jessica Kingsley Publishers. Smith, T. (2001). Discrete trial training in the treatment of autism. Focus on autism and other developmental disabilities , 16(2), 86-92. Spain, D., & Blainey, S. H. (2015). Group social skills interventions for adults with high-functioning autism spectrum disorders: A systematic review. Autism , 19(7), 874-886. Shaw, K. A. (2025). Prevalence and early identification of autism spectrum disorder among children aged 4 and 8 years—Autism and Developmental Disabilities Monitoring Network, 16 Sites, United States, 2022. MMWR. Surveillance Summaries , 74. Sundberg, M. L., Juan, B. S., Dawdy, M., & Argüelles, M. (1990). The acquisition of tacts, mands, and intraverbals by individuals with traumatic brain injury. The Analysis of verbal behavior , 8, 83–99. Thiemann, K. S., & Goldstein, H. (2001). Social stories, written text cues, and video feedback: Effects on social communication of children with autism. Journal of applied behavior analysis , 34(4), 425-446. Virues-Ortega, J., Pérez-Bustamante, A., & Tarifa-Rodriguez, A. (2022). Evidence-based Applied Behavior Analysis (ABA) autism treatments: An overview of comprehensive and focused meta-analyses. Handbook of autism and pervasive developmental disorder: Assessment, diagnosis, and treatment , 631-659. Watkins, C. L., Pack-Teixeira, L., & Howard, J. S. (1989). Teaching intraverbal behavior to severely retarded children. The Analysis of verbal behavior , 7, 69–81. Wang, T., Ma, Y., Du, X., Li, C., Peng, Z., Wang, Y., & Zhou, H. (2024). Digital interventions for autism spectrum disorders: A systematic review and meta‐analysis. Pediatric Investigation , 8(3), 224-236. Wang, T., Ma, P. H., Ge, H., Sun, Y., Kwok, T. T. O., Liu, X., ... & Zhang, W. (2025). The use of gamified interventions to enhance social interaction and communication among people with autism spectrum disorder: A systematic review and meta-analysis. International Journal of Nursing Studies , 105037. Xu, F., Gage, N., Zeng, S., Zhang, M., Iun, A., O’Riordan, M., & Kim, E. (2024). The use of digital interventions for children and adolescents with autism Spectrum disorder—a Meta-analysis. Journal of Autism and Developmental Disorders , 1-17. Zhang, M., Ding, H., Naumceska, M., & Zhang, Y. (2022). Virtual reality technology as an educational and intervention tool for children with autism spectrum disorder: current perspectives and future directions. Behavioral Sciences , 12(5), 138. Zhang, Z. F., & Wang, H. P. (2005). Development of the Autism Children's Behavior Checklist and related research. Journal of Special Education Research , 28, 145-166. (In Chinese) Zimmerman, K. N., Ledford, J. R., Gagnon, K. L., & Martin, J. L. (2020). Social stories and visual supports interventions for students at risk for emotional and behavioral disorders. Behavioral Disorders , 45(4), 207-223. Zung, W. W. (1971). A rating instrument for anxiety disorders. Psychosomatics: Journal of Consultation and Liaison Psychiatry . Zeidan, J., Fombonne, E., Scorah, J., Ibrahim, A., Durkin, M. S., Saxena, S., ... & Elsabbagh, M. (2022). Global prevalence of autism: A systematic review update. Autism research , 15(5), 778-790. Additional Declarations No competing interests reported. Cite Share Download PDF Status: Posted Version 1 posted You are reading this latest preprint version Research Square lets you share your work early, gain feedback from the community, and start making changes to your manuscript prior to peer review in a journal. As a division of Research Square Company, we’re committed to making research communication faster, fairer, and more useful. We do this by developing innovative software and high quality services for the global research community. Our growing team is made up of researchers and industry professionals working together to solve the most critical problems facing scientific publishing. Also discoverable on Platform About Our Team In Review Editorial Policies Advisory Board Help Center Resources Author Services Accessibility API Access RSS feed Manage Cookie Preferences © Research Square 2026 | ISSN 2693-5015 (online) Privacy Policy Terms of Service Do Not Sell My Personal Information {"props":{"pageProps":{"initialData":{"identity":"rs-8504009","acceptedTermsAndConditions":true,"allowDirectSubmit":true,"archivedVersions":[],"articleType":"Research Article","associatedPublications":[],"authors":[{"id":592988013,"identity":"cba5e709-e1e4-4fe4-911b-bd1611af1e33","order_by":0,"name":"Yingjian Zhang","email":"","orcid":"","institution":"Hangzhou Normal University","correspondingAuthor":false,"prefix":"","firstName":"Yingjian","middleName":"","lastName":"Zhang","suffix":""},{"id":592988018,"identity":"614e6ea0-9709-4f89-9065-1896da2e734e","order_by":1,"name":"Huixin Lu","email":"","orcid":"","institution":"Hangzhou Normal University","correspondingAuthor":false,"prefix":"","firstName":"Huixin","middleName":"","lastName":"Lu","suffix":""},{"id":592988019,"identity":"9abc77c8-f648-48ec-9d16-1d9676eb4782","order_by":2,"name":"Zhiguo Hu","email":"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAZAAAAAyAQMAAABI0h/eAAAABlBMVEX///8AAABVwtN+AAAACXBIWXMAAA7EAAAOxAGVKw4bAAAAuElEQVRIiWNgGAWjYBACAxDxsQHCkSBaC+NMkrUw85KkxVwi/fFr2x11eQYHmA/e5mGwyyOoxXJGQpp17hm2YoMDbMnWPAzJxYQddiPhmHFuG0/ihgM8ZtI8DAcSGwhrSWwztmyTAGrh/0aslmTmx4xtBiBb2IjUcuYZG2NvW0LizMNsxpZzDJKJ0HI8/fGHn211iX3Hmx/eeFNhR1gLELBBooMZbAIR6kFqPxCnbhSMglEwCkYsAAD9jDuhhILo/QAAAABJRU5ErkJggg==","orcid":"","institution":"Hangzhou Normal University","correspondingAuthor":true,"prefix":"","firstName":"Zhiguo","middleName":"","lastName":"Hu","suffix":""}],"badges":[],"createdAt":"2026-01-03 02:53:24","currentVersionCode":1,"declarations":"","doi":"10.21203/rs.3.rs-8504009/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-8504009/v1","draftVersion":[],"editorialEvents":[],"editorialNote":"","failedWorkflow":false,"files":[{"id":102982075,"identity":"e2210b6b-20a3-4736-86b2-b6e6446187b5","added_by":"auto","created_at":"2026-02-19 09:12:15","extension":"png","order_by":1,"title":"Figure 1","display":"","copyAsset":false,"role":"figure","size":268689,"visible":true,"origin":"","legend":"\u003cp\u003eTraining steps of digital intervention program\u003c/p\u003e","description":"","filename":"1.png","url":"https://assets-eu.researchsquare.com/files/rs-8504009/v1/ca0570589503c7c3d20a1c36.png"},{"id":102982102,"identity":"3541bf18-2e28-403d-846e-ad3b87412851","added_by":"auto","created_at":"2026-02-19 09:12:21","extension":"png","order_by":2,"title":"Figure 2","display":"","copyAsset":false,"role":"figure","size":23049,"visible":true,"origin":"","legend":"\u003cp\u003eResults of the ATEC scale score . Error bars show standard errors (SE) of the mean. *** \u003cem\u003ep\u003c/em\u003e \u0026lt; 0.001\u003c/p\u003e","description":"","filename":"2.png","url":"https://assets-eu.researchsquare.com/files/rs-8504009/v1/51b2eed0ae6e0818df22ef37.png"},{"id":102982103,"identity":"8338ef09-fe43-4fe1-9286-6c6398ad439b","added_by":"auto","created_at":"2026-02-19 09:12:21","extension":"png","order_by":3,"title":"Figure 3","display":"","copyAsset":false,"role":"figure","size":21127,"visible":true,"origin":"","legend":"\u003cp\u003eResults of the communication behavior scale score . Error bars show standard errors (SE) of the mean. ** \u003cem\u003ep\u003c/em\u003e \u0026lt; 0.01\u003c/p\u003e","description":"","filename":"3.png","url":"https://assets-eu.researchsquare.com/files/rs-8504009/v1/e8beea81e287c8b4b38a09bf.png"},{"id":102982098,"identity":"e6ff69f0-c278-4d2b-8235-56e2d4d6e346","added_by":"auto","created_at":"2026-02-19 09:12:20","extension":"png","order_by":4,"title":"Figure 4","display":"","copyAsset":false,"role":"figure","size":26849,"visible":true,"origin":"","legend":"\u003cp\u003eResults of the intraverbal behavior assessment between the two groups. Error bars show standard errors (SE) of the mean. ** \u003cem\u003ep\u003c/em\u003e \u0026lt; 0.01\u003c/p\u003e","description":"","filename":"4.png","url":"https://assets-eu.researchsquare.com/files/rs-8504009/v1/b75acf0f3dab0f9dab3270f4.png"},{"id":105931056,"identity":"8e0c50d0-df42-42a7-bb1e-2d3092c26f65","added_by":"auto","created_at":"2026-04-01 14:12:28","extension":"pdf","order_by":0,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":1127816,"visible":true,"origin":"","legend":"","description":"","filename":"manuscript.pdf","url":"https://assets-eu.researchsquare.com/files/rs-8504009/v1/3f312934-d3a9-40cd-bd26-9984d0f20715.pdf"}],"financialInterests":"No competing interests reported.","formattedTitle":"Digital Intervention Training for Intraverbal Behavior in Children with Autism Spectrum Disorder","fulltext":[{"header":" Highlights","content":"\u003cp\u003e• Developed Conversation Train-based digital game for ASD children (n=40)\u003cbr\u003e\u0026nbsp;• Improved dialogue selection (+67%) and real conversation skills (+85%)\u003cbr\u003e\u0026nbsp;• Reduced ATEC scores (Z=-3.551, p\u0026lt;0.001) vs control group\u003cbr\u003e\u0026nbsp;• B/S architecture enables multi-device accessibility\u003c/p\u003e"},{"header":"1 Introduction","content":"\u003cp\u003eAutism spectrum disorder (ASD) is a neurodevelopmental disorder characterized by persistent deficits in social communication, repetitive and stereotyped behaviors, and restricted interests (American Psychiatric Association, \u003cspan citationid=\"CR1\" class=\"CitationRef\"\u003e2013\u003c/span\u003e). In recent years, the global prevalence of ASD has increased significantly (Zeidan et al., \u003cspan citationid=\"CR41\" class=\"CitationRef\"\u003e2022\u003c/span\u003e; Ameis, 2018). The latest epidemiological surveys showed that its prevalence has reached 1%\u0026ndash;2%, severely affecting children\u0026rsquo;s social adaptation and qualities of life (Shaw et al., 2025). Among the core symptoms of ASD, intraverbal behavior impairments are particularly prominent, which are manifested by difficulties in initiating conversations, insufficient topic maintenance, and deficits in nonverbal communication.These impairments directly hinder social connections with others and have become key factors affecting the functional prognosis of children with ASD (Cui \u0026amp; Wang, 2023; Moraleda-Sepulveda, 2025).\u003c/p\u003e \u003cp\u003eThe term \"intraverbal behavior\" first appeared in Skinner\u0026rsquo;s \u003cem\u003eVerbal Behavior\u003c/em\u003e (Skinner, 1957), which was classified into six main functional categories: mand, tact, echoic, intraverbal, textual, and transcription. Children with ASD exhibit various degrees of deficits in intraverbal behavior, including inability to understand turn-taking rules in conversations, difficulty in adjusting language content according to context, and lack of ability to actively maintain topics (ALM\u0026aring;S et al., \u003cspan citationid=\"CR3\" class=\"CitationRef\"\u003e2022\u003c/span\u003e; Chen \u0026amp; Yang, \u003cspan citationid=\"CR7\" class=\"CitationRef\"\u003e2014\u003c/span\u003e). Notably, as the core of social communication, the lack of such ability further leads to secondary problems such as peer interaction difficulties and family interaction pressure for children with ASD (Callanan, 2021).\u003c/p\u003e \u003cp\u003eFor a long time, traditional offline interventions, such as applied behavior analysis and speech-language therapy, have been the mainstream of language intervention for ASD, with their effectiveness partially verified (Brignell et al., \u003cspan citationid=\"CR5\" class=\"CitationRef\"\u003e2018\u003c/span\u003e; Smith, \u003cspan citationid=\"CR27\" class=\"CitationRef\"\u003e2001\u003c/span\u003e; Virues-Ortega et al., \u003cspan citationid=\"CR32\" class=\"CitationRef\"\u003e2022\u003c/span\u003e). However, these intervention strategies have the following limitations: (1) the shortage of professional therapists make such intervention services less accessible ;(2) high costs of such face-to-face interventions are great financial challenges most families face (Bekele et al., \u003cspan citationid=\"CR4\" class=\"CitationRef\"\u003e2013\u003c/span\u003e); (3) the standardization of offline interventions is insufficient, and differences in therapists\u0026rsquo; skills and operations may affect the stability of intervention effects (Dawson, 2010); (4) traditional interventions mostly adopt structured tabletop training, lacking skill transferring to natural scenarios, which fails to meet the needs in real interactive contexts for children with ASD (Panceri, 2021).\u003c/p\u003e \u003cp\u003eWith the development of digital technology, digital interventions based on serious game and visualization have gradually become a research hotspot in the field of ASD (Zhang et al., \u003cspan citationid=\"CR37\" class=\"CitationRef\"\u003e2022\u003c/span\u003e). Digital interventions, with their advantages of repeatability, standardized operation, and cross-scenario application, provide new ideas for addressing the pain points of traditional interventions (LaLonde et al., \u003cspan citationid=\"CR17\" class=\"CitationRef\"\u003e2020\u003c/span\u003e). Systematic reviews have shown that digital interventions have significant effects on improving social attention (Xu et al., \u003cspan citationid=\"CR36\" class=\"CitationRef\"\u003e2024\u003c/span\u003e) and IQ (Wang et al., \u003cspan citationid=\"CR34\" class=\"CitationRef\"\u003e2024\u003c/span\u003e) in children with ASD, and their gamification design can effectively trigger children\u0026rsquo;s intrinsic motivation and extend intervention participation time (Rezayi et al., \u003cspan citationid=\"CR23\" class=\"CitationRef\"\u003e2025\u003c/span\u003e). However, existing studies utilizing digital interventions mostly focus on training of the general social skills such as cognition and emotion recognition in children with ASD. To our knowledge, no study to date has used digital game to intervene in the intraverbal behavior in children with ASD.\u003c/p\u003e \u003cp\u003eTo address this research gap, the present study aimed to develop a digital game-based intervention program grounded in dialogue structure modeling and conduct a randomized controlled trial. The game was designed based on the idea of \u003cem\u003eLearning Conversations with Trains\u003c/em\u003e (\u003cem\u003eThe Conversation Train\u003c/em\u003e)(Shaul, \u003cspan citationid=\"CR26\" class=\"CitationRef\"\u003e2014\u003c/span\u003e), utilizing the potential interest in trains of children with ASD. Specifically, the locomotive, power carriage, passenger carriage, track converter, and train tail are analogized to greetings, dialogue-promoting utterances, conversational turns, topic shifts, and conversation endings in a dialogue, respectively. Through analogical reasoning methods, dialogue rules were structured and visualized, in order to improve the abilities of intraverbal behavior in children with ASD.We hypothesize that compared with conventional intervention, the intervention model incorporating a digital game could be more effective in improving the intraverbal behaviors abilities in children with ASD.\u003c/p\u003e"},{"header":"2 Methods","content":"\u003cdiv id=\"Sec3\" class=\"Section2\"\u003e \u003ch2\u003e2.1 Participants\u003c/h2\u003e \u003cp\u003eA total of 40 children with ASD aged 5\u0026ndash;10 years were recruited and randomly assigned to the experimental group and the control group. The inclusion criteria were as follows: (1) Children aged 5\u0026ndash;10 years; (2) A diagnosis of ASD by a professional clinician; (3) Having intraverbal behavior deficits reported by rehabilitation therapists; (4) Having moderate-to-high functioning as evaluated by the therapist.The exclusion criteria included: (1) children with severe ASD; (2) those with other neurodevelopmental disorders or physical disabilities or chronic conditions; (3) children with a family history of other genetic disorders; (4) those receiving medication currently.\u003c/p\u003e \u003cp\u003e All the parents or legal guardians of the participants have signed the written informed consent form.\u003c/p\u003e \u003cp\u003eThe participants in the control group received only daily training, while the participants in the experimental group received both daily training and digital intervention. In the experimental group, two participants withdrew due to illness, remaining 18 participants who finally completed the post-test. In the control group, four participants did not complete the routine intervention and dropped out.Another two participants in the control group finished the daily training but did not take part in the post-test; however, their rehabilitation therapists completed the assessment scales of these two participants. Eventually, fourteen participants\u0026rsquo; full data in the control group were collected. To compensate for the relatively small sample size of the control group, 16 data (including the data from interactive dialogue with the two children who did not complete the post-test) were analyzed for the assessments of scales.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003e2.2 Ethical Approval and accordance\u003c/strong\u003e \u003cp\u003e\u003cp\u003e This study was approved by the Ethics Review Committee of Hangzhou Normal University Affiliated Hospital. The ethics committee that reviewed this study conducted an examination of the submitted documents based on relevant laws and regulations, the Helsinki Declaration and other ethical principles. After voting, they agreed to the relevant contents of the study design and approved the conduct of this research.\u003c/p\u003e \u003cp\u003e The informed consent forms for publication have been obtained from all participants.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec4\" class=\"Section2\"\u003e \u003ch2\u003e2.3 Experimental Materials\u003c/h2\u003e \u003cp\u003eThis study adopted commonly used scales for assessing ASD symptoms. In addition, we also developed a scenario test to evaluate the intraverbal behavior in children with ASD.\u003c/p\u003e \u003cdiv id=\"Sec5\" class=\"Section3\"\u003e \u003ch2\u003e2.3.1 Measures\u003c/h2\u003e \u003cp\u003eThe \u003cem\u003eChildhood Autism Rating Scale (CARS)\u003c/em\u003e is used to assess changes in children\u0026rsquo;s autism symptoms (Schopler et al., \u003cspan citationid=\"CR25\" class=\"CitationRef\"\u003e1980\u003c/span\u003e), covering core domains such as interpersonal relationships, adaptability, sensory responses, and emotional-behavioral responses. It includes 15items, each rated in a 4-point Likert scale (1\u0026thinsp;=\u0026thinsp;normal, 4\u0026thinsp;=\u0026thinsp;severely abnormal). A higher total score indicates more severe autism symptoms. The Cronbach\u0026rsquo;s α coefficient for the scale in the present study was 0.823.\u003c/p\u003e \u003cp\u003eThe \u003cem\u003eAutism Treatment Evaluation Checklist (ATEC)\u003c/em\u003e was developed by Rimland et al. (Rimland \u0026amp; Edelson, \u003cspan citationid=\"CR24\" class=\"CitationRef\"\u003e1999\u003c/span\u003e), consisting of 77 items which were divided into 4 subscales: expressive/language communication (14 items), social ability (20 items), perceptual/cognitive ability (18 items), and health/physical/behavior (25 items). The total score ranges from 0 to 179, with higher scores indicating more severe ASD symptoms (Fang et al., \u003cspan citationid=\"CR11\" class=\"CitationRef\"\u003e2019\u003c/span\u003e). The Cronbach\u0026rsquo;s α coefficient in this study was 0.927.\u003c/p\u003e \u003cp\u003eThe \u003cem\u003eBasic Communication Behavior Assessment Scale\u003c/em\u003e for Children with Autism developed by Zhang and Wang (\u003cspan citationid=\"CR38\" class=\"CitationRef\"\u003e2005\u003c/span\u003e), was adopted to assess the development of basic communication behaviors in autistic children with no or limited verbal ability. It classifies communication behaviors into six dimensions: requesting, refusing, attracting attention, answering questions, requesting explanations, and social phrases, with a total of 90 observable behaviors.Only the 'answering questions' dimension was assessed in this study, which refers to the ability to provide verbal responses to causal questions (e.g., those beginning with 'why').A child\u0026rsquo;s language ability was evaluated based on a 4-point Likert scale (0\u0026thinsp;=\u0026thinsp;never, 1\u0026thinsp;=\u0026thinsp;occasionally, 2\u0026thinsp;=\u0026thinsp;frequently, 3\u0026thinsp;=\u0026thinsp;always). The Cronbach\u0026rsquo;s α coefficient in the present study was 0.743.\u003c/p\u003e \u003cp\u003eThe \u003cem\u003eParenting Stress Index-Short Form (PSI-SF)\u003c/em\u003e is a simplified version of the \u003cem\u003eParenting Stress Index\u003c/em\u003e developed through a series of exploratory factor analyses (Luo et al., \u003cspan citationid=\"CR18\" class=\"CitationRef\"\u003e2021\u003c/span\u003e). It includes 36 items, grouped into three 12-item factors: Parental Distress (PD), Parent-Child Dysfunctional Interaction (PCDI), and Difficult Child (DC). Each item is rated on a 5-point Likert scale (1\u0026thinsp;=\u0026thinsp;strongly disagree, 5\u0026thinsp;=\u0026thinsp;strongly agree), with higher scores indicating more parenting stress. The Cronbach\u0026rsquo;s α coefficient in this study was 0.908.\u003c/p\u003e \u003cp\u003eThe \u003cem\u003eSelf-Rating Anxiety Scale (SAS)\u003c/em\u003e was employed to measure current anxiety symptoms (Zung, \u003cspan citationid=\"CR40\" class=\"CitationRef\"\u003e1971\u003c/span\u003e). The scale consisted of 20 items.For each item, individuals were required to indicate the frequency with which they experience the symptom on a 4-point Likert scale (1\u0026thinsp;=\u0026thinsp;none or a little of the time; 2\u0026thinsp;=\u0026thinsp;some of the time; 3\u0026thinsp;=\u0026thinsp;good part of the time; 4\u0026thinsp;=\u0026thinsp;most or all of the time). The raw t otal score ranges from 20 to 80, which are then transformed to the standard score when multiplied by 1.25. Higher scores indicate higher anxiety levels. The Cronbach\u0026rsquo;s α coefficient in this study was 0.948.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec6\" class=\"Section3\"\u003e \u003ch2\u003e2.3.2 Digital Game Materials\u003c/h2\u003e \u003cp\u003eThe digital game consisted of three stages: understanding stage, learning stage, and practice stage. In the three stages,dialogue materials in the digital program were compiled, evaluated, and standardized by the researchers. Based on the analogy between trains and dialogues, different dialogue contents were designed, involving different life scenarios for children, such as school, family, and supermarket. Each scenario included 6 dialogues for training interactive conversations in children with ASD. The dialogue contents (e.g., \u0026ldquo;school\u0026rdquo;, \u0026ldquo;family\u0026rdquo;) were close to daily life, making it easy for children to understand. The alternative dialogue materials used in the three phases of this intervention study were all systematically compiled by the researchers based on common interaction scenarios in daily social settings. After compiling the materials, a standardized evaluation questionnaire was created using the Wenjuanxing platform. Ten postgraduate students majoring in psychology were recruited to form an evaluation team to assess the content. Following consistency testing, dialogue materials with an evaluation consistency coefficient of 80% or higher were finally selected and included in the formal intervention system.\u003c/p\u003e \u003cp\u003eThe learning videos in Phase 1 were recorded using train toys as the main experimental props to capture original images, which were then processed through standardized post-editing. The train image materials used in the connection training game were collected by photographing train toys to obtain original images, followed by removing irrelevant backgrounds and applying standardized cropping.The dialogue learning materials in Phase 2 were improved and rebuilt based on previously evaluated and approved basic materials. A multiple-choice question format was used for the standardized design of dialogue scenarios to help children learn and practice social dialogue strategies through selective learning and training. The real-scenario dialogue materials in Phase 3 also came from the previously evaluated and approved material library and were chosen based on the main interaction needs of real social situations.\u003c/p\u003e \u003cp\u003eThe intervention materials in the three stages presented a hierarchical characteristic of stepwise progression: Phase 1 focused on two core scenarios, namely schools and intervention institutions, including 12 segments of standardized dialogue content; Phase 2 expanded to family and supermarket scenarios on the basis of Phase 1, adding 12 new segments of dialogue materials (totaling 24 segments); Phase 3 further included hospital and amusement park scenarios, adding another 12 segments of dialogue materials (totaling 36 segments). All dialogue materials were targetedly designed around key interaction behaviors in social scenarios to ensure a high degree of adaptability between the content and children's daily life experiences.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec7\" class=\"Section3\"\u003e \u003ch2\u003e2.3.3 Materials for the Scenario Test\u003c/h2\u003e \u003cp\u003eIn addition, the researchers developed a scenario test to evaluate the children\u0026rsquo;s intraverbal behavior ability based on different dialogue scenarios (Watkins et al., \u003cspan citationid=\"CR33\" class=\"CitationRef\"\u003e1989\u003c/span\u003e; Sundberg et al., \u003cspan citationid=\"CR30\" class=\"CitationRef\"\u003e1990\u003c/span\u003e). The test included two parts: dialogue selection and real dialogue. In dialogue selection part, children with ASD were asked to choose the correct dialogue response from given options (e.g., Hello or Goodbye). In real dialogue, they were required to verbally respond to a daily dialogue. Each dialogue included five parts: greeting, mutual greeting, situational dialogue, topic shifting, and farewell, see Table\u0026nbsp;\u003cspan refid=\"Tab1\" class=\"InternalRef\"\u003e1\u003c/span\u003e for an example of a dialogue. The dialogue materials were close to daily life, including scenarios such as school, family, and supermarket, with simple questions and answers suitable for children\u0026rsquo;s cognition.\u003c/p\u003e \u003cp\u003e \u003cdiv class=\"gridtable\"\u003e\u003ctable float=\"Yes\" id=\"Tab1\" border=\"1\"\u003e \u003ccaption language=\"En\"\u003e \u003cdiv class=\"CaptionNumber\"\u003eTable 1\u003c/div\u003e \u003cdiv class=\"CaptionContent\"\u003e \u003cp\u003eExample of Intraverbal Behavior test materials\u003c/p\u003e \u003c/div\u003e \u003c/caption\u003e \u003ccolgroup cols=\"4\"\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c1\" colnum=\"1\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c2\" colnum=\"2\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c3\" colnum=\"3\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c4\" colnum=\"4\"\u003e\u003c/div\u003e \u003cthead\u003e \u003ctr\u003e \u003cth align=\"left\" colname=\"c1\"\u003e \u003cp\u003eAnalogy: Different components of trains\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c2\"\u003e \u003cp\u003eDialogue phase\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c3\"\u003e \u003cp\u003eProgram dialogue\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c4\"\u003e \u003cp\u003eChild\u0026rsquo;s response\u003c/p\u003e \u003c/th\u003e \u003c/tr\u003e \u003c/thead\u003e \u003ctbody\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eLocomotive\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003egreeting\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003eProgram: \u0026ldquo;Hello!\u0026rdquo;\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003eC: _____\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003ePower carriage\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003emutual greeting\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003eP: \u0026ldquo;Did you have a good rest last night?\u0026rdquo;\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003eC: _____\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003ePassenger carriage\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003esituational dialogue\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003eP: \u0026ldquo;I feel pretty good.\u0026rdquo;\u003c/p\u003e \u003cp\u003eP: \u0026ldquo;Do you want to draw?\u0026rdquo;\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003eC: _____\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eThe track converter\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003etopic shifting\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003eP: \u0026ldquo;That's great.\u0026rdquo;\u003c/p\u003e \u003cp\u003eP: \u0026ldquo;By the way, what would you like for lunch?\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003eC: _____\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eThe train tail\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003efarewell\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003eP: \u0026ldquo;OK. See you later!\u0026rdquo;\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003eC: _____\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003c/tbody\u003e \u003c/colgroup\u003e \u003c/table\u003e\u003c/div\u003e \u003c/p\u003e \u003c/div\u003e \u003c/div\u003e \u003cdiv id=\"Sec8\" class=\"Section2\"\u003e \u003ch2\u003e2.4 Experimental Design and Procedures\u003c/h2\u003e \u003cdiv id=\"Sec9\" class=\"Section3\"\u003e \u003ch2\u003e2.4.1 Design and Implementation of the Digital Game\u003c/h2\u003e \u003cp\u003eBased on the idea of the book \u003cem\u003eThe Conversation Train\u003c/em\u003e (\u003cem\u003elearning conversations with trains\u003c/em\u003e)(Shaul, \u003cspan citationid=\"CR26\" class=\"CitationRef\"\u003e2014\u003c/span\u003e), this study designed a digital game program to intervene in the intraverbal behavior of children with ASD. The digital program included three stages: (1) Understanding stage: Children with ASD learned the specific components of a complete dialogue and understood dialogue rules through the correspondence between different train carriages and different dialogue parts. That is, the locomotive represents the start of a dialogue (greetings, e.g., \u0026ldquo;Hello\u0026rdquo;); the power carriage promotes the continuation of the dialogue (mutual greetings, e.g., \u0026ldquo;How are you feeling today?\u0026rdquo;); the passenger carriage represents daily conversations (situational dialogue, e.g.,\u0026ldquo;What would you like for dinner?\u0026rdquo;); the track converter represents topic shifts (topic shifting, e.g.,\u0026ldquo;By the way, is there a painting class today?\u0026rdquo;) ; and the train tail represents the end of a dialogue (mutual farewells, e.g,\u0026ldquo;Goodbye\u0026rdquo;). (2) Learning stage: Children learned correct dialogue content through dialogue selection. (3) Practice stage: The ASD children\u0026rsquo;s real intraverbal behavior ability was trained. See Fig.\u0026nbsp;\u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e1\u003c/span\u003e for the illustration.\u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003cp\u003eThe intervention game was developed by professional programmers utilizing a Browser/Server (B/S) architecture. The front-end implementation employed HTML5, CSS3, and JavaScript for web development, while the back-end system used Spring Boot to develop RESTful APIs. The platform was hosted on a cloud server to ensure accessibility, with load balancing and HTTPS security encryption implemented through Nginx, and cross-environment consistency ensured by Docker containerization technology. The overall architecture supports horizontal expansion to meet performance requirements under high concurrency scenarios, while providing complete permission control and audit log functions, complying with the Level 2.0 / Level 3.0 security standards. The user interface of the digital game was shown in Fig.\u0026nbsp;\u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e1\u003c/span\u003e.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec10\" class=\"Section3\"\u003e \u003ch2\u003e2.4.2 Design and Process of the Intervention Experiment\u003c/h2\u003e \u003cp\u003eThis study adopted a pre-test-post-test randomized controlled experimental design, with a 2 (group: experimental group, control group) \u0026times; 2 (time: T0, T1) mixed factorial design. The dependent variables were scores of the five scales and the assessment score of intraverbal behavior ability of the participants.\u003c/p\u003e \u003cp\u003eThe research process included three stages: pre-test assessment, intervention, and post-test assessment. The pre-test assessment was conducted before the interventions (T0), including filling in the questionnaires by the children\u0026rsquo; therapists and caregivers and completing the intraverbal behavior test by the children themselves. Rehabilitation therapists completed the following three scales: \u003cem\u003eChildhood Autism Rating Scale (CARS)\u003c/em\u003e, \u003cem\u003eAutism Treatment Evaluation Checklist (ATEC)\u003c/em\u003e, and \u003cem\u003eBasic Communication Behavior Assessment Scale for Children with Autism\u003c/em\u003e. The primary caregivers (typically their mothers) of children with ASD completed the \u003cem\u003eParenting Stress Index-Short Form (PSI-SF)\u003c/em\u003e and \u003cem\u003eSelf-Rating Anxiety Scale (SAS)\u003c/em\u003e.\u003c/p\u003e \u003cp\u003eIn the pre-test, all ASD children underwent a test of the intraverbal behavior ability in quiet rooms of the ASD rehabilitation institutions.The test was administered by a trained researcher to a child, with the assistance of the child\u0026rsquo;s therapist. During the test, the researcher presented the test materials to the children on a tablet, including 5 dialogue selections (20 dialogues in total) and 6 real dialogues (30 dialogues in total).\u003c/p\u003e \u003cp\u003eIn the dialogue selection stage, children with ASD were required to select the correct response from options under the dialogue. Each dialogue selection was displayed for 10 seconds. After the dialogue was presented, the researcher asked the child to choose the response; if the child failed to choose within 10 seconds, it was considered an incorrect selection. For illiterate children, the researcher first read the dialogue content and options to them, then asked the child to choose and started timing.\u003c/p\u003e \u003cp\u003eIn the real dialogue stage, the researcher interacted with the child strictly according to the experimental materials. After the researcher spoke the dialogue, the child needed to respond within 5 seconds; if no response was made within 5 seconds, the researcher repeated the dialogue and waited another 5 seconds. A correct response within the given time was scored 1 point; incorrect or no response scored 0 point.\u003c/p\u003e \u003cp\u003eDuring the test, the therapist also recorded the responses of the children, in order to measure inter-evaluator reliability. After the test, the records completed by the researchers and the therapists were compared to determine the consistency. Consistency was calculated by the following formula: (number of consistent records / total number of consistent and inconsistent records) \u0026times; 100%. The inter-evaluator agreement for all participants in the study was over 90%.\u003c/p\u003e \u003cp\u003eAfter the pre-test, the participants in the experimental group and the control group received different interventions. The participants in the experimental group received the digital intervention training in addition to the daily training in the ASD rehabilitation institutions. The digital game lasted approximately 15 minutes per day, with 5 days per week. In total, there were 30 times of digital game sessions, lasting about one and a half months. the participants in the control group received only daily training as usual in the institutions. After the participants in the experimental group completed the digital intervention training (T1), all the participants were evaluated again. The procedure of the post-test was the same as that in the pre-test.\u003c/p\u003e \u003c/div\u003e \u003c/div\u003e \u003cdiv id=\"Sec11\" class=\"Section2\"\u003e \u003ch2\u003e2.5 Data analysis\u003c/h2\u003e \u003cp\u003eDue to the limited sample size and the non-normal distribution of the results obtained in this study, parametric analyses such as analysis of variance (ANOVA) were not appropriate. Therefore, a non-parametric approach was employed using a 2 (group: experimental, control) \u0026times; 2 (time: T0, T1) mixed design, with the Mann-Whitney U test applied to compare differences in scores between groups across time points. Data are presented as median (Mdn), interquartile range (IQR), test statistic (U), and significance level (p). Analyses were conducted for scales、dialogue selection and authentic dialogue scores to confirm comparable baseline levels of verbal behavioral ability in children with autism spectrum disorder (ASD), thereby establishing a foundation for subsequent intervention.\u003c/p\u003e \u003c/div\u003e"},{"header":"3 Results","content":"\u003cdiv id=\"Sec13\" class=\"Section2\"\u003e \u003ch2\u003e3.1 Results of Scales\u003c/h2\u003e \u003cp\u003eThe mean total scale scores at the pre-test and post-test in the two groups were shown in Table\u0026nbsp;2.\u003c/p\u003e \u003cp\u003e \u003cdiv class=\"gridtable\"\u003e\u003ctable float=\"No\" id=\"Taba\" border=\"1\"\u003e \u003ccolgroup cols=\"6\"\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c1\" colnum=\"1\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c2\" colnum=\"2\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c3\" colnum=\"3\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c4\" colnum=\"4\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c5\" colnum=\"5\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c6\" colnum=\"6\"\u003e\u003c/div\u003e \u003cthead\u003e \u003ctr\u003e \u003cth align=\"left\" colname=\"c1\" morerows=\"1\" rowspan=\"2\"\u003e\u0026nbsp;\u003c/th\u003e \u003cth align=\"left\" colspan=\"2\" nameend=\"c3\" namest=\"c2\"\u003e \u003cp\u003eEXP\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/th\u003e \u003cth align=\"left\" colspan=\"2\" nameend=\"c6\" namest=\"c5\"\u003e \u003cp\u003eCTL\u003c/p\u003e \u003c/th\u003e \u003c/tr\u003e \u003ctr\u003e \u003cth align=\"left\" colname=\"c2\"\u003e \u003cp\u003eT0\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c3\"\u003e \u003cp\u003eT1\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/th\u003e \u003cth align=\"left\" colname=\"c5\"\u003e \u003cp\u003eT0\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c6\"\u003e \u003cp\u003eT1\u003c/p\u003e \u003c/th\u003e \u003c/tr\u003e \u003c/thead\u003e \u003ctbody\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003ePSI\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e99.17(18.57)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e101.00(17.48)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e104.17(20.12)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e106.19(24.22)\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eSAS\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e40.50(9.38)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e40.43(6.56)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e43.42(8.26)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e44.23(6.26)\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eATEC\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e100.3(29.08)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e76.78(28.08)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e104.56(23.27)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e98.88(42.64)\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eCARS\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e28.28(6.99)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e30.22(9.64)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e30.28(7.47)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e32.31(11.35)\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eCB\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e11.06(7.27)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e15.67(6.32)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e10.72(7.84)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e10.81(5.59)\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003c/tbody\u003e \u003c/colgroup\u003e \u003ctfoot\u003e \u003ctr\u003e\u003ctd colspan=\"6\"\u003e\u003cem\u003e(Note: EXP=experimental group;CTL=control group;CARS=childhood autism rating scale;ATEC=autism treatment evaluation checklist;PSI=parenting stress index;SAS=self-rating anxiety scale;CB=communication behavior assessment scale for children with autism.)\u003c/em\u003e\u003c/td\u003e\u003c/tr\u003e \u003c/tfoot\u003e \u003c/table\u003e\u003c/div\u003e \u003c/p\u003e \u003cdiv id=\"Sec14\" class=\"Section3\"\u003e \u003ch2\u003e3.1.1 Baseline comparison between the two groups\u003c/h2\u003e \u003cp\u003eMann-Whitney U tests on pre-test scale data in the two groups showed no significant differences on scores of PSI (U\u0026thinsp;=\u0026thinsp;134.5, \u003cem\u003ep\u003c/em\u003e\u0026thinsp;=\u0026thinsp;0.389), SAS (U\u0026thinsp;=\u0026thinsp;126.0, \u003cem\u003ep\u003c/em\u003e\u0026thinsp;=\u0026thinsp;0.265), ATEC (U\u0026thinsp;=\u0026thinsp;159.0, \u003cem\u003ep\u003c/em\u003e\u0026thinsp;=\u0026thinsp;0.938), CARS (U\u0026thinsp;=\u0026thinsp;135.0, \u003cem\u003ep\u003c/em\u003e\u0026thinsp;=\u0026thinsp;0.406), and communication behavior scale (U\u0026thinsp;=\u0026thinsp;156, \u003cem\u003ep\u003c/em\u003e\u0026thinsp;=\u0026thinsp;0.864) between the experimental group and control group, indicating that the two groups were comparable at the baseline.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec15\" class=\"Section3\"\u003e \u003ch2\u003e3.1.2 Comparison of pre-test and post-test in the two groups\u003c/h2\u003e \u003cp\u003eFor the experimental group, after receiving the digital training intervention in addition to the daily intervention, the post-test score of ATEC (Mdn\u0026thinsp;=\u0026thinsp;83, IQR\u0026thinsp;=\u0026thinsp;51\u0026ndash;99) was significantly lower (Z = -3.551, \u003cem\u003ep\u003c/em\u003e\u0026thinsp;\u0026lt;\u0026thinsp;0.001, effect size r\u0026thinsp;=\u0026thinsp;0.59) than the pre-test score (Mdn\u0026thinsp;=\u0026thinsp;108, IQR\u0026thinsp;=\u0026thinsp;72\u0026ndash;127). For the control group, there was no significant difference (Z = -0.776, \u003cem\u003ep\u003c/em\u003e\u0026thinsp;=\u0026thinsp;0.438, effect size r\u0026thinsp;=\u0026thinsp;0.13) between the pre-test score (Mdn\u0026thinsp;=\u0026thinsp;104, IQR\u0026thinsp;=\u0026thinsp;95\u0026ndash;116) and the post-test ATEC score (Mdn\u0026thinsp;=\u0026thinsp;99, IQR\u0026thinsp;=\u0026thinsp;71\u0026ndash;141). These results indicated that the introduction of the digital training significantly reduced the severity of ASD symptoms in the experimental group. See Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003e for the illustration.\u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003cp\u003eFor the communication behavior scale scores, there was a significant difference (Z = -2.943, \u003cem\u003ep\u003c/em\u003e\u0026thinsp;=\u0026thinsp;0.003, effect size r\u0026thinsp;=\u0026thinsp;0.49) between the pre-test score (Mdn\u0026thinsp;=\u0026thinsp;9, IQR\u0026thinsp;=\u0026thinsp;8\u0026ndash;16) and post-test score (Mdn\u0026thinsp;=\u0026thinsp;16, IQR\u0026thinsp;=\u0026thinsp;11\u0026ndash;20) in the experimental group. In the control group, no significant difference (Z = -0.647, \u003cem\u003ep\u003c/em\u003e\u0026thinsp;=\u0026thinsp;0.517, effect size r\u0026thinsp;=\u0026thinsp;0.01) between the pre-test score (Mdn\u0026thinsp;=\u0026thinsp;10, IQR\u0026thinsp;=\u0026thinsp;7\u0026ndash;13) and the post-test score (Mdn\u0026thinsp;=\u0026thinsp;9, IQR\u0026thinsp;=\u0026thinsp;7\u0026ndash;15) was observed. See Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003e for the illustration.\u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003cp\u003eFor the PSI scale scores, there was no significant difference (Z = -0.501, \u003cem\u003ep\u003c/em\u003e\u0026thinsp;=\u0026thinsp;0.616, effect size r\u0026thinsp;=\u0026thinsp;0.08) between the pre-test score (Mdn\u0026thinsp;=\u0026thinsp;99, IQR\u0026thinsp;=\u0026thinsp;84\u0026ndash;119) and the post-test score (Mdn\u0026thinsp;=\u0026thinsp;100, IQR\u0026thinsp;=\u0026thinsp;87\u0026ndash;116) in the experimental group. In the control group, no significant difference (Z = -0.028, p\u0026thinsp;=\u0026thinsp;0.977, effect size r\u0026thinsp;=\u0026thinsp;0.004) was found between the pre-test score (Mdn\u0026thinsp;=\u0026thinsp;108, IQR\u0026thinsp;=\u0026thinsp;89\u0026ndash;119) and the post-test score (Mdn\u0026thinsp;=\u0026thinsp;112, IQR\u0026thinsp;=\u0026thinsp;97\u0026ndash;123).\u003c/p\u003e \u003cp\u003eFor the SAS scale score, no significant difference (Z = -0.207, \u003cem\u003ep\u003c/em\u003e\u0026thinsp;=\u0026thinsp;0.836, effect size r\u0026thinsp;=\u0026thinsp;0.03) was found between the pre-test score (Mdn\u0026thinsp;=\u0026thinsp;39, IQR\u0026thinsp;=\u0026thinsp;35\u0026ndash;44) and the post-test score (Mdn\u0026thinsp;=\u0026thinsp;39, IQR\u0026thinsp;=\u0026thinsp;36\u0026ndash;45) in the experimental group. In the control group, there was also no significant difference (Z = -0.313, \u003cem\u003ep\u003c/em\u003e\u0026thinsp;=\u0026thinsp;0.754, effect size r\u0026thinsp;=\u0026thinsp;0.05) between the pre-test score (Mdn\u0026thinsp;=\u0026thinsp;44, IQR\u0026thinsp;=\u0026thinsp;36\u0026ndash;48) and the post-test score (Mdn\u0026thinsp;=\u0026thinsp;44, IQR\u0026thinsp;=\u0026thinsp;41\u0026ndash;49) .\u003c/p\u003e \u003cp\u003eFor the CARS scale score, no significant difference (Z = -0.825, \u003cem\u003ep\u003c/em\u003e\u0026thinsp;=\u0026thinsp;0.410, effect size r\u0026thinsp;=\u0026thinsp;0.14) was found between the pre-test score (Mdn\u0026thinsp;=\u0026thinsp;29, IQR\u0026thinsp;=\u0026thinsp;22\u0026ndash;32) and the post-test score (Mdn\u0026thinsp;=\u0026thinsp;31, IQR\u0026thinsp;=\u0026thinsp;22\u0026ndash;38) in the experimental group. In the control group, there was no significant difference (Z = -0.906, \u003cem\u003ep\u003c/em\u003e\u0026thinsp;=\u0026thinsp;0.365, effect size r\u0026thinsp;=\u0026thinsp;0.16) between the pre-test score (Mdn\u0026thinsp;=\u0026thinsp;31, IQR\u0026thinsp;=\u0026thinsp;24\u0026ndash;34) and the post-test score (Mdn\u0026thinsp;=\u0026thinsp;30, IQR\u0026thinsp;=\u0026thinsp;26\u0026ndash;44) scores.\u003c/p\u003e \u003c/div\u003e \u003c/div\u003e \u003cdiv id=\"Sec16\" class=\"Section2\"\u003e \u003ch2\u003e3.2 Results of the Intraverbal Behavior Ability Test\u003c/h2\u003e \u003cp\u003eTable\u0026nbsp;\u003cspan refid=\"Tab2\" class=\"InternalRef\"\u003e3\u003c/span\u003e showed the mean scores in the intraverbal behavior ability test in the experimental group and the control group.\u003c/p\u003e \u003cp\u003e \u003cdiv class=\"gridtable\"\u003e\u003ctable float=\"Yes\" id=\"Tab2\" border=\"1\"\u003e \u003ccaption language=\"En\"\u003e \u003cdiv class=\"CaptionNumber\"\u003eTable 3\u003c/div\u003e \u003cdiv class=\"CaptionContent\"\u003e \u003cp\u003eMean scores of the intraverbal behavior assessment (Mean (SD))\u003c/p\u003e \u003c/div\u003e \u003c/caption\u003e \u003ccolgroup cols=\"5\"\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c1\" colnum=\"1\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c2\" colnum=\"2\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c3\" colnum=\"3\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c4\" colnum=\"4\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c5\" colnum=\"5\"\u003e\u003c/div\u003e \u003cthead\u003e \u003ctr\u003e \u003cth align=\"left\" colname=\"c1\" morerows=\"1\" rowspan=\"2\"\u003e\u0026nbsp;\u003c/th\u003e \u003cth align=\"left\" colspan=\"2\" nameend=\"c3\" namest=\"c2\"\u003e \u003cp\u003eEXP\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colspan=\"2\" nameend=\"c5\" namest=\"c4\"\u003e \u003cp\u003eCTL\u003c/p\u003e \u003c/th\u003e \u003c/tr\u003e \u003ctr\u003e \u003cth align=\"left\" colname=\"c2\"\u003e \u003cp\u003eT0\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c3\"\u003e \u003cp\u003eT1\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c4\"\u003e \u003cp\u003eT0\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c5\"\u003e \u003cp\u003eT1\u003c/p\u003e \u003c/th\u003e \u003c/tr\u003e \u003c/thead\u003e \u003ctbody\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eDialogue selection\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e10.78(5.23)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e18.00(2.63)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e13.28(3.31)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e12.71(5.76)\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eReal intraverbal behavior\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e14.33(9.71)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e26.50(4.25)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e21.57(6.32)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e18.70(9.35)\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003c/tbody\u003e \u003c/colgroup\u003e \u003ctfoot\u003e \u003ctr\u003e\u003ctd colspan=\"5\"\u003e\u003cem\u003e(Note: EXP=experimental group;CTL=control group.)\u003c/em\u003e\u003c/td\u003e\u003c/tr\u003e \u003c/tfoot\u003e \u003c/table\u003e\u003c/div\u003e \u003c/p\u003e \u003cdiv id=\"Sec17\" class=\"Section3\"\u003e \u003ch2\u003e3.2.1 Baseline comparison between the two groups\u003c/h2\u003e \u003cp\u003eThe Mann-Whitney U test showed no significant difference in pre-test scores between the experimental group and the control group. Specifically, there was no significant difference in pre-test scores of dialogue selection between the two groups (U\u0026thinsp;=\u0026thinsp;90.00, \u003cem\u003ep\u003c/em\u003e\u0026thinsp;=\u0026thinsp;0.180), and no significant difference in pre-test scores of real dialogue between the two groups (U\u0026thinsp;=\u0026thinsp;142.00, \u003cem\u003ep\u003c/em\u003e\u0026thinsp;=\u0026thinsp;0.276). Therefore, the two groups were comparable at the baseline.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec18\" class=\"Section3\"\u003e \u003ch2\u003e3.2.2 Comparison of pre-test and post-test in the two groups\u003c/h2\u003e \u003cp\u003eResults showed that the post-test score of dialogue selection (Mdn\u0026thinsp;=\u0026thinsp;19, IQR\u0026thinsp;=\u0026thinsp;17\u0026ndash;20) was significantly higher than the pre-test score (Mdn\u0026thinsp;=\u0026thinsp;10, IQR\u0026thinsp;=\u0026thinsp;6\u0026ndash;15) in the experimental group (Z = -4.097, \u003cem\u003ep\u003c/em\u003e\u0026thinsp;=\u0026thinsp;0.001, effect size r\u0026thinsp;=\u0026thinsp;0.97); the post-test score of real dialogue (Mdn\u0026thinsp;=\u0026thinsp;29, IQR\u0026thinsp;=\u0026thinsp;24\u0026ndash;30) was also significantly higher than the pre-test score (Mdn\u0026thinsp;=\u0026thinsp;13, IQR\u0026thinsp;=\u0026thinsp;7\u0026ndash;24) in the experimental group (Wilcoxon test: Z = -3.758, \u003cem\u003ep\u003c/em\u003e\u0026thinsp;=\u0026thinsp;0.001, effect size r\u0026thinsp;=\u0026thinsp;0.89). An effect size r\u0026thinsp;\u0026gt;\u0026thinsp;0.5 indicates that the intervention effect has moderate or higher practical significance, suggesting that the digital game significantly improved the intraverbal behavior ability in children with ASD. See Fig.\u0026nbsp;\u003cspan refid=\"Fig4\" class=\"InternalRef\"\u003e4\u003c/span\u003e for the illustration.\u003c/p\u003e \u003cp\u003eThe control group received only daily intervention without the digital game training. There was no significant difference between the post-test score (Mdn\u0026thinsp;=\u0026thinsp;11, IQR\u0026thinsp;=\u0026thinsp;9\u0026ndash;20) and pre-test score (Mdn\u0026thinsp;=\u0026thinsp;14, IQR\u0026thinsp;=\u0026thinsp;10\u0026ndash;16) of dialogue selection (Z = -0.392, \u003cem\u003ep\u003c/em\u003e\u0026thinsp;=\u0026thinsp;0.701, effect size r\u0026thinsp;=\u0026thinsp;0.10); Neither significant difference was found between the post-test score (Mdn\u0026thinsp;=\u0026thinsp;17, IQR\u0026thinsp;=\u0026thinsp;7\u0026ndash;28) and pre-test score (Mdn\u0026thinsp;=\u0026thinsp;22, IQR\u0026thinsp;=\u0026thinsp;14\u0026ndash;27) of real dialogue (Z = -0.991, \u003cem\u003ep\u003c/em\u003e\u0026thinsp;=\u0026thinsp;0.329, effect size r\u0026thinsp;=\u0026thinsp;0.26) in the control group.\u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003c/div\u003e \u003c/div\u003e"},{"header":"4 Discussion","content":"\u003cp\u003eThis study demonstrated that the digital game intervention combined with conventional training can significantly improve the intraverbal behavior of children with ASD .The results showed that the participants in the experimental group (contrary to the control group) had significantly reduced ATEC scores and increased CB scores. More importantly, The test of intraverbal behavior ability further showed that after the interventions, the correct response rates of the participants in the experimental group in dialogue selection and real dialogue increased by 67% and 85% respectively, which were significantly better than those of the control group. These results were consistent with the conclusion of the systematic review that digital game intervention can improve social skills in children with ASD (Gao et al., \u003cspan citationid=\"CR13\" class=\"CitationRef\"\u003e2025\u003c/span\u003e), and also in line with the meta-analysis finding that structured digital intervention is more targeted than traditional therapy (Wang et al., \u003cspan citationid=\"CR35\" class=\"CitationRef\"\u003e2025\u003c/span\u003e).\u003c/p\u003e \u003cp\u003eSpecifically, the significant reduction in ATEC scores in the experimental group suggested that digital game intervention may effectively alleviate the core symptoms of ASD through structured interaction. This effect may stem from the intervention program\u0026rsquo;s structured and visual decomposition of abstract social dialogue rules. The game in our study used different carriages to represent different components of dialogue, and designed three stages (i.e.,\"understanding-learning-practice\") to facilitate the neuroplasticity. In addition, the real-time feedback in the game (e.g., \"on track / off track\" prompts) strengthens the learning acquisition of correct dialogue patterns, which is consistent with the mechanism that social interaction training can activate the mirror neuron system in children with ASD (Dapretto et al., \u003cspan citationid=\"CR9\" class=\"CitationRef\"\u003e2006\u003c/span\u003e). In contrast, there was no significant change in ATEC scores in the control group, which aligns with the view that conventional intervention has limited short-term improvement effects on language ability in ASD children (Reichow et al., \u003cspan citationid=\"CR22\" class=\"CitationRef\"\u003e2018\u003c/span\u003e).\u003c/p\u003e \u003cp\u003eThe significant increase in CB scores in the experimental group indicates that digital game intervention significantly improved the actual communication behavior in children with ASD. This result echoes the reduction in the ATEC score, indicating that the digital intervention not only alleviated core symptoms but also directly promoted the functional improvement of intraverbal behavior ability. This may be attributed to the design of the real dialogue practice module in the digital program, which provides structured language application opportunities through scenario simulation andreal-time feedback \u0026mdash;a \"learning by doing\" model that essentially incorporates the embodiment theory in digital intervention (Kumar, \u003cspan citationid=\"CR16\" class=\"CitationRef\"\u003e2025\u003c/span\u003e). Mechanistically, the improvement in CB scores may be related to the cognitive scaffolding role of the train carriage metaphor in the digital game. The game decomposes dialogue structures into visualized carriage units, helping children with ASD transform abstract social rules into concrete operations. This \"visual-semantic mapping\" intervention strategy is similar to the principle of traditional social story therapy (Thiemann \u0026amp; Goldstein, \u003cspan citationid=\"CR31\" class=\"CitationRef\"\u003e2001\u003c/span\u003e; Zimmerman et al., \u003cspan citationid=\"CR39\" class=\"CitationRef\"\u003e2020\u003c/span\u003e).\u003c/p\u003e \u003cp\u003e It is worth noting that, the progress of the participants in the experimental group in dialogue selection and real dialogue provided more direct and specific evidences in support of the enhanced ability of the ASD children to transfer training content to actual expression. This outcome might stem from the vivid design of the digital intervention, where train-themed visual symbols transform abstract conversational structures into tangible visual representations. Previous studies have shown that immersive visualization interfaces can reduce social cognitive load in children with ASD (Bacchetta et al., 2024). Such design enables children to rapidly internalize the logical sequence required for conversations, laying the foundation for skill transfer. In addition, the reduced standard deviation of scores in the experimental group indicates that the digital intervention has good effects on children with different ability levels, providing a basis for clinical promotion.This finding supports the view that language intervention in ASD children should incorporate repetitive reinforcement task design (Reichow et al., \u003cspan citationid=\"CR22\" class=\"CitationRef\"\u003e2018\u003c/span\u003e) and highlights the advantages of digital game intervention in standardized training.\u003c/p\u003e \u003cp\u003eThe parental SAS scores in the experimental group showed a modest numerical change (T0\u0026thinsp;=\u0026thinsp;40.50 (9.38), T1\u0026thinsp;=\u0026thinsp;40.43 (6.56)) but this difference did not reach statistical significance (Z = -0.207, p\u0026thinsp;=\u0026thinsp;0.836). Despite the lack of statistical significance, this subtle trend may reflect potential indirect benefits of the digital intervention, which could be associated with the significant improvement in children\u0026rsquo;s communication abilities (evidenced by the increased Communication Behavior Scale score). When parents observed specific and visible progress in their children\u0026rsquo;s intraverbal behavior\u0026mdash;particularly active expression in real dialogue\u0026mdash;it may have alleviated their concerns about their children\u0026rsquo;s social communication disorders and enhanced their confidence and self-efficacy in supporting interactive communication (Karst \u0026amp; Van Hecke, \u003cspan citationid=\"CR15\" class=\"CitationRef\"\u003e2012\u003c/span\u003e), aligning with the \"child behavior-parental well-being\" bidirectional influence model (Callanan et al., \u003cspan citationid=\"CR6\" class=\"CitationRef\"\u003e2021\u003c/span\u003e). The absence of statistical significance for SAS scores is consistent with the non-significant change in PSI scores, likely due to these scales measuring broader, more persistent aspects of parental stress and anxiety (e.g., long-term caregiving burdens, economic pressures) that are less responsive to short-term interventions targeting specific child communication skills. Additionally, individual differences in parental stress responses and the relatively small sample size may have contributed to the lack of measurable statistical effects, despite the observed directional trend.\u003c/p\u003e \u003cp\u003eThe results of this study highlight the great potential of digital therapy, especially gamified programs based on clear theoretical frameworks, in language intervention for children with ASD. This study provided an innovative idea: utilizing the visual learning advantages and intrinsic motivation for games common in children with ASD (Grynszpan et al., \u003cspan citationid=\"CR14\" class=\"CitationRef\"\u003e2014\u003c/span\u003e) to transform abstract social language rules into concrete, structured, and repeatedly practicable digital tasks. This study proved that the digital gamified intervention according to the \"conversation train\" model, which combines dialogue decomposition, rules learning, and active verbal output in three stages, was effective and has significant advantages over traditional methods. Its browser/server architecture-based digital platform ensures the usability of the program. The high standardization of study content, the gradual difficulty setting, and the real-time feedback mechanisms, made the learning progress smoothly and reduced potential confoundings caused by trainers or the environment (Spain \u0026amp; Blainey, \u003cspan citationid=\"CR28\" class=\"CitationRef\"\u003e2015\u003c/span\u003e). Moreover, the program supported multi-terminal access, allowing children to use it flexibly in different places, such as home and school, which overcomes the limitations of locations and resources of therapy, thus greatly improves the accessibility. The convenient use in family environments also enables parents to participate in or assist their children\u0026rsquo;s training, and the observable progress would eventually help them alleviate the anxiety and enhance their confidence, thus forming a more supportive intervention environment.\u003c/p\u003e \u003cp\u003eThe gamified intervention demonstrated significant improvements in intraverbal behavior among children with ASD. Clinically, this approach offers three key advantages: (1) standardized protocols reduce therapist dependency; (2) flexible implementation across home/school settings; (3) real-time feedback enhances engagement. For policy makers, these findings support integrating digital tools into standard ASD intervention frameworks, potentially reducing healthcare disparities in resource-limited regions.\u003c/p\u003e \u003cp\u003eThis study also has some limitations. First, the sample size in our study (18 ASD children in the experimental group, 16 in the control group) is relatively small. Future research enrolling more children with ASD is needed to further demonstrate our findings. (2) We only conducted the pre-test and post-test to explore the efficacy of the digital intervention. Future studies should conduct follow-up assessments (e.g., 3 months, 6 months) to investigate the long-term effect of the digital intervention.(3) The study only used scales and behavioral tests, which are subjective measurements and might be influenced by the reporters\u0026rsquo; memory bias and raters\u0026rsquo; evaluation bias. In the future, it is necessary to utilize multimodal objective measurement, such as eye-tracking technology and multi-channel physiological signal recording, to evaluate the effect of digital intervention objectively and also to understand the associated psychological and physiological mechanism.\u003c/p\u003e"},{"header":"5 Conclusion","content":"\u003cp\u003eThis study developed a digital game intervention program based on the idea of \u003cem\u003eLearning Conversations with Trains\u003c/em\u003e, and designed a randomized controlled study to explore the effect of the digital intervention in children with ASD. The results demonstrated that the digital intervention can effectively improve the abilities of intraverbal behavior and core symptoms (as indexed by ATEC and CB scores) in ASD children and enhance the interactive and conversational skills of children with autism. The findings in this study provide empirical evidence for the efficacy of digital therapy in language intervention in children with ASD. It also have important implications for the new intervention strategies in children with ASD.\u003c/p\u003e"},{"header":"Declarations","content":"\u003ch2\u003eFunding:\u003c/h2\u003e \u003cp\u003eResearch on Key Technologies of Brain-Computer Digital Regulation for Children with ADHD and Autism under the \"Sharpshooter\" Research and Development Program of Zhejiang Province(2023C03002)\u003c/p\u003e \u003cp\u003eConflicts of Interest: All authors declare no potential conflicts of interest.\u003c/p\u003e \u003cp\u003e Ethics Approval: The study was approved.(Approval No.: 2024(E2)-HS-013]).\u003c/p\u003e \u003cp\u003e Consent: Written informed consent was obtained from all participants\u0026rsquo; parents or legal guardians.\u003c/p\u003e \u003cp\u003eMaterials and/or Code Availability: The digital game intervention materials and code are available from the corresponding author upon reasonable request.\u003c/p\u003e\u003ch2\u003eAuthor Contribution\u003c/h2\u003e\u003cp\u003eConceptualization: Zhiguo Hu; Methodology: Yingjian Zhang; Data Collection: Yingjian Zhang; Writing: Yingjian Zhang and Huixin Lu\u0026rdquo;]\u003c/p\u003e\u003ch2\u003eData Availability\u003c/h2\u003e\u003cp\u003eThe digital game intervention materials and code are available from the corresponding author upon reasonable request.\u003c/p\u003e"},{"header":"References","content":"\u003col\u003e\n\u003cli\u003eAmerican Psychiatric Association. (2013). \u003cem\u003eDiagnostic and statistical manual of mental disorders (DSM-5\u0026reg;)\u003c/em\u003e. Washington, DC: Author.\u003c/li\u003e\n\u003cli\u003eAmeis, S. H., Kassee, C., Corbett‐Dick, P., Cole, L., Dadhwal, S., Lai, M. C., ... \u0026amp; Correll, C. U. (2018). Systematic review and guide to management of core and psychiatric symptoms in youth with autism. \u003cem\u003eActa Psychiatrica Scandinavica\u003c/em\u003e, 138(5), 379-400.\u003c/li\u003e\n\u003cli\u003eALM\u0026aring;S, I. K., Smith, D. P., Eldevik, S., et al. (2022). Emergent intraverbal and reverse intraverbal behavior following listener training in children with autism spectrum disorder. Analysis of Verbal Behavior, 38(1), 1\u0026ndash;23. \u003c/li\u003e\n\u003cli\u003eBekele, E., Zheng, Z., Swanson, A., Crittendon, J., Warren, Z., \u0026amp; Sarkar, N. (2013). Understanding how adolescents with autism respond to facial expressions in virtual reality environments. \u003cem\u003eIEEE transactions on visualization and computer graphics\u003c/em\u003e, 19(4), 711-720.\u003c/li\u003e\n\u003cli\u003eBrignell, A., Chenausky, K. V., Song, H., Zhu, J., Suo, C., \u0026amp; Morgan, A. T. (2018). Communication interventions for autism spectrum disorder in minimally verbal children. \u003cem\u003eCochrane Database of Systematic Reviews\u003c/em\u003e, (11).\u003c/li\u003e\n\u003cli\u003eCallanan, J., Signal, T., \u0026amp; McAdie, T. (2021). What is my child telling me? Reducing stress, increasing competence and improving psychological well-being in parents of children with a developmental disability. \u003cem\u003eResearch in developmental disabilities\u003c/em\u003e, 114, 103984.\u003c/li\u003e\n\u003cli\u003eChen, G. X., \u0026amp; Yang, X. J. (2014). A study on conversational ability of autistic children. \u003cem\u003eChinese Journal of Special Education\u003c/em\u003e, 11, 45-52. (In Chinese)\u003c/li\u003e\n\u003cli\u003eCui, M., Ni, Q., \u0026amp; Wang, Q. (2023). Review of intervention methods for language and communication disorders in children with autism spectrum disorders. \u003cem\u003ePeerJ\u003c/em\u003e, 11, e15735.\u003c/li\u003e\n\u003cli\u003eDapretto, M., Davies, M. S., Pfeifer, J. H., Scott, A. A., Sigman, M., Bookheimer, S. Y., \u0026amp; Iacoboni, M. (2006). Understanding emotions in others: mirror neuron dysfunction in children with autism spectrum disorders. \u003cem\u003eNature neuroscience\u003c/em\u003e, 9(1), 28-30.\u003c/li\u003e\n\u003cli\u003eDawson, G., Rogers, S., Munson, J., Smith, M., Winter, J., Greenson, J., ... \u0026amp; Varley, J. (2010). Randomized, controlled trial of an intervention for toddlers with autism: the Early Start Denver Model. \u003cem\u003ePediatrics\u003c/em\u003e, 125(1), e17-e23.\u003c/li\u003e\n\u003cli\u003eFang, H., Yanling, R., Li, C., \u0026amp; Xiaoyan, K. (2019). Reliability and validity of the Chinese version of autism treatment evaluation checklist. \u003cem\u003eSichuan Mental Health\u003c/em\u003e, 32(6), 518. (In Chinese)\u003c/li\u003e\n\u003cli\u003eJespersen, A. E., Lumbye, A., Vinberg, M., Glenth\u0026oslash;j, L., Nordentoft, M., W\u0026aelig;hrens, E. E., ... \u0026amp; Miskowiak, K. W. (2024). Effect of immersive virtual reality-based cognitive remediation in patients with mood or psychosis spectrum disorders: study protocol for a randomized, controlled, double-blinded trial. \u003cem\u003eTrials\u003c/em\u003e, \u003cem\u003e25\u003c/em\u003e(1), 82.\u003c/li\u003e\n\u003cli\u003eGao, J., Song, W., Huang, D., Zhang, A., \u0026amp; Ke, X. (2025). The effect of game-based interventions on children and adolescents with autism spectrum disorder: A systematic review and meta-analysis. \u003cem\u003eFrontiers in Pediatrics\u003c/em\u003e, 13, 1498563.\u003c/li\u003e\n\u003cli\u003eGrynszpan, O., Weiss, P. L., Perez-Diaz, F., \u0026amp; Gal, E. (2014). Innovative technology-based interventions for autism spectrum disorders: a meta-analysis. \u003cem\u003eAutism\u003c/em\u003e, 18(4), 346-361.\u003c/li\u003e\n\u003cli\u003eKarst, J. S., \u0026amp; Van Hecke, A. V. (2012). Parent and family impact of autism spectrum disorders: A review and proposed model for intervention evaluation. \u003cem\u003eClinical child and family psychology review\u003c/em\u003e, 15, 247-277.\u003c/li\u003e\n\u003cli\u003eKumar, A. (2025). Unique models of embodied cognition and eco-social niches proposed to validate hypothesis of social attunement and mis-attunement with a focus on autism. \u003cem\u003eFrontiers in Psychiatry\u003c/em\u003e, 16, 1562061.\u003c/li\u003e\n\u003cli\u003eLaLonde, K. B., Jones, S., West, L., \u0026amp; Santman, C. (2020). An evaluation of a game-based treatment package on intraverbals in young children with autism. \u003cem\u003eBehavior Analysis in Practice\u003c/em\u003e, 13, 152-157.\u003c/li\u003e\n\u003cli\u003eLuo, J., Wang, M. C., Gao, Y., Zeng, H., Yang, W., Chen, W., ... \u0026amp; Qi, S. (2021). Refining the parenting stress index\u0026ndash;short form (PSI-SF) in Chinese parents. \u003cem\u003eAssessment\u003c/em\u003e, 28(2), 551-566.\u003c/li\u003e\n\u003cli\u003eMoraleda-Sepulveda, E., Pulido-Garc\u0026iacute;a, N., Loro-Vicente, N., \u0026amp; Santos-Muriel, N. (2025). Effectiveness of Intensive Linguistic Intervention in Autism Spectrum Disorder: A Case Study. \u003cem\u003eChildren\u003c/em\u003e, 12(2), 182.\u003c/li\u003e\n\u003cli\u003ePanceri, J. A. C., Freitas, \u0026Eacute;., de Souza, J. C., da Luz Schreider, S., Caldeira, E., \u0026amp; Bastos, T. F. (2021). A new socially assistive robot with integrated serious games for therapies with children with autism spectrum disorder and down syndrome: a pilot study. \u003cem\u003eSensors\u003c/em\u003e, 21(24), 8414.\u003c/li\u003e\n\u003cli\u003ePeretti, S., Pino, M. C., Caruso, F., \u0026amp; Di Mascio, T. (2024). evaluating the potential of immersive virtual reality-based serious games interventions for autism: A pocket guide evaluation framework. \u003cem\u003eEducation Sciences\u003c/em\u003e, \u003cem\u003e14\u003c/em\u003e(4), 377.\u003c/li\u003e\n\u003cli\u003eReichow, B., Barton, E. E., Boyd, B. A., \u0026amp; Hume, K. (2018). Early intensive behavioral intervention (EIBI) for young children with autism spectrum disorders (ASD). \u003cem\u003eCochrane database of systematic reviews\u003c/em\u003e, (10).\u003c/li\u003e\n\u003cli\u003eRezayi, S., Shahmoradi, L., \u0026amp; Tehrani-Doost, M. (2025). Systematic Review and Thematic Analysis of Digital Games for Cognitive Enhancement in Children with Autism Spectrum Disorder: Toward a Conceptual Framework. \u003cem\u003eCognitive Computation\u003c/em\u003e, 17(1), 60.\u003c/li\u003e\n\u003cli\u003eRimland, B., \u0026amp; Edelson, S. M. (1999). Autism treatment evaluation checklist. \u003cem\u003eJournal of Intellectual Disability Research\u003c/em\u003e.\u003c/li\u003e\n\u003cli\u003eSchopler, E., Reichler, R. J., DeVellis, R. F., \u0026amp; Daly, K. (1980). Toward objective classification of childhood autism: Childhood Autism Rating Scale (CARS). \u003cem\u003eJournal of autism and developmental disorders\u003c/em\u003e.\u003c/li\u003e\n\u003cli\u003eShaul, J. (2014). \u003cem\u003eThe Conversation Train: A Visual Approach to Conversation for Children on the Autism Spectrum\u003c/em\u003e. Jessica Kingsley Publishers.\u003c/li\u003e\n\u003cli\u003eSmith, T. (2001). Discrete trial training in the treatment of autism. \u003cem\u003eFocus on autism and other developmental disabilities\u003c/em\u003e, 16(2), 86-92.\u003c/li\u003e\n\u003cli\u003eSpain, D., \u0026amp; Blainey, S. H. (2015). Group social skills interventions for adults with high-functioning autism spectrum disorders: A systematic review. \u003cem\u003eAutism\u003c/em\u003e, 19(7), 874-886.\u003c/li\u003e\n\u003cli\u003eShaw, K. A. (2025). Prevalence and early identification of autism spectrum disorder among children aged 4 and 8 years\u0026mdash;Autism and Developmental Disabilities Monitoring Network, 16 Sites, United States, 2022. \u003cem\u003eMMWR. Surveillance Summaries\u003c/em\u003e, 74.\u003c/li\u003e\n\u003cli\u003eSundberg, M. L., Juan, B. S., Dawdy, M., \u0026amp; Arg\u0026uuml;elles, M. (1990). The acquisition of tacts, mands, and intraverbals by individuals with traumatic brain injury. \u003cem\u003eThe Analysis of verbal behavior\u003c/em\u003e, 8, 83\u0026ndash;99.\u003c/li\u003e\n\u003cli\u003eThiemann, K. S., \u0026amp; Goldstein, H. (2001). Social stories, written text cues, and video feedback: Effects on social communication of children with autism. \u003cem\u003eJournal of applied behavior analysis\u003c/em\u003e, 34(4), 425-446.\u003c/li\u003e\n\u003cli\u003eVirues-Ortega, J., P\u0026eacute;rez-Bustamante, A., \u0026amp; Tarifa-Rodriguez, A. (2022). Evidence-based Applied Behavior Analysis (ABA) autism treatments: An overview of comprehensive and focused meta-analyses. \u003cem\u003eHandbook of autism and pervasive developmental disorder: Assessment, diagnosis, and treatment\u003c/em\u003e, 631-659.\u003c/li\u003e\n\u003cli\u003eWatkins, C. L., Pack-Teixeira, L., \u0026amp; Howard, J. S. (1989). Teaching intraverbal behavior to severely retarded children. \u003cem\u003eThe Analysis of verbal behavior\u003c/em\u003e, 7, 69\u0026ndash;81.\u003c/li\u003e\n\u003cli\u003eWang, T., Ma, Y., Du, X., Li, C., Peng, Z., Wang, Y., \u0026amp; Zhou, H. (2024). Digital interventions for autism spectrum disorders: A systematic review and meta‐analysis. \u003cem\u003ePediatric Investigation\u003c/em\u003e, 8(3), 224-236.\u003c/li\u003e\n\u003cli\u003eWang, T., Ma, P. H., Ge, H., Sun, Y., Kwok, T. T. O., Liu, X., ... \u0026amp; Zhang, W. (2025). The use of gamified interventions to enhance social interaction and communication among people with autism spectrum disorder: A systematic review and meta-analysis. \u003cem\u003eInternational Journal of Nursing Studies\u003c/em\u003e, 105037.\u003c/li\u003e\n\u003cli\u003eXu, F., Gage, N., Zeng, S., Zhang, M., Iun, A., O\u0026rsquo;Riordan, M., \u0026amp; Kim, E. (2024). The use of digital interventions for children and adolescents with autism Spectrum disorder\u0026mdash;a Meta-analysis. \u003cem\u003eJournal of Autism and Developmental Disorders\u003c/em\u003e, 1-17.\u003c/li\u003e\n\u003cli\u003eZhang, M., Ding, H., Naumceska, M., \u0026amp; Zhang, Y. (2022). Virtual reality technology as an educational and intervention tool for children with autism spectrum disorder: current perspectives and future directions. \u003cem\u003eBehavioral Sciences\u003c/em\u003e, 12(5), 138.\u003c/li\u003e\n\u003cli\u003eZhang, Z. F., \u0026amp; Wang, H. P. (2005). Development of the Autism Children\u0026apos;s Behavior Checklist and related research. \u003cem\u003eJournal of Special Education Research\u003c/em\u003e, 28, 145-166. (In Chinese)\u003c/li\u003e\n\u003cli\u003eZimmerman, K. N., Ledford, J. R., Gagnon, K. L., \u0026amp; Martin, J. L. (2020). Social stories and visual supports interventions for students at risk for emotional and behavioral disorders. \u003cem\u003eBehavioral Disorders\u003c/em\u003e, 45(4), 207-223.\u003c/li\u003e\n\u003cli\u003eZung, W. W. (1971). A rating instrument for anxiety disorders. \u003cem\u003ePsychosomatics: Journal of Consultation and Liaison Psychiatry\u003c/em\u003e.\u003c/li\u003e\n\u003cli\u003eZeidan, J., Fombonne, E., Scorah, J., Ibrahim, A., Durkin, M. S., Saxena, S., ... \u0026amp; Elsabbagh, M. (2022). Global prevalence of autism: A systematic review update. \u003cem\u003eAutism research\u003c/em\u003e, 15(5), 778-790.\u003c/li\u003e\n\u003c/ol\u003e"}],"fulltextSource":"","fullText":"","funders":[],"hasAdminPriorityOnWorkflow":false,"hasManuscriptDocX":true,"hasOptedInToPreprint":true,"hasPassedJournalQc":"","hasAnyPriority":false,"hideJournal":true,"highlight":"","institution":"","isAcceptedByJournal":false,"isAuthorSuppliedPdf":false,"isDeskRejected":"","isHiddenFromSearch":false,"isInQc":false,"isInWorkflow":false,"isPdf":false,"isPdfUpToDate":true,"isWithdrawnOrRetracted":false,"journal":{"display":true,"email":"[email protected]","identity":"researchsquare","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":true,"externalIdentity":"","sideBox":"","snPcode":"","submissionUrl":"/submission","title":"Research Square","twitterHandle":"researchsquare","acdcEnabled":true,"dfaEnabled":false,"editorialSystem":"","reportingPortfolio":"","inReviewEnabled":false,"inReviewRevisionsEnabled":true},"keywords":"Autism spectrum disorder, ASD, intraverbal behavior, digital intervention, digital games","lastPublishedDoi":"10.21203/rs.3.rs-8504009/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-8504009/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"\u003cp\u003eIntraverbal deficits represent a core characteristic of children with Autism Spectrum Disorder(ASD), significantly impairing their social communication abilities. Traditional face-to-face interventions are often constrained by limited accessibility and insufficient generalization of skills, while current digital approaches rarely focus specifically on intraverbal behavior. This study aimed to develop a gamified digital intervention grounded in the \"Conversation Train\" framework and to conduct a systematic evaluation of its effectiveness. A total of 40 children with ASD, aged 5 to 10 years, were randomly assigned to either an experimental group or a control group. The experimental group received standard training supplemented with 30 sessions of gamified digital intervention(15 minutes per day, five days per week), whereas the control group received standard training only. Outcome assessments included standardized instruments—the Autism Treatment Evaluation Checklist(ATEC) and the Communication Behavior Checklist (CB)—as well as a researcher-developed intraverbal behavior scenario test comprising two subtests: dialogue selection and real-time dialogue performance. No significant differences were observed between groups in baseline demographic or clinical characteristics (all p\u0026gt;0.05). Following the intervention, the experimental group showed significantly lower ATEC scores (Z=-3.551,p\u0026lt;0.001), higher CB scores (Z=-2.943,p=0.003), and improved performance on both subtests of the scenario test (dialogue selection: Z=-4.097, p\u0026lt;0.001; real-time dialogue: Z=-3.758,p\u0026lt;0.001), with moderate to large effect sizes (r = 0.49–0.97) across all measures. In contrast, the control group exhibited no significant changes. These findings indicate that the proposed gamified digital intervention effectively enhances intraverbal behavior and reduces core symptoms of ASD in children, offering a feasible, accessible, and scalable alternative to conventional face-to-face interventions.\u003c/p\u003e","manuscriptTitle":"Digital Intervention Training for Intraverbal Behavior in Children with Autism Spectrum Disorder","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2026-02-19 09:11:39","doi":"10.21203/rs.3.rs-8504009/v1","editorialEvents":[{"type":"communityComments","content":0}],"status":"published","journal":{"display":true,"email":"[email protected]","identity":"researchsquare","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":true,"externalIdentity":"","sideBox":"","snPcode":"","submissionUrl":"/submission","title":"Research Square","twitterHandle":"researchsquare","acdcEnabled":true,"dfaEnabled":false,"editorialSystem":"","reportingPortfolio":"","inReviewEnabled":false,"inReviewRevisionsEnabled":true}}],"origin":"","ownerIdentity":"163ae164-3f6a-4921-afc7-35968abfeadd","owner":[],"postedDate":"February 19th, 2026","published":true,"recentEditorialEvents":[],"rejectedJournal":[],"revision":"","amendment":"","status":"posted","subjectAreas":[],"tags":[],"updatedAt":"2026-04-01T14:10:34+00:00","versionOfRecord":[],"versionCreatedAt":"2026-02-19 09:11:39","video":"","vorDoi":"","vorDoiUrl":"","workflowStages":[]},"version":"v1","identity":"rs-8504009","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-8504009","identity":"rs-8504009","version":["v1"]},"buildId":"XKTyCvWXoU3ODBz1xrDgd","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}