INTRODUCTION
Autism spectrum disorder (ASD) is a neurodevelopmental disorder characterized by persistent difficulties in reciprocal social communication and interaction accompanied by limeISSN 1976-3026 OPEN ACCESS ited and repetitive patterns of behavior and interests [1] During development, social communication—the ability to exchange information and emotions through verbal and non-verbal means—is a critical and persistent challenge [2-4]. Although early research [5-7] found no specific deficits in structural language (e.g., phonology, syntax, or vocabulary) associated with social communication in patients with ASD, pragmatic language impairments (i.e., the appropriate social use of language) are consistently observed, even when structural language skills appear intact. Social communication refers to the ability to use language and non-verbal cues to share information, and feelings with others, requiring cognitive skills such as understanding that communication is a means of interpersonal connection [8,9].
Language can be broadly divided into two main components: structural language and non-verbal language, the latter encompassing pragmatic language such as emotional reciprocity, forming relationships, joint attention, and interpreting facial expressions [2-4]. In ASD, symptoms range from a complete absence of language to mild language delays or deficits. Among these, pragmatic impairments are particularly persistent throughout development, emphasizing the need to consider structural and pragmatic components as complementary rather than contrasting [10,11]. While structural language deficits may improve with age or targeted interventions such as speech therapy, pragmatic language impairments tend to become more pronounced over time and remain a major challenge [12]. Consequently, pragmatic language skills are critical determinants of effective social communication [13].
ASD diagnoses in children have been increasing worldwide [14]. While pharmacological options are available, they are largely limited to reducing repetitive behaviors, and no medication has proven effective in improving primary difficulties in ASD, such as socialization and communication [2,15]. As a result, nonpharmacological therapies for children with ASD, including speech therapy and applied behavioral analysis (ABA), remain central for improving language and communication skills. Therapeutic interventions should begin promptly after diagnosis, since early treatment can enhance neuroplasticity and positively influence developmental outcomes in young children [14,16,17]. However, despite the importance of early intervention and long-term care, these traditional therapeutic treatments impose significant economic strain and place additional burdens on families, in part owing to the spatial and temporal constraints of in-person sessions [18,19].
In contrast, compared with traditional therapeutic treatments, digital therapeutics do not impose spatial and temporal constraints, enabling patients to access therapy in their own homes via tablets or smartphones [20]. Moreover, mobile forms of therapy that function at home may provide an affordable method to meet the demand for early intervention in children with ASD [21] and also support continuity of care, which are critical challenges in traditional in-person sessions. Thus, increasing attention has been paid to digital therapeutic devices that can address the limitations of conventional treatment methods. However, despite this growing interest, few studies have explored the use of digital technology to enhance social communication skills in children with ASD, and many lack a foundation in child development theory [21,22].
Therefore, the present pilot trial aimed to verify the effectiveness of the “AVATA Cure” treatment device in improving the communication skills of children with ASD. The device was developed based on the principles of augmentative and alternative communication and evidence-based therapist-centered therapy, with modules designed to address both structural language and non-verbal communication. The program emphasized social communication, one of the central difficulties for children with ASD. Early studies regarding digital interventions demonstrated that emotion recognition training increased social interaction and understanding in children with ASD [23]. Additionally, the integration of gamification elements, such as scoring systems and positive feedback, enhances engagement and motivation, enabling children to practice social communication in a structured yet enjoyable environment [24].
Using the AVATA Cure, children with ASD can interact through speech and touch responses and practice joint attention by following the gaze of animated characters. These features are designed to facilitate social communication activities and strengthen interaction skills through digital therapy. By providing consistent practice opportunities while reducing the need for frequent in-person sessions, the AVATA Cure offers a potentially cost-effective and scalable solution for families seeking early intervention [25]. Thus, this study aims to evaluate the effectiveness of AVATA Cure in improving social communication, a key challenge for children with ASD, in a cohort of Korean children, demonstrating its potential as an accessible, structured intervention that complements traditional therapies.
METHODS
Participants
Between July and December 2023, 33 children with ASD were recruited from the Department of Psychiatry, CHA Bundang Medical Center (Seongnam, Republic of Korea). The inclusion criteria were: 1) age 1-7 years; 2) diagnosis of ASD based on detailed psychiatric interviews and Diagnostic and Statistical Manual of Mental Disorders, Fifth Edition (DSM-5) criteria; and 3) ability to use mobile applications on tablet PCs. The exclusion criteria were: 1) serious medical, sensory, or motor disabilities, and 2) reluctance or discomfort in using tablet PCs.
All study procedures were approved by the Institutional Review Board of CHA Bundang Medical Center (no. 2023-04-044-002). Participants and/or their guardians received a full explanation and a written description of the study procedures, and provided written informed consent before study initiation. In the case of children under 6 years of age, guardian consent was obtained; for children over 6 years of age with decision-making capacity, consent was obtained from both the participant and their guardian.
Of the 33 enrolled participants, three did not complete the study, and one had missing data at the second visit. Consequently, the analysis included 29 participants who completed all assessments.
Study design
At the first visit, a psychiatrist confirmed each child’s eligibility for the study. Eligible children and their guardians were given a comprehensive description of the study procedures and signed informed consent forms. A researcher trained in clinical psychology assessed the children’s communication skills using the Autism Diagnostic Observation Schedule-2 (ADOS-2); Sequenced Language Scale for Infants (SELSI) or Preschool Receptive-Expressive Language Scale (PRES); Social Communication Questionnaire (SCQ); and Korean Vineland Adaptive Behavior Scales 2nd edition (K-Vineland-II). The participants then received a tablet to use the digital AVATA Cure application at home for 8 weeks. Follow-up assessments of the participants’ communication skills were conducted 4 and 8 weeks from the date of their first visit by the same researcher. The Consolidated Standards of Reporting Trials (CONSORT) flow diagram of the study participants is shown in Figure 1.
Intervention: digital therapeutics (AVATA Cure) application
The AVATA Cure application was the digital intervention applied in the present study and was delivered via a digital tablet. This application includes several modules targeting social communication abilities, such as joint attention and receptive, expressive, as well as pragmatic language (Figure 2A). For example, the joint attention module trains children to follow the gazes of characters and interact with specific objects. The receptive language module uses visual cues and simple instructions to improve comprehension, whereas the expressive language module uses visual aids to encourage children to form words or sentences. The pragmatic language module focuses on context-appropriate communication by simulating real-world conversational scenarios.
The application follows a predesigned protocol tailored for children with ASD, enabling interaction through speaking, touching, and other reactions on the tablet. Given the common ASD characteristics of restricted behaviors, specific interests, and hyperfocus, some children may be at risk for side effects such as overindulgence [26,27]. To mitigate this risk, AVATA Cure includes limits on maximum daily usage and session duration. Individual app usage times were recorded for each participant (Figure 2B). Additional details of the therapeutic intervention are provided in the Supplementary Figure 1.
All participants received a tablet preloaded with the AVATA Cure application at the first visit and began using it immediately. Participants and their parents were advised to engage with AVATA Cure for at least 30 min per day, 4 days per week, for a total of 960 min over the 8-week intervention period. Most children receive autism-related therapies, such as speech therapy, ABA, and sensory integration therapy. Participants and their guardians were instructed to maintain any ongoing autism-related therapies that had been initiated prior to study enrollment without substantial changes until the end of the study period. Information on concurrent therapies was collected via guardian reports and included the type of therapy, approximate duration of each session, and frequency of sessions during the study period. These concurrent therapies were part of routine clinical care, varied according to individual clinical needs, and were not standardized by the study protocol.
Social communication skill assessments
All participants responded to four standardized clinical outcome measures, including ADOS-2, SELSI or PRES, SCQ, and K-Vineland-II.
ADOS-2
The ADOS-2 is a semi-structured, standardized, observational instrument that uses questions and activities to evaluate communicative, social, and repetitive behaviors relevant to ASD diagnosis. It comprises five modules: Toddler, 1, 2, 3, and 4, which are selected primarily based on the individual’s expressive language level and secondarily on chronological age and task relevance to the individual’s interests and abilities. The Toddler Module is designed for preverbal children or those using single words aged 12-30 months, Module 1 for preverbal or single-word users aged 31 months and older, Module 2 for children using phrase speech, Module 3 for children or adolescents with fluent speech, and Module 4 for adolescents or adults with fluent speech. The evaluator determines the appropriate module for each individual based on the manual’s criteria, and scores are entered into a normed algorithm to assess ASD likelihood and symptom severity, with lower severity scores indicating less impairment [14]. The present study used the “reciprocal social interaction” total score as part of the social affect category. Previous studies have demonstrated strong internal consistency across all modules, with Cronbach’s alpha values ranging from 0.82 to 0.91 [28].
SELSI or PRES
The SELSI and PRES are scales used to measure language development, with the choice of scale determined according to the participants’ age and language level. SELSI was administered to children aged 4-35 months or to those older than 35 months whose language skills were low enough to warrant this scale. PRES was used for children aged 2-6 years. Both instruments evaluate the detailed aspects of language. In the present study, the receptive and expressive “pragmatics” scores were used to assess the participants’ ability to understand and express pragmatic understanding and expression by understanding the communication situation or the flow of the story, as outlined in the SELSI and PRES manuals [29,30]. The SELSI demonstrates strong internal consistency, with Cronbach’s alpha values of 0.99 for both receptive and expressed domains [30]. Similarly, the PRES shows strong internal consistency, with Cronbach’s alpha values of 0.95 for both receptive and expressive domains [29].
SCQ
The SCQ is used to evaluate communication skills and social functioning in children [31]. It comprises 40 items that comprehensively assess various aspects of social interaction abilities as well as verbal and non-verbal communication skills. Each item is answered with “yes” or “no.” In the present study, the SCQ was completed by the guardians of the children, and the resulting scores were used for analysis. Higher scores indicated greater communication deficits. Previous studies have demonstrated the strong internal consistency of this instrument, with a Cronbach’s alpha reaching 0.93 [32].
K-Vineland-II
The K-Vineland-II assesses adaptive behavior and personal and social functioning across four domains that assess each developmental step in communication, socialization, daily liv-ing skills, and motor adaptive skills. Each domain consists of subdomains (receptive, expressive, and written communication; personal, domestic, and community daily living skills; interpersonal relationship, play and leisure time, and coping skills; gross and fine motor skills) [33]. According to the K-Vineland-II manual, the “interpersonal relationship” and “play and leisure time” subdomains of the socialization scale include items that measure the ability to understand and use non-verbal communication to manage social interactions, capturing qualitative impairments in social interaction characteristic of children with ASD [34]. The present study collected V-scale scores for these two subdomains. Overall, the K-Vineland-II demonstrates strong internal consistency, with a Cronbach’s alpha value ranging from 0.96 to 0.99 [34].
Statistical analysis
To evaluate the effectiveness of AVATA Cure, changes in children’s communication skill evaluation scores were assessed at three time points: baseline (0-week; V1), mid-intervention (4-week; V2), and post-intervention (8-week; V3). Statistical analyses were performed using data from the 29 participants who completed all assessments (complete case analysis). Analyses were conducted using IBM SPSS Statistics for Windows, version 29.0 (IBM Corp.), with statistical significance set at p <0.05.
The Shapiro-Wilk test was used to test the normality of the data. For normally distributed data, changes in measurement scores over time were assessed using one-way repeated-measures analysis of variance (ANOVA). Sphericity was assessed using Mauchly’s test for one-way repeated-measures ANOVA, and when the assumption was violated, Greenhouse-Geissercorrected degrees of freedom were applied and reported. If the data were not normally distributed, the Friedman test examined overall differences, pairwise comparisons between visits were conducted using the Wilcoxon signed-rank test. Posthoc analyses were adjusted using the Bonferroni correction.
RESULTS
Demographic and clinical characteristics of the study participants
Table 1 shows the demographic and clinical characteristics of the study participants, including AVATA Cure total usage time and other autism-related therapy times, such as speech, ABA, and sensory integration therapy. The participants included 20 males (68.97%) and 9 females (31.03%) with a median age of 4 years (range: 1-6 years). Among the participants, 22 (75.86%) were verbal and 7 (24.14%) were non-verbal. The participants used the AVATA Cure application for a median of 13.8 hours (range: 2.05-32.83 hours) through the 8-week treatment period. In addition, the participants spent a median of 16.3 hours (range: 0-80.5 hours) on other therapies.
Changes in social communication skills across visits
Table 2 and Figure 3 present the changes in social communication-related measures across the three visits (0-week, V1; 4-week, V2; 8-week, V3) highlighting whether significant improvements occurred over time following the AVATA Cure digital therapeutics intervention. In addition to statistical significance, effect sizes were examined to quantify the magnitude of change over time; the effect size estimates are presented in Table 2. Exploratory dose-response analyses did not reveal statistically significant associations between changes in clinical outcomes and AVATA Cure app usage time or concurrent therapy intensity (Supplementary Table 1). Exploratory subgroup analyses based on verbal status are also presented (Supplementary Table 2).
ADOS-2 reciprocal social interaction
The Friedman test revealed a significant change in ADOS-2 scores (χ²=8.51, p=0.014), indicating notable differences in scores between visits. The Wilcoxon signed-rank test with Bonferroni correction was conducted to examine differences between each visit, the ADOS-2 “reciprocal social interaction” score significantly decreased from V1 (median=8, range=2-19) to V3 (median=7, range=1-17), indicating a positive effect.
SELSI/PRES receptive and expressive pragmatics
The Friedman test revealed a significant increase in the SELSI/PRES “receptive pragmatics” scores (χ²=29.62, p<0.001), indicating differences in scores between visits. Post-hoc analyses using the Wilcoxon signed-rank test with Bonferroni correction demonstrated a statistically reliable increase from V1 (median=4, range=1-13) to V2 (median=6, range=1-14) and from V2 to V3 (median=6, range=1-13). Similarly, “expressive pragmatics” scores significantly increased (χ²=22.61, p< 0.001), and post-hoc analysis with Bonferroni correction, demonstrated a significantly increased from V1 (median=6, range=2-19) to V2 (median=8, range=2-20) and from V2 to V3 (median=8, range=2-21). These findings indicated improvements both receptive and expressive pragmatic abilities.
SCQ
One-way repeated-measures ANOVA to examine the differences in SCQ scores revealed a significant main effect of time (F(2, 56)=8.61, p<0.001). Post-hoc analysis with Bonferroni correction revealed a significant decrease in V3 from V1. This indicated an improvement.
K-Vineland-II interpersonal relationship and play and leisure time
One-way repeated-measures ANOVA revealed a significant main effect on the K-Vineland-II “interpersonal relationship” score over time. Post-hoc analysis with Bonferroni correction revealed a significant increase from V1 to V3 (F(2, 56)=3.75, p=0.030). Furthermore, the “play and leisure time” scores also showed a significant main effect. Because Mauchly’s test indicated a violation of the sphericity assumption for this subdomain, the Greenhouse-Geisser correction was applied. Post-hoc analysis with Bonferroni correction revealed significant increases from V1 to V2 and V1 to V3 (F(1.57, 43.88)=10.04, p<0.001). Thus, both “interpersonal relationship” and “play and leisure time” scores showed improvements.
DISCUSSION
This pilot clinical trial is the first in Republic of Korea to evaluate the feasibility and preliminary efficacy of AVATA Cure, a mobile-based digital therapeutic, in enhancing social communication skills in children with ASD. Across the 8-week intervention, significant improvements were observed in multiple measures of social communication, including reciprocal social interaction, caregiver-reported communication, and adaptive social functioning. These findings suggest the potential of AVATA Cure as a feasible and scalable adjunct to conventional therapies in children with ASD.
A notable outcome was an improvement in reciprocal social interaction, as measured by ADOS-2. Gains in this domain suggest that children with ASD are better able to initiate and maintain shared exchanges such as eye contact, joint engagement, and affective reciprocity, which are essential foundations for later social development. Pragmatic language abilities, including both receptive and expressive domains in the SELSI/PRES, also significantly increased, indicating enhanced understanding and expression in social communication through comprehension of situational contexts or narrative flow. Receptive language improvements indicate a better grasp of communication situations or the flow of a story, whereas expressive language advancements indicate a better ability to convey thoughts and engage in conversations. These findings are consistent with the notion that pragmatic language deficits in children with ASD are key challenges that can be addressed through structured interventions [12,35]. The SCQ score also significantly decreased, indicating a positive improvement. Additionally, both interpersonal relationship and play and leisure time scores on the K-Vineland-II significantly increased, indicating improvements. These findings are consistent with earlier research showing that better social communication skills positively influence broader socialization and daily living skills [33,36].
Previous studies on digital therapeutics for autism have demonstrated the association between higher compliance with game-based mobile therapy and improved social communication [21]. Our results suggest that AVATA Cure, by encouraging interactions through speaking, touching, and joint attention via gaze-following of characters across multiple modules, may support the development of social communication skills in children with ASD. For instance, modules focusing on the joint attention of AVATA Cure may strengthen children’s ability to coordinate attention with social partners by repeatedly engaging them in behaviors, such as sustained gaze, gaze shifting, and pointing, which are considered foundational skills for social communication [37,38]. Moreover, activities involving AVATA Cure’s conversation and theory of mind-related modules may support pragmatic language development by providing structured opportunities for turn-taking and understanding others’ intentions, which are important for social communication skills [8,39]. However, the mechanisms underlying these improvements remain unclear, highlighting the need for further research to examine how AVATA Cure enhances social communication, potentially by targeting specific cognitive or social processes. Previous studies have linked digital therapeutics with increased adaptability and communication competencies in children with ASD [3]. While some digital interventions have been reported to enhance socialization, they mainly target emotion recognition, and others have not demonstrated improvements in communication outcomes. Our study is novel for its direct focus on pragmatic and functional social communication, demonstrating positive effects in this critical domain [3,40].
Socialization and communication are critical skills for children with ASD [2]. Previous studies indicate that children with ASD experience more pronounced impairments in social and pragmatic abilities compared with other developmental disorders [13]. The improvements observed in pragmatic language in the present study suggest that the AVATA Cure digital intervention, which focused on social communication skills, may specifically target these pragmatic deficits. Currently, no effective pharmacological treatments are available to improve the social communication skills of children with ASD, and traditional in-person therapies face limitations such as spatial, temporal, and economic constraints. Digital therapeutics overcome these barriers by enabling children to access interventions at home via tablets or smartphones [20,41]. This flexibility is particularly valuable for early intervention, providing an affordable and scalable approach to meet the growing demand for ASD therapies [21].
Several limitations should be acknowledged in this study. First, the sample was relatively small, and no control group was included, which limited the ability to draw causal inferences. Therefore, it was not possible to conclusively attribute the observed improvements solely to the intervention in our study, as they may be attributable to natural developmental changes or concurrent therapies. Furthermore, the relatively small sample size, restricted age range, and limited variation in participants’ language abilities may constrain the generalizability of our findings given the diversity of the ASD population. Therefore, future studies with larger and more diverse samples of children with ASD could enable subgroup analyses at the language level (e.g., verbal vs. nonverbal children). In the future, randomized controlled trials with larger sample sizes are required to corroborate these findings. Nonetheless, the potential of AVATA Cure as a complementary tool alongside other therapies should be further explored, including its specific impact and interactions with traditional interventions. Second, some assessments were based on the guardians’ subjective reports, which may be subject to response bias. To enhance the reliability of the assessment results, standardized observational instruments such as the ADOS-2 and interviewtype K-Vineland-II were also administered by a trained psychologist.
Despite these limitations, the significant improvements observed across multiple clinical measures provide preliminary evidence that AVATA Cure can improve social communication in children with ASD, complement existing therapies, and offer a scalable and accessible treatment option. Although some mean or median changes were modest, the observed effect sizes suggested that these improvements may be clinically meaningful in the context of a short-term pilot study involving children with ASD. Notably, in some assessments in this study, such as the SELSI/PRES, both receptive and expressive pragmatic language showed significant improvement over a 4-week period, with further improvement observed at the 8-week mark. Compared with traditional face-to-face therapies, which can be time-consuming and financially burdensome, AVATA Cure offers a practical alternative by reducing in-person session costs. Previous studies have demonstrated that digital interventions can significantly lower therapy-associated costs while maintaining accessibility for families in geographically isolated or underserved areas [19,20]. These economic advantages, combined with the targeted focus on social communication skills, suggest that AVATA Cure could be particularly beneficial for families facing financial and logistical challenges in accessing early interventions. Moreover, previous research has emphasized the importance of early intervention in children with ASD, as well as the financial burden of long-term and continuous treatment [14,16-18]. While AVATA Cure was effective for a relatively short period in the present study, it may also offer a cost-effective solution for maintaining long-term therapeutic benefits.
Taken together, these findings suggest that AVATA Cure could be used as an adjunctive therapy in clinical practice, alongside pre-existing interventions for ASD (e.g., speech therapy and ABA) to reinforce social communication skills between in-person sessions and across home or educational settings. Although there are no standardized guidelines for the clinical implementation of digital therapeutics for ASD, prior studies have emphasized the importance of providing caregivers with clear instructions regarding device use and appropriate intervention duration [40,42]. Therefore, clinicians should review the intended therapeutic targets and safety considerations (e.g., appropriate intervention duration and potential overuse) and provide guardians with guidance on supervised use and procedures for reporting concerns.
In conclusion, the results of this pilot study provide preliminary evidence that the AVATA Cure can enhance social communication skills in young children with ASD and complement traditional therapies. Notably, improvements were detectable as early as 4 weeks, suggesting that short-term interventions may have a meaningful impact. Future randomized controlled trials with larger samples, extended follow-up periods, and mechanistic analyses are warranted to confirm the efficacy of digital therapeutics and explore how they facilitate social communication gains. If validated, the AVATA Cure has the potential to evolve into a scalable and cost-effective tool for improving social communication, thereby addressing one of the most critical challenges in ASD intervention.
PDF Links
PubReader
ePub Link
XML Download
Full text via DOI
Download Citation
Supplement1
Print
