The use of the social robot NAO to enhance communication skills in children with autism spectrum disorder: A systematic review of studies from 2013–2019 to 2020

Abstract

Although numerous studies have examined the field of Socially Assistive Robots (SARs), less attention has been paid to evaluating the role of the NAO robot in enhancing the communication skills of children with autism spectrum disorder (ASD) across two distinct time periods (2013–2019 and 2020–2025), highlighting advancements in both SAR-based interventions and NAO’s technological capabilities. A systematic search was conducted across international databases, using strict inclusion and exclusion criteria, resulting in the selection of 44 relevant studies. This review synthesizes findings on the types of interventions implemented, NAO’s functional characteristics, and its effectiveness in facilitating social communication. The results indicate a positive impact on children’s communication abilities, with more recent studies (2020–2025) reporting greater improvements, possibly due to advancements in NAO’s capabilities and the refinement of intervention methodologies. To the best of our knowledge, this is the first systematic review to compare NAO-assisted interventions across two different periods, offering novel insights into the evolving role of SARs in autism therapy.

Keywords

robot NAO autism spectrum disorder types of interventions characteristics of the NAO roles of the NAO

Introduction

Socially Assistive Robots (SARs)/NAO

The history of robotics is interdisciplinary, as the branches of engineering, politics, philosophy, and learning theories intersect.¹ The word “robot” was first used in literature in a science fiction work titled Rossum’s Universal Robots.² In 1962, the first industrial robots used in engineering were introduced at Ford’s Canton automobile plant to increase production.³ According to Roshidul et al.,⁴ there are five generations of robotics: Robots, robotic arms, walking robots, behavior-based robots, and humanoid robots. In particular, humanoid robots could also be called Socially Assistive Robots (SARs) due to their ability to simulate empathy by mimicking human movements and understand emotions when programmed.⁵ Socially Assistive Robots (SARs) are robotics platforms with audio, video, and motion capabilities.⁶ Robotics has been introduced in recent years in the field of education through interventions implemented in the school context.⁷ More specifically, robotics is linked to the pedagogical theory of constructivism, according to which children learn through their experiences and actively participate in the creation of knowledge.⁸

A significant number of studies have been implemented over the last 10 years on the utilization of SARs and specifically for the NΑΟ. In particular, the NAO robot has a small size, namely, 58 cm in height, two legs, and “rounded” characteristics.⁹ Furthermore, the NAO has 25 liberty degrees, speaks 20 languages, and has low cost.¹⁰ The NAO was designed by Soft Bank Robotics (formerly Aldebaran) in 2006, and continuous upgrades have been made.¹¹ In addition, it is equipped with four directional microphones and speakers for better human-user interaction.¹² The NAO robot rotates its head, creating eye contact with the user, and at the same time shows slow movements of the hands and body, significantly increasing its interactivity with the user.¹³ The NAO, moreover, presents special characteristics, such as advanced communication, display of physical cues, and human face recognition.¹⁴ These qualities are important for enhancing the interaction of the person with others, in meeting the needs of everyday life and in effective communication with those around them.¹⁵

The use of SARs, especially the NAO, seems to improve the social and communication skills of children with special educational needs, as studies in recent years have shown.¹⁶ The World Health Organization (WHO) reports that 10% of the world’s population has a disability, a fairly high proportion.¹⁷ In particular, 14% of the population in the USA and at least 18% in Europe have been diagnosed with some type of disability.¹⁸ The number of studies on the use of social robots in special education is small globally. The majority of research on the use of social robots in special education is on children with autism spectrum disorder (ASD), while there are far fewer studies on other cases.¹⁹ The term “autism spectrum disorder” (ASD) refers to a wide range of symptoms, ranging from mild to more severe.²⁰ More specifically, ASD is a long-term neurodevelopmental disorder that manifests in childhood and is characterized by a lack of social interaction and repetition of an unusual range of sensorimotor behaviors.²¹ The two main symptoms of ASD are limited and repetitive sensorimotor behaviors and limited social interaction.²²

Interventions using NAO

There are two literature reviews published in related literature in international journals with significant impact factor to date on the use of NAO and refer to interventions implemented in typically developing children and older people.^23,24 First, a literature review by Robaczewski et al.²⁴ collected the interventions that have been conducted using the NAO in order to evaluate its potential as SARs. For this purpose, the researchers distinguished six categories: social interaction, emotionality, intervention, assisted teaching, mild cognitive impairment, and intellectual disability. The results showed the effectiveness of NAO in the majority of cases. Second, the literature review by Amirova et al.2²³ presents a summary of qualitative and quantitative research over the decade (2010–2020) on the potential of the NAO, the ways of communicating with children of typical and elderly individuals, its use in teaching, social skills development, and care of elderly individuals, as well as in human-robot interaction experiments in which the NAO was involved.

Despite the growing body of research on the applications of NAO in various populations, no systematic literature review has specifically examined its role in developing communication and social skills in children with special needs. Existing reviews have primarily focused on typically developing children and older adults, overlooking their impact in special education settings. Additionally, no study has comparatively analyzed interventions across different time periods to assess potential advancements in the effectiveness of NAO-based interventions. Given the increasing adoption of NAO in special education and the technological upgrades introduced in recent years, understanding how its impact has evolved over time is crucial. This review aims to bridge these gaps by systematically analyzing studies from two distinct time periods (2013–2019 and 2020–2025), providing insights into both the effectiveness of NAO-assisted interventions and their progression over time.

Study aims

According to the study, the primary aim of this systematic review is to evaluate the effectiveness of NAO-assisted interventions in fostering communication skills in children with ASD, with a focus on two distinct time periods: 2013–2019 and 2020–2025. This division is not arbitrary but based on key developments in the field. First, the integration of social robots in special education significantly expanded after 2019, leading to an increase in studies and potentially more refined intervention methodologies.²⁵ Additionally, from 2020 onwards, the COVID-19 pandemic accelerated interest in educational technologies, including social robots, which correlated with a rapid increase in research on their applications.²⁶ This thematic review aims to understand the effectiveness of the NAO social robot and not to generalize the results to other humanoid robots. More specifically, the aim is to understand whether NAO interventions are effective and to interpret when and why they are effective.

Technological advancement of the NAO robot is a crucial factor in this division. In 2019–2020, SoftBank Robotics released enhanced versions of NAO featuring improved natural language processing, greater computational power, and advanced sensory capabilities.²⁷ The technological advancements of NAO, particularly in natural language processing and sensory capabilities, have likely facilitated more effective interactions with children with ASD, contributing to the significantly greater improvements in communication skills reported in recent studies (2020–2025).²⁸ This analysis will assess whether these technological advancements and the increasing familiarity of educators and researchers with social robot-based interventions contributed to more impactful outcomes.

Specifically, the review examines the following dimensions across two distinct periods: 2013–2019 and 2020–2025:

(1) What types of intervention with the NAO robot have been used to develop communication skills in children with ASD, and what is their effectiveness?

(2) Which specific characteristics of the NAO (movements, sounds, and facial expressions) are most often used in interventions, and how do they affect the development of communication skills?

(3) What is the role of the NAO robot in interventions to support communication in children with ASD (e.g., interlocutor, mediator, and imitator)?

Through the study of the existing literature, the review aims to identify effective interventions and best practices in the use of ΝΑΟ, as well as to highlight how ΝΑΟ is a useful tool for special education, while identifying potential gaps in the literature and suggesting directions for future research.

Materials and methods

A systematic review was conducted to collect, select and comparatively analyze the effectiveness of different intervention methods using the NAO social robot to develop communication skills in children with ASD. The collection and analysis of a wide range of research articles seem to enhance the development of new theories, making the above research method a powerful approach.²⁹ To present a comprehensive and well-structured systematic review, the research was based on the PRISMA methodology (Preferred Reporting Items for Systematic Reviews and Meta-Analyses).^30,31

Search strategy

The Scopus, Google Scholar, Web of Science, Wiley Library, and high impact databases were used for this systematic review. Two limitations exist: the date limitation between 2013 and 2025 and the language limitation, which was English. In addition, the Springer, Taylor, and Francis journals were examined in the literature. In particular, we focused on the International Journal of Social Robotics (published by Springer), where we found articles related to the topic of the review. However, the articles that will be presented and examined below were collected from the Google Scholar, Scopus, and Wiley Library databases, because in the other databases searched for studies related to our review, no relevant papers were found as articles. The following terms were used to search for articles (Table 1).

Table 1.

Finding NAO intervention articles with the appropriate phrases.

Research strategy	Key term
Google Scholar	(Social robot NAO and interventions on children with autism spectrum for cultivating communication skills) AND (Interventions with social robot NAO and children with autism spectrum )
Scopus	(Social robot NAO and children with autism spectrum) AND (Interventions for children in the autism spectrum utilizing the NAO)
Wiley Library	(Social robot NAO and children with autism spectrum) AND (Interventions with the use of the NAO robot for the development of communication skills of children with autism spectrum disorders)

Eligibility criteria

Screening of research articles with specific inclusion and exclusion criteria was the second phase of the systematic review (Figure 1). The criteria used for the exclusion of studies were as follows:

(1) Research papers presented at conferences. Papers published in conference proceedings were not examined but in papers submitted to scientific journals because full peer review is provided and the research is presented with the required depth in methodology, results, and discussion, enhancing the reliability and validity of our review.

(2) Books and book chapters on social robotics.

(3) Research papers on the attitudes of teachers and parents towards the social robot NAO.

(4) Research papers using social robot NAO to diagnose autism.

(5) Research papers in which the social robot NAO is used as a tutor for teaching school lessons.

(6) Systematic review of the literature on the use of social robots.

Figure 1.

PRISMA flow diagram for study selection.

Studies to be included based on criteria:

(1) Studies published within a specific time frame (from 2013 to 2019 and from 2020 to 2025).

(2) Studies written in English.

(3) Studies in which the social robot NAO was used as a teaching assistant for children in the autism spectrum.

(4) Studies on NAO social robot used to develop the communication skills of children in the autism spectrum.

(5) Studies that included all levels of education (preschool, primary, and secondary education).

Study selection, data extraction, and quality assessment

The database research was conducted from January to March 2025. Initially, 1.329 articles were found in all three databases and in the two time periods searched (Figure 1). In the first stage, 134 duplicate studies were removed. In the second stage, the abstracts of the 1.195 remaining articles were examined in terms of inclusion and exclusion criteria, resulting in 92 articles for further evaluation. Following detailed screening, 1.103 studies were excluded due to the irrelevance of their scope. In particular, there were 287 systematic reviews and systematic literature reviews on the use of social robots, 71 book chapters and books on social robots and specifically the NAO robot, 569 articles presented at conferences, 75 articles on the attitudes of teachers and parents toward the social robot NAO, 82 articles where the social robot NAO was a tutor during teaching, and 16 articles used the social robot NAO to diagnose autism. At this stage, 9 papers were removed because of language restriction. In the third stage, the full texts of the 83 studies were fully examined against the inclusion and exclusion criteria. The result of the third and final stage was to produce 44 studies that were fully in line with the aim of our research.

The researchers used a predetermined form to extract and collect data. This form included information about the researchers in each study, forms of intervention, types of communication skills, role of the NAO, and specific characteristics of the NAO. Quality is ensured through objectivity in the selection of sources, the use of reliable and valid data, and the systematic and comprehensive coverage of the topic. More specifically, the initial review and selection of articles was carried out by all members of the writing team independently. Any discrepancies related to the eligibility of articles were resolved through discussion and agreement among all of us, thus ensuring that a consensus was reached. Accuracy for data extraction was ensured through cross-checking by at least three team members for each extracted item, thus minimizing the potential for errors and at the same time enhancing the reliability of the data. In addition, the method used in this narrative research was the synthesis of a systematic qualitative analysis of the findings. Finally, no meta-analysis was conducted owning to the different methodological approaches used in the research evaluated.

The methodological quality of the included studies was assessed using the CASP (Critical Appraisal Skills Programme) checklist for systematic reviews, which helps to assess the validity, reliability, and transparency of the methodology of the included studies.³²

Results

General characteristics of included studies

Before analyzing the main results of this systematic review, some general characteristics of the studies are mentioned (see Table 2). First, the majority of interventions were implemented in Europe (N = 18) compared to those implemented in China-Japan (N = 10), United States-Latin America (N = 8), and Asia (N = 8). Furthermore, the vast majority of interventions were implemented with primary schoolchildren and a minority of studies were implemented with secondary schoolchildren (N = 3). In addition, in the majority of studies, children’s parents were not involved (they had given their permission for their children to participate in the intervention) and in only eight studies parents were asked to complete questionnaires and participate in semi-structured interviews about the improvement of children’s communication skills. Moreover, the place of implementation of the interventions was in the vast majority of cases specially designed spaces within the school (classrooms, special computer laboratories, etc.) and only in some studies was the research carried out in specially designed spaces outside the school (N = 2). Finally, the sample size of participants involved in the interventions examined was small. More specifically, 25 studies had double-digit numbers of participants, 1 study had triple-digit numbers of participants, and 18 studies had single-digit numbers of participants.

Table 2.

Studies on the utilization of the social robot NΑO in the development of communication skills of children with autism.

Author/Date	Sample	Methodology	Types of intervention	Types of communication skills	Role of the NAO	Special characteristics of the NAO
Huskens et al.³³	6 children with autism spectrum disorder (ASD) between 8 and 14 years of age	Ten-minute individual sessions where the child sits on a table opposite the robot in a classroom. Data was collected from video cameras, from observation sheets completed by the observers	Use of questions facilitated by the robot. One-to-one type intervention (1:1)	Enhancing children’s verbal communication, especially self-initiated questions at a satisfactory level	Acts as an interlocutor	The tone of the voice
Warren et al.³⁴	6 children with ASD with an average age of 3.46 years	Four laboratory visits of the children with the presence of their parents for the period of 2 weeks where each child sat at the desk and opposite them was the robot. Data were collected from the Social Communication Questionnaire (SCQ) and Social Responsiveness Scale (SRS) questionnaires completed by the children’s parents and from data obtained from the eye-tracker	Use of interactive games with the reward of watching a video clip (e.g., Dora the Explorer) with the correct answer. One-to-one type intervention (1:1)	Enhancing verbal communication and understanding of oral instructions at a satisfactory level	Acts as an interlocutor	The tone of the voice
Anzalone et al.³⁵	16 children with ASD and 16 typical development (TD) children age 7–12 years	An intervention where each child sat at the desk and across from him was the robot which indicated to the child where he would look. The data was derived from systems that measured the angle of inclination of each child’s head and the RGB-D system	Interactive activities using 3D figures to depict animals and objects. One-to-one type intervention (1:1)	Enhancement of visual contact and verbal and non-verbal communication to a satisfactory level	Acts as an interlocutor and imitator	Simple movement of its head
Bekele et al.³⁶	6 children with ASD with a mean age range of 2.78–4.9 years and 6 TD children with a mean age range of 2.18–4.96 years	An intervention where each participant were introduced to the experiment room and seated in a chair opposite the robot. Data was collected from observation sheets, robot data, and questionnaires. Social Communication Questionnaire (SCQ) and Social Responsiveness Scale (SRS) questionnaires were completed by the children’s parents	Use of questions facilitated by the robot. One-to-one type intervention (1:1)	Enhancing children’s verbal communication, especially self-initiated questions at a satisfactory level	Acts as an interlocutor	Simple movement of its head
Barakova et al.³⁷	6 boys ages 8–12 years with ASD	The children participated in 30-min sessions as pairs which were recorded by a video camera and the children sat in chairs and opposite them was the robot which was placed on a table. The data was collected from the video and from data from the robot	Use of LEGO-game-based therapy. Intervention in groups	Enhancing verbal communication with the team member(s) and enhancing the production of oral language at a very good level	Interlocutor and as a mediator of interaction with others	Hand and head simple movements of the robot
Kaboski et al.³⁸	8 individuals with ASD and 8 TD peers ages 12–17 years	Participants were divided into pairs (one child with ASD and one child with TD) and trained in robot programming with the presence of a facilitator. Data were collected from observation protocols completed by the observers	Participation in cooperative activities (how to cooperate with each other and how to program a robot). Intervention in groups	Reinforcement of verbal communication and attention ability to a fairly satisfactory degree	Interlocutor and as mediator of interaction with others	Hand and head simple movements of the robot
Aziz et al.³⁹	2 children with ASD and 1 typically developing child aged 5 years old	An intervention where every child participated with the robot in activities. Data were collected from interviews conducted by each child	Interactive dance, singing and everyday activities (e.g., teaching the correct way to sneeze and wait their turn). One-to-one type intervention (1:1)	Enhancement of eye contact and verbal and non-verbal communication to a satisfactory level	Acts as an interlocutor	The tone of the robot’s voice
Srinivasan et al.⁴⁰	36 children with ASD between 5 and 12 years of age	Three groups of children who participated with the help of the robot in different activities. Data were collected from observation lists such as assessment battery for children, 2nd edition (MABC-2) and from the Vineland Adaptive Behavior Scales (VABS) questionnaire completed by parents	Imitation activities. Intervention in groups	Reinforcement of verbal and non-verbal communication and attention skill to a fairly satisfactory degree	Acts as an imitator	Simple movements of the hands, the head, and the tone of the voice
Chevalier et al.⁴¹	1 child 8-year-old with ASD	For a period of 7 months, we videotaped the child in 20-minute sessions with the student sitting in a chair and across the table from him was the robot. Data was collected from the video recording and from observation protocols collected by the researcher	Imitation activities. One-to-one type intervention (1:1)	Enhancement of visual contact and verbal and non-verbal communication to a satisfactory level	Acts as an imitator and the interlocutor	Simple movements of the hands, the head, and the change of body posture
So et al.⁴²	13 children aged 7–12 years with ASD	For a period of 1 month, we videotaped each child in 40-minute sessions with the students sitting in a chair and across the table from him was the robot. Data was collected from the video recording and from observation protocols collected by the researcher	Imitation games. One-to-one type intervention (1:1)	Strengthening non-verbal communication and gestures at a satisfactory level	Acts as an imitator and the interlocutor	Simple movements of the hands, the head, and the change of body posture
David et al.⁴³	5 children with ASD aged 3c5 years old	Daily 10-minute sessions where each child would sit in a chair and the robot would sit at a desk opposite him. Data were collected from observation protocols and from the Communication Subscale (CS), Social Interaction Subscale (SIS), Play Subscale (PS), and Stereotype Behaviors Subscale (SBS)	Use of question-and-answer sessions between child and robot to a satisfactory degree. One-to-one type intervention (1:1)	Enhancing verbal communication to a satisfactory level	Acts as an interlocutor	Simple movements of the head
Desideri et al.⁴⁴	2 children with ASD 9 years old	The sessions included the same training activities (5–15 min each). There were 12 sessions for each participant in the study. The intervention was implemented in one room and a table was placed in the center of the treatment room along with two chairs, one for the child and one for the educator. In addition, there was an observer in the room who completed an observation record and a camera that recorded the sessions. Data were collected from observation protocols, robot sensors, and video analysis	Use singing activities. One-to-one type intervention (1:1)	Enhancement of eye contact and verbal and non-verbal communication to a satisfactory level	Acts as an imitator and the interlocutor	Simple movements of its head and hands
So et al.⁴⁵	45 children with ASD aged 4–6 years old	The intervention lasted 9 weeks, with half-hour sessions and was implemented in a school room with the children sitting in chairs and the robot on the opposite table. Data was collected from video observation and an observation protocol completed by an observer in the classroom	Games with fairytale content with robot and child interaction. One-to-one type intervention (1:1)	Enhancing verbal communication and understanding of oral instructions at a satisfactory level	Acts as interlocutor	The tone of the voice
Taheri et al.⁴⁶	6 children with ASD aged 6–7 years old	A 3-month intervention program where the six participants were divided into three pairs and each pair attended twelve 30-minute sessions. Each pair of children sat upright and opposite them was the robot on the table. Data were collected from video data, interviews with the children’s parents, observation protocols, and questionnaires	Use of specially designed discovery games. Intervention in groups	Enhancing verbal communication with the team member(s), enhancing the production of oral language at a very good level	Mediator of interaction with others	The tone of the voice
Chung4⁴⁷	14 children with ASD aged 9–11 years old	Twelve half-hour sessions were implemented on a weekly basis. The children were in the classroom and the robot was on a table. Data were collected from the Social Responsiveness Scale (SRS), observation protocols completed by a classroom-based observer, and data from the NAO	Dance and singing activities based on a fairytale at a satisfactory degree. One-to-one type intervention (1:1)	Enhancement of eye contact and verbal communication to a satisfactory level	Acts as imitator and as interlocutor	Simple movements of the hands and head
Zhang et al.⁴⁸	20 children with ASD and 20 children TD with a mean age range of 5.08–8.83 years old	The intervention lasted 25 min and each child participated in different activities with the presence of the robot on the floor. Data were collected from observation protocols and from an interview with each child individually	Use interactive games (finding token found in cups) from the robot. One-to-one type intervention (1:1)	Enhancing verbal communication and understanding of oral instructions at a satisfactory level	Acts as interlocutor	The tone of the voice
Anzalone et al.⁴⁹	42 children with ASD with a mean age of 7.94 years and 16 TD children with a mean age of 8.06 years	The intervention was implemented in a specially designed room, where the robot is standing upright on the table and the child is facing the table standing upright as well. In the room there is a researcher, completes observation protocols. Data was collected from a video camera, from observation protocols, from robot sensors and by interviewing a subgroup of children with ASD 6 months after the intervention	Use of interactive game (card game for animal identification) facilitated by the robot. One-to-one type intervention (1:1)	Enhancing verbal communication and non-verbal communication and understanding of oral instructions at a satisfactory level	Acts as an imitator and the interlocutor	Simple movement of the head and hands
Qidwai et al.⁵⁰	15 children with ASD from 7–11 years old	Small groups of children were introduced to the room by a teacher and a monitor and made on a carpet unattended by the robot, which was placed on the floor. The data was collected from video observation, from interviews with teachers who were present during the intervention and with observation protocols	Use of song and dance activities. Intervention in groups	Strengthening attention ability, verbal and non-verbal communication, comprehension, and oral production to a very satisfactory level	Acts as mediator and interlocutor	Hand movements and facial expressions that the NAO uses to encourage communication
Marino et al.⁵¹	14 children with ASD aged 4–8 years old	Twelve 90-minute sessions were implemented in a specially equipped room, in which the robot and three cameras were located. The intervention was implemented in small groups of children sitting opposite the robot, with the presence of a teacher and an observer. Data were collected from the emotional understanding tests, the emotional vocabulary test, direct observation, and the completion of a specific protocol	Use of interactive activities by the robot. Intervention in groups	Enhancement of eye contact, attention ability, verbal communication, comprehension, and speech production to a very satisfactory level	Acts as interlocutor and as a mediator	The special movements of the hands and the movements of the body
David et al.⁵²	11 children with ASD aged between 3 and 5 years old	Eight sessions were implemented in a treatment room with the absence of the robot, a teacher, and an observer. The OZ wizard method was used, where the researcher can control the robot’s movements and dialogues without being visible to the child. The children were sitting in a chair and the robot was placed on a table opposite them. Data was collected from the camera, from the robot’s sensors, and from observation protocols	Use of interactive activities everyday life (such as waiting his tur and, moving objects to the right place). One-to-one type intervention (1:1)	Strengthened attention ability, verbal and non-verbal communication, comprehension, and oral production to a very satisfactory level	Acts as interlocutor and as imitator	Hand movements and facial expressions used by the NAO to encourage communication
Romero-García et al.⁵³	1 child with ASD aged 5 years old	The child was put in a specially designed room and sat upright and opposite him was the robot placed on a table. The Q-CHAT-NAO questionnaire system was used, a system of six questions answered by the child and not by parents or therapists as in the previous version	Use of questions and answers. One-to-one type intervention (1:1)	Enhancing oral and non-verbal communication, speech production, and comprehension to a very satisfactory level	Acts as interlocutor	The tone of voice robot
Chung⁵⁴	15 children with ASD aged 5–11 years old	Twelve half-hour sessions were implemented, where the first three sessions were implemented by a trainer, the first four to nine sessions were implemented with the use of robots and the last three sessions were implemented again with a trainer. Data was collected from video and observation protocols	Use of songs and interactive exercises facilitated by the robot. One-to-one type intervention (1:1)	Enhancement of visual contact, attention skills, verbal and non-verbal communication, and oral comprehension to a very satisfactory level	Acts as interlocutor and as a mediator	Facial expressions
Kim et al.⁵⁵	6 children with ASD with a mean age of 9.67 years	Thirty-minute sessions were implemented for a period of 3 months where each child participated in one session every two to three weeks in a specially designed room. The instructor was in the specially designed room and the robot was placed on the floor. The data was collected from video analysis and specially designed observation protocols	Use of interactive games such as rock-paper-scissors or they play catch by throwing and catching small beanbags. One-to-one type intervention (1:1)	Strengthened attention ability, verbal and non-verbal communication, comprehension, and oral production to a very satisfactory level	Acts interlocutor and as imitator	Gestures and body movements (such as when the child fails, the robot hits his head with his hand) and facial expressions
Taheri et al.⁵⁶	4 children with ASD with a mean age of 5.75 years	A 9-week intervention lasting 20–30 min each in a specially designed room with the presence of a robot, musical instruments, and an instructor. Data were collected from the Early Social Communication Scale (ESCS), imitation exercises, observation protocols, video analysis, and from questionnaires completed by the parents of the participants	Use of music activities with drum and xylophone. One-to-one type intervention (1:1)	Enhancement of eye contact, attention ability, verbal and non-verbal communication, and oral comprehension to a very satisfactory level	Acts as interlocutor	The specific movements of the hands and the postures of the body
Barnes et al.⁵⁷	12 TD children and 3 children with ASD with a mean age of 9.25 years	The intervention was implemented in three rooms, where in the first room were the parents of the children, in the second room were the children who would participate in the intervention, and in the last room were the children who did not participate. In the room was the robot placed on the table and opposite it was each child, in the presence of the trainer and an observer. Data were collected from semi-structured interviews with four special education professionals and from observation protocols completed by the observer	Use of musical, singing, dancing, and imitation games. One-to-one type intervention (1:1)	Enhancement non-verbal communication, verbal communication, and comprehension at a very satisfactory level	Act as interlocutor and as mediator	The specialized movements of the hands and the body movement of the NAO
Taheri et al.⁵⁸	20 children with ASD with a mean age of 4.98 years and 20 children with TD with a mean age of 5.30 years	The intervention took place in a specially designed room at the Social and Cognitive Robotics Lab., Sharif University of Technology, Iran. In the room were the children’s parents, the child, a handler, and a therapist using the “Wizard of Oz” method, with the robot placed on the table. The data will be collected by observation protocol and video analysis	Use of dance activities. One-to-one type intervention (1:1)	Enhancing eye contact and verbal and non-verbal communication	Act interlocutor	Special movements of the body and hands
Urdanivia Alarcon et al.⁵⁹	3 children with ASD aged 9–13 years	The intervention was implemented in a specially designed space with the presence of a robot, the therapist, and an observer. The robot is on the table and the child is sitting in a chair opposite it. Data was collected from video analysis, with an instrument to assess the child’s emotional expressions and by completing a primary observation	Use of interactive activities of everyday life (waiting my turn, listening to instructions, etc.). One-to-one type intervention (1:1)	Enhancement of attention ability, verbal and non-verbal communication, nd oral comprehension to a very satisfactory level	Acts as an interlocutor and as a imitator	The tone of voice, the movements of the hands, and the general posture of the body
de Souza Jeronimo et al.⁶⁰	20 children with ASD aged 9–11 years	The intervention was implemented in a specially adapted room with the presence of the robot, the teacher, a monitor, and musical instruments. The child was seated in a chair opposite the robot which was on the table. Data was collected by special observation protocols	Use of songs, musical instruments, and interactive activities. One-to-one type intervention (1:1)	Strengthened attention ability, verbal and non-verbal communication, comprehension, and oral production to a very satisfactory level	Acts as mediator	Hand movements and facial expressions
Nagae & Lee6⁶¹	11 children with ASD average age 6.9 years	A half-hour intervention with three parts that was implemented in a specially designed room, where the child and the trainer were located. The Wizard of Oz method was used with the child sitting in a chair and the robot placed on top of the desk. Data was collected from the robot’s sensors, video, and observation protocols	Using playing cards with emotions and interactive games. One-to-one type intervention (1:1)	Enhancing eye contact, speech production, comprehension, and recognition of emotions at a very satisfactory level	Act as mediator	The specialized movements of the hands and the body movement of the NAO
van Straten et al.⁶²	293 children with ASD aged 7–10 years	A 20-minute intervention for each child took place in a room separated from the main hall of the museum by glass doors, where the child was asked to sit on the floor in front of the robot, at a distance where he felt comfortable. In the room, were the instructor and an observer. The study was based on the approach of the “Wizard of Oz.” The data were collected by questionnaires, the robot’s sensors, and observation protocols	Use of questions and answers. One-to-one type intervention (1:1)	Enhancement of verbal and non-verbal communication and oral comprehension to a very satisfactory level	Acts as an interlocutor	Facial expressions
Papadopoulou et al.⁶³	19 children with ASD with a mean age of 8.58 years and 21 children with TD with a mean age of 8.86 years	Intervention in a total of 24 sessions in a specially designed room with the presence of the child, the robot, placed on the table, a teacher, and an observer. Data were collected from the robot’s sensors, video analysis, and the observation protocol	Use of interactive exercises such as like reading a fairytale that every child likes and memorizing pictures from it. One-to-one type intervention (1:1)	Enhancement of eye contact, attention ability, verbal and non-verbal communication, and oral comprehension to a very satisfactory level	Act as interlocutor and as mediator	Special movements of the hands and head
Korneder et al.⁶⁴	3 children with ASD aged 5 years	The intervention was implemented in a special clinic room with two tables and four chairs. In the room, there was an instructor, a researcher, the child sitting in a desk chair, and the robot, which was placed on top of the table	Use of questions and answers from the robot to the child. One-to-one type intervention (1:1)	Enhancement verbal and non-verbal communication and oral comprehension to a very satisfactory level	Acts as an interlocutor	Facial expressions
Telisheva et al.⁶⁵	34 children aged 3–12 years old, with different cases (ASD)	The experiment room was small and bright, with no furniture. In the intervention, each child was sitting on the ground in front of the robot who could see it clearly, the researcher, and a trainer. Data was collected from video analysis and observation protocols	Use of imitation and storytelling activities. One-to-one type intervention (1:1)	Strengthened attention ability, verbal and non-verbal communication, comprehension, and oral production to a very satisfactory level	Act as imitator and as interlocutor	Hand movements and specific facial expressions
Baraka et al.⁶⁶	11 children with ASD aged 2–7 years	In a specially designed room, face-to-face sessions were implemented in the presence of the child, the trainer, and an observer. The child was seated on a table in an office and the robot was placed on a table opposite him. The data was collected from sensors on the robot and from the first	Use of storytelling activities. One-to-one type intervention (1:1)	Strengthened attention ability, verbal and non-verbal communication, comprehension, and oral production to a very satisfactory level	Acts as interlocutor and as imitator	Specially adapted hand movements, body posture, and facial expression
Mutawa et al.⁶⁷	12 children with ASD aged 3–6 years	The intervention lasted for half an hour in a specially designed room and the children played in groups of three. In the room, there was the therapist, the observer, and the robot which was placed on the table and the child sat upright facing it.	Use of animal card games. Intervention in groups	Strengthening the ability to pay attention, to communicate verbally and non-verbally with group members, and to understand and produce spoken language at a very satisfactory level	Acts as mediator	Movements of the hands and head and facial expression
Al-Nafjan et al.⁶⁸	5 children with ASD aged 4–11 years	The intervention was implemented in a specially designed room with the presence of a therapist, an observer, and the robot. Each child sat either on a chair or on the floor and the robot was placed opposite him/her. Data was collected from video analysis and special observation protocols	Use of interactive social activities (such as introducing yourself and telling your age). One-to-one type intervention (1:1)	Enhancement of eye contact, attention ability, verbal and non-verbal communication, and oral comprehension to a very satisfactory level	Acts as mediator	Movements of the hands and head
Amirova et al.⁶⁹	16 children with ASD aged 5–12 years	Ten sessions of 15 min each were carried out in a specially designed room with the researcher, the therapist, and the child present, who was seated on the floor and the robot facing him. Data were collected from video analysis, observation protocols, and semi-structured interviews with children’s parents	Use of songs, stories, and generally interactive activities. One-to-one type intervention (1:1)	Strengthened attention ability, verbal and non-verbal communication, comprehension, and oral production to a very satisfactory level	Act as imitator and interlocutor	Specially designed hand movements and facial expression
Kostrubiec et al.⁷⁰	23 children with ASD with a median age of 5.52 years	In the specially designed room of the psychologist, the intervention was implemented with the absence of a therapist, an observer, and the robot, which was placed on the floor and opposite the child. Data were collected from assessment scales (Social communication questionnaire, Conner’s parent rating scale, Vineland adaptive behavior scales, and Wechsler intelligence scale for children IV) and observation protocols	Use of song activities from the robot. One-to-one type intervention (1:1)	Enhancement of visual contact, attention skills, verbal and non-verbal communication, and oral comprehension to a very satisfactory level	Acts as imitator	The movements of the hands and the movements of the body
Lorenzo & Lorenzo-Lledó⁷¹	4 children with ASD aged 11 years	The intervention was carried out in a specially designed room with the presence of a therapist, an observer, and the robot. Data were collected by camera and observation protocols	Use of ball games with the robot. One-to-one type intervention (1:1)	Enhancing verbal and non-verbal communication, speech production, and comprehension at a very satisfactory level	Act as imitator and interlocutor	The specialized movements of the hands and the body movement of the NAO
Chung et al.⁷²	60 children with ASD aged 5–11 years	The intervention was implemented in a total of six 30-min sessions in a specially designed room with the presence of a therapist and the robot. Data were collected from video analysis, semi-structured interviews, and assessment scale Social Responsiveness Scale	Use interactive activities (songs, dance, math quiz, etc.) facilitated by the robot. One-to-one type intervention (1:1)	Strengthened attention ability, verbal and non-verbal communication, comprehension, and oral production to a very satisfactory level	Acts as imitator	Special hand movements, body posture, and facial expressions
Lin et al.⁷³	2 children with ASD aged 6 years and 1 child with ASD aged 10 years	The intervention was implemented for 12 weeks with sessions of 40–60 min in a specially designed room with the presence of a heater and the robot placed above a table. Data were collected from the robot’s sensors and from assessment scales (BLO and ESCS)	Use of interactive activities with the language of the body. One-to-one type intervention (1:1)	Enhancement of visual contact, attention skills, verbal and non-verbal communication, and oral comprehension to a very satisfactory level	Act as mediator and interlocutor	The special movements of the hands and the posture of the body
Lorenzo-Lledó et al.⁷⁴	4 children with ASD with a median age of 11 years	An intervention of 3 hours implemented in a specially designed room with the presence of the therapist, two observers, with the robot on the floor and the child sitting opposite it. Data was collected using the observation protocols and the robot’s sensors	Use of imitation activities. One-to-one type intervention (1:1)	Enhancement of eye contact, attention ability, verbal and non-verbal communication, and oral comprehension to a very satisfactory level	Acts as interlocutor and as imitator	The specially adapted hand movement and the body movements
Zhao et al.⁷⁵	21 children with ASD aged 5–8 years and 21 children with TD aged 5–8 years	The intervention was implemented in a specially designed room and there was the therapist, a robot, and the child sitting opposite a screen with a robot or an adult present. Data were collected from a questionnaire completed by 15 children with ASD and 11 children with TD	Use of music activities by the robot. One-to-one type intervention (1:1)	Enhancement of eye contact, attention ability, verbal and non-verbal communication, and oral comprehension to a very satisfactory level	Act as interlocutor	The specially movements of the hands and head
Al-Nafjan et al.⁷⁶	7 children with ASD aged 5–9 years	The intervention was implemented in 30 sessions in a specially designed room with the presence of a therapist, an observer, and the robot sitting opposite the child. The data were collected by a primary observer, the robot’s sensors, and from camera	Use of interactive applause exercises. One-to-one type intervention (1:1)	Strengthened attention ability, verbal and non-verbal communication, comprehension, and oral production to a very satisfactory level	Act as interlocutor and mediator	Special hand and head movements and facial expressions

The types of interventions utilizing the NAO robot for the development of communication skills and their effectiveness

Across all intervention types, recent studies (2020–2025) report significantly greater improvements than earlier research (2013–2019). This is largely attributed to advancements in the capabilities of NAO robot’s and the increasing complexity of interactive tasks. One-to-one interventions using structured prompts and interactive games show notable gains in verbal and non-verbal communication, while group-based interventions foster social engagement and cooperative skills.

Since the use of Artificial Intelligence (AI) applications and, in particular, the utilization of social robots such as the social robot NAO in the field of education have been studied and recognized by many researchers,^14,77–82 the use of the social robot NAO in enhancing communication skills and social interaction of children with ASD has been studied for the last 13 years and more systematically for the last 6 years. In particular, different types of interventions carried out in recent years, especially in children with ASD, have evaluated verbal communication, non-verbal communication, and social interaction of children with ASD in a series of interventions implemented in specially designed classrooms in the school setting, yielding useful findings with important implications for the use of the NAO social robot in the field of special education.⁸³ It is worth noting that the majority of studies (see Table 2) focus on interventions for primary school children (aged 3–12 years).

The first two types of interventions evaluated were as follows:

(1) Interventions in which each child entered a specially designed room and the robot was placed opposite, with the presence of a teacher-therapist and the researcher. These interventions can be described as one-to-one interventions and symbolized as 1:1.

(2) Group interventions in which two or more children entered a specially designed room and the robot was placed opposite, with the presence of a teacher-therapist and the researcher.

Two other types of interventions one-to-one that were implemented and evaluated were:

(1) Interventions one-to-one (1:1) with oral questions-prompts.

(2) Interventions one-to-one (1:1) with interactive games and activities.

These types of interventions were evaluated in combination to assess the communication skills of children with ASD.

One-to-one (1:1) iterventions

Interventions one-to-one (1:1) with oral questions-prompts

Studies^33,36,43 for the period 2013–2019 that the interaction of the NAO with the child with ASD in the form of oral questions and prompts enhances verbal communication to a satisfactory degree in order to develop his social skills. On the other hand, studies^53,62,64 for the period 2020–2025 in the same type of intervention were more effective. More specifically, the results showed that child interaction with the recent version of the NAO using questions significantly enhances oral communication, oral production and comprehension.

One-to one (1:1) iterventions with interactive games and activities

Studies^{34,45,46,48,49} for the period 2013–2019 showed that the interaction between the child and the social robot NAO in simple interactive games, such as Dora the Explorer, fairytale games, discovery games, card games, and free games (such as rock-paper-scissors or play catch by throwing and catching small bean bags), enhanced verbal communication and understanding of oral instructions at a satisfactory level. In contrast, studies^55,57,61,71 with the same type of intervention in the period 2020–2025 seem to be more effective in relation to the results of studies carried out in the previous period. Interactive games were more complex than their counterparts in the previous period, such as music and dance games, imitation games, ball games, and card games with emotions. In particular, the results showed an enhancement in eye contact, oral language production, and comprehension to a very satisfactory level with the use of the recent version of the NAO robot.

Research^{35,39,41,44,47} implemented in the period 2013–2019 highlights the interaction of the social robot NAO with the child with ASD in a series of interactive activities, such as interactive activities using 3D figures to depict animals and objects, imitation activities, dance and singing activities, daily life activities (e.g., teaching the correct way to sneeze and wait their turn), and activities involving a structured story which enhanced visual contact and verbal and non-verbal communication to a satisfactory level. In addition, the results of studies^{52,54,56,59,60,63,65,66,68–70,72–76} in the period 2020–2025 on the same type of intervention showed enhancement of eye contact, attention ability, verbal and non-verbal communication, and oral comprehension to a very satisfactory level in relation to the corresponding studies implemented in the period 2013–2019. In particular, these studies used recent versions of the NAO and the structure of some activities was more complex, such as the rock-paper-scissors game, robot body language games, and storytelling activities.

Group-based interventions

Interventions implemented in groups using oral questions and prompts were not found in either period.

A study by Barakova et al.³⁷ on the interaction of the social robot NAO with children in a series of games, such as Lego-based play and discovery games, showed an enhancement in verbal communication with group members, comprehension, and production of spoken language to a very high extent. As for the period 2020–2025, the research of Mutawa et al.⁶⁷ showed that the interaction of the children’s group with the new version social robot NAO in card games with animals enhances oral communication with the group member(s), eye contact, enhances attention ability, comprehension, and oral production to a very satisfactory level in relation to the results of the previous period.

Studies carried out during the period 2013–2019 with the interaction of the NAO with a group of children in interactive skills such as cooperative activities (how to cooperate with each other and how to program a robot) and imitation activities with the aim of enhancing verbal communication and attention skills to a fairly satisfactory level.^38,40 On the other hand, the studies by Qidwai et al.⁵⁰ and Marino et al.⁵¹ in the same type of intervention showed that the interaction of the group of children with the new version of the NAO enhanced eye contact, attention ability, verbal communication, comprehension, and production of speech at a very satisfactory level compared to the previous time period.

Finally, interventions in which children are in groups and participate in interactive game-activities enhance communication skills to a more satisfactory extent than one-to-one interventions. In addition, regarding one-to-one interventions, it seems that interventions using interactive games and activities are more effective in developing the communication skills of children with ASD than interventions using oral language and prompts.

Characteristics of the NAO (tone of voice, hand, head, and body gesture movements and facial expressions) most frequently used in interventions and to what extent they influence the development of communication skills

The studies examined demonstrated certain characteristics of the social robot NAO, such as the tone of voice, movements of the hands and head, and more recently changes in facial expressions, due to the recent versions of NAO. These characteristics enhance the communication and social skills of children with ASD to a very satisfactory level and other characteristics to a satisfactory level depending on the time period we studied. However, there is an increase in communication skills in interventions which children with ASD are grouped.

Tone of voice as a communicative feature in interventions

The results of the studies by Huskens et al.³³ and Warren et al.³⁴ which focused on the tone of voice of the social robot NAO in interventions involving robot-initiated questions to children with ASD and their participation in interactive activities, respectively, reported improvements in verbal communication and oral comprehension to a satisfactory extent. In addition, three studies^39,45,48 also focused on the tone of voice of the social robot NAO in interventions that incorporated interactive activities such as singing and dancing, as well as interactive games. These studies reported improvements in verbal communication, comprehension of spoken language, and oral instructions to a satisfactory extent. Finally, the study by Taheri et al.⁴⁶ in which the participants were divided into groups during the intervention, focused on the tone of the NAO’s voice during specially designed games. The findings indicated a marked enhancement in oral communication among group members, contributing to a notably high level of spoken language production.

In the studies by Romero-García et al.⁵³ and Urdanivia Alarcon et al.⁵⁹ conducted in between 2020 and 2025 and in which the new version of the NAO was used, we examined tone voice and vocal expression in conjunction with specific hand and head movements, respectively. The results of these studies showed notable enhancements in both verbal and non-verbal communication, as well as oral language production and comprehension reaching a highly satisfactory level.

Hand, head, and body movements in interventions

Three studies^35,36,43 which focused on the simple head movements of the NAO during interactive activities, which involved 3D figures representing animals and objects, as well as questions posed by the robot to participants with ASD. These studies showed enhancements in eye contact and verbal and non-verbal communication to a satisfactory level. The results of four studies^35,37,44,47 which focused on the NAO social robot’s head and hand movements during LEGO-based play, imitation activities, singing activities, interactive games, and storybook-based dance and singing activities, respectively, showed improvements in eye contact and verbal and non-verbal communication at a satisfactory level. Furthermore, the studies conducted by Chevalier et al.⁴¹ and So et al.⁴² focused on the simple movements of the head, hands, and body posture of the NAO in imitation activities and games. The results of these studies showed enhancements of non-verbal communication and the use of gestures to a satisfactory extent. Τhe results of the studies by Kaboski et al.³⁸ and Srinivasan et al.⁴⁰ in which participants engaged in groups and engaged in participatory and imitation activities, respectively, showed enhancements in non-verbal communication, verbal communication, and attention skills reaching a moderately satisfactory extent.

The results of six studies^{56–58,61,70,73,74} which focused on the specific head, hand, and body posture movements of the recent version of the social robot NAO used in music games, musical activities, imitation activities, card game with emotions, singing activities, ball game, and interactive body language activities, respectively, demonstrated enhancements in eye contact, non-verbal and verbal communication, and oral comprehension. In addition, four studies^63,68,71,75 focused on the specific head and hand movements (excluding body posture) of the recent versions of the NAO robot conducted during interactive exercises, in social interactive exercises, in imitation activities, and in music and singing activities, respectively, demonstrated significant enhancements in significant non-verbal communication, attention skills, and eye contact at a very satisfactory level. Lastly, the study by Marino et al.⁵¹ in which participants engaged in groups based in interactive activities focused on the specific hand movements and body posture of the recent version of the NAO robot. The results of the research showed significant enhancements in eye contact, verbal communication, comprehension, and oral communication production to a very satisfactory extent.

Facial expressions in interventions

In the period 2013–2019, no research focused on the facial expressions of the social robot NAO as the earlier versions of the robot used in these interventions did not support such capabilities.

The research by Chung⁵⁴ highlighted that the facial expressions of the recent version of the NAO robot used in interactive exercises and activities significantly enhanced eye contact, attention skills, verbal and non-verbal communication, and oral comprehension to a very satisfactory level. Similar results were observed in the studies by Korneder et al.⁶⁴ and van Straten et al.⁶² in which children with ASD participated in question-and-answer activities with the NAO robot. The results demonstrated that the robot’s facial expressions enhanced verbal and non-verbal communication and oral comprehension to a very satisfactory extent.

Furthermore, eight studies^{52,55,60,65,66,69,72,76} utilized the recent version of the NAO in a variety of activities, including daily life activities, in complex interactive games, singing-themed activities, imitation activities, fairytale activities, activities with the theme of singing, interactive activities, and activities with applause, respectively. The results showed that the combination of the NAO’s facial expressions and specialized hand and head movements enhanced the attention skills of participating children with ASD, verbal and non-verbal communication, comprehension, and oral language production to a very satisfactory extent. Research conducted by Qidwai et al.⁵⁰ and Mutawa et al.⁶⁷ in which children with ASD participated in groups based in singing activities and card games, respectively, showed significant enhancements in attention skills, verbal and non-verbal communication, and speech production to a very satisfactory extent.

The integration of NAO’s characteristics voice tone, movement, and facial expressions has showed considerable potential in enhancing communication skills in children with ASD. While individual features contribute to progress, the combination of multiple characteristics appears to produce the most substantial enhancements, particularly in group-based interventions settings. Future research should focus on further enhancing NAO’s capabilities to optimize its contribution to social and communicative development.

The role of social robot NAO in interventions to enhance communication skills: Mediator, interlocutor, imitator, and their combinations

The research examined showed that NAO was used various roles in interventions including mediator, interlocutor, imitator, and their combination. These roles enhance verbal and non-verbal communication, eye contact, attention skills, comprehension, and oral communication. The degree of enhancement ranged from satisfactory to very satisfactory, depending on the duration and time period of each intervention.

NΑO as an interlocutor

The results of four studies^33,34,36,43 in which NAO used as an interlocutor in interventions, such as questions and answers between the robot and the child, demonstrated enhancements eye contact and verbal communication to a satisfactory extent. Furthermore, in three studies^39,45,48 NAO was used as an interlocutor in interventions such as interactive activities (dance, singing and daily activities), fairytale-based games and interactive games, respectively. The results showed a satisfactory level of enhancement of oral and non-verbal communication, eye contact, and comprehension of verbal instructions. On the other hand, in three studies^53,62,64 utilized the recent version of NAO as an interlocutor, in question-and-answer activities with the child. The results showed significant enhancement of verbal and non-verbal communication and oral comprehension at a very satisfactory level. In addition, the results of three studies^56,58,75 in which NAO was used as an interlocutor in song games, drumming and xylophone music activities, music and singing activities, and music activities, respectively, demonstrated improvement in eye contact, attention skills, verbal and non-verbal communication, and oral comprehension to a very satisfactory level.

NΑO as an imitator

In the study by Srinivasan et al.,⁴⁰ the participating children were divided into groups and the NAO robot was employed as an imitator in imitation activities. The results of the research showed significant enhancements of verbal and non-verbal communication and attention skills to a fairly satisfactory extent. Furthermore, research by Kostrubiec et al.⁷⁰ and Chung et al.⁷² highlight the role of NAO used as an imitator in singing activities and other interactive activities (music, dance, and math quiz), respectively. More specifically, the results showed enhancements in attention skills, verbal and non-verbal communication, comprehension, and production of spoken language at a very satisfactory level.

NΑO as an imitator and interlocutor

Τhe results of the studies by Chevalier et al.⁴¹ and So et al.⁴² in which the NAO social robot was used as an imitator and interlocutor at the same time in imitation activities and games, respectively, showed enhancement of non-verbal communication and gestures at significant satisfactory extent. In addition, three studies^35,47,49 in which the NAO social robot was used in 3D interactive activities, in singing, dancing, and story-based activities and interactive games, respectively. The results of these studies showed significant enhancement of verbal and non-verbal communication, eye contact, and understanding of verbal instruction at a satisfactory level.

In addition, in three studies,^65,66,74 the social robot NAO was utilized as both an interlocutor and imitator in imitation and storytelling activities, respectively. The results of the research highlight significant enhancements that eye contact, attention ability, verbal and non-verbal communication, and oral comprehension of children with ASD to a very satisfactory level. Furthermore, the results of five studies^{52,55,59,69,71} in which children with ASD used in interactive activities, interactive games, and interactive activities of daily living, as well as singing, storytelling, and ball games, respectively, showed significant enhancement of attention skills, verbal and non-verbal communication, comprehension, and oral language production at a very satisfactory extent.

NΑO as a mediator

Τhe research by Taheri et al.⁴⁶ involved children participating in groups, with the social robot NAO serving as a mediator in an intervention by specially designed games. The results revealed an improvement in oral communication within the group. On the other hand, in three studies^60,61,68 the new version of NAO was used in activities such as songs and musical instruments, in card games with emotions and in interactive social activities, respectively. The results of the research revealed significant enhancements of eye contact, attention skills, verbal and non-verbal communication, and oral comprehension at a very satisfactory level in relation to the corresponding studies of the previous period. In addition, the study by Mutawa et al.⁶⁷ where the participants were grouped and the NAO was utilized in activities involving animal cards showed strengthening of attention skills, verbal and non-verbal communication with group members, comprehension, and oral language production at a very satisfactory extent.

NΑO as a mediator and an interlocutor

Τhe results of the studies by Barakova et al.³⁷ and Kaboski et al.³⁸ in which children participated in groups with the NAO serving as both interlocutor and mediator and the interventions involving LEGO-based play and cooperative activities, respectively, revealed an enhancement in oral communication and attention skills to a fairly satisfactory extent. In contrast, the studies by Qidwai et al.⁵⁰ and Marino et al.⁵¹ in which children participated in group activities with NAO such as singing, dancing, and interactive activities demonstrated significant enhancement of eye contact, attention skills, verbal communication, comprehension, and oral production at a very satisfactory level. Furthermore, the results of five studies in which the NAO was used in activities such as songs and interactive activities, in games with music and dance games, in interactive activities (such as reading a fairytale), imitation games, and imitation activities, respectively, demonstrated significant enhancement of eye contact, verbal and non-verbal communication, and oral comprehension to a very satisfactory level.^{57,63,72,73,76}

Lastly, the studies reviewed indicate that NAO’s role in interventions as an interlocutor, imitator, mediator, or in a combination of these roles significantly enhances various communication skills in children, especially those with ASD. While all roles contribute to verbal and non-verbal communication, eye contact, attention skills, and comprehension, the effectiveness varies depending on the type of intervention and the nature of the activities involved. Future research could further investigate how different combinations of NAO’s roles may optimize learning and social interaction outcomes.

Discussion

Recent advancements in NAO’s design and capabilities, along with improved pedagogical practices, are expected to significantly enhance the effectiveness of interventions focused on communication. Initially, a positive impact is suggested to result from both technological developments in robotics and the adoption of improved pedagogical approaches. New versions of the NAO (post-2020) feature significantly enhanced capabilities—such as more natural hand and head movements and facial expressions—which contribute to improved communication outcomes.^{50–52,54–76} In addition, modern versions of the robot offer more natural interactions with children, enhancing robot-child interaction to a significantly degree compared to older versions.^{51,55–57,59,61,66,70–74} These improvements facilitate for more natural and engaging interactions, thus increasing the effectiveness of the interventions.

In addition, group interventions that incorporate interactive games have been found to be more effective than individual interventions, enhancing verbal and non-verbal communication, speech production, eye contact, and concentration while children showed higher levels of engagement when interacting simultaneously with the robot and their peers.^{37,38,40,50,51,67} The social dynamics of group settings may amplify learning by fostering shared experiences and encouraging cooperative engagement.⁸⁴ Regarding non-verbal communication, results indicate that children show more positively when the robot combines expressive movements with facial expressions.^{55,60,65,66,69,72,76} Furthermore, the results of the studies indicates that body language (hand movements, head movements, and posture) is more important than the robot’s tone of voice in enhancing communication.^{35,36,38,40–43,56–58,61,63,68,70,71,73–75} Thus, interventions that emphasize NAO’s expressive body language appear to result in superior communication outcomes.

Moreover, the role of the NAO in the intervention (as an interlocutor, imitator, mediator, or a combination of these) significantly influences the interaction and enhances children’s engagement in the activities. More specifically, when the robot acts both as an imitator and an interlocutor, children tend to participate more and develop skills that are retained after the intervention, compared to the other roles that utilized in the other interventions, suggesting that the interactive and mimetic skills of the NAO contribute more effectively to learning than passive mediating roles.^{41,42,47,49,52,55,59,65,66,69,71,74} Among the various techniques applied, the most effective seem to be: interactive storytelling, imitation games, cooperative activities, musical activities, emotion-based card games, and interactive social activities.^{34,35,37–42,44–52,54–57,59–61,63,65–76} These interventions are hypothesized to be more effective than other activities, and they are consistent with pedagogical strategies frequently used by teachers, such as storytelling, turn-taking activities, imitation-based learning, and training of gestures and facial expressions. More specifically, interactive storytelling activities using NAO have been utilized to tell social stories aimed at helping children with ASD understand social rules.^45,63,65,66

In relation to turn-taking activities, the robot guides children through activities where they have to wait their turn, thereby fostering social interaction and enhancing communication.^39,52,59,68 The strategy of imitation-based learning and gesture training refers to the use of specific movements or expressions of the NAO, where children are encouraged to imitate them, thereby enhancing their verbal and non-verbal communication skills.^{40–42,57,73,74} Thus, it is particularly effective in developing both verbal and non-verbal communication skills. Finally, gesture and facial expression training involves NAO’s body postures and facial expressions, aiming to enhance children’s non-verbal behaviors and improve their ability to interpret the emotions of others.^{50,52,55,60,65,66,67,69,72,76} These interventions suggest that developments in NAO design are significantly improving communication, particularly for children with ASD. Furthermore, promoting more natural interactions through the features of the newer versions of the NAO, combined with group and interactive activities, appears to increase their engagement. In addition, the role of the NAO as a mimic and conversationalist, as well as the emphasis on body language, are considered key factors in the development of communication skills of children with ASD, with the new versions of the robot enhancing them to a significant degree.

The present review confirms that the NAO social robot is an effective tool to support communication in children with ASD. Moreover, technological advances in the design of the NAO have gradually enhanced its effectiveness. However, further empirical studies are needed to validate these hypotheses and investigate. There were also limitations in our literature review. The first limitation of the present literature review is the limited number of databases searched and the lack of meta-analysis. In addition, the three dimensions on the basis of which the review was carried out are totally subjective, which may limit their representative compared to other potential categories. More specifically, there are some studies which could have been classified into more than one category. Therefore, the reliability of these dimensions could be called into question. In addition, papers published in conference proceedings were not considered, but only papers submitted to scientific journals. More specifically, papers published in journals are provided with full peer review and the research is presented with the required depth in methodology, results, and discussion, enhancing the credibility and validity of our review. However, conference papers can potentially provide important state of the art findings. Lastly, although methodological quality was not an exclusion criterion, some studies failed to report the gender of the participants, which suggests that the results should be interpreted with caution.

Future research investigating the role of the NAO social robot for children with ASD should prioritize the exploration of group dynamic in the development of communication skills, given the very limited number of studies (Ν = 6) conducted to date. In particular, studies involving group-based interventions seem to be more effective in developing the communication skills of children with ASD compared to one-to-one interventions (1:1), highlighting the need for further investigation into the potential benefits of group interventions. In addition, future research should focus on participants with ASD over the age of 12 years, as only a small number of the studies (Ν = 3) we examined included participants in this age group. More specifically, the majority of the studies reviewed involved preschool and elementary school children and there are few literature reports on secondary school children with ASD. Finally, the duration of the interventions in relatively short (up to 3 months) making it difficult to reliably assess whether the enhancement of communication skills are temporary or long-lasting. Therefore, future research should prioritize interventions of longer duration and include follow-up assessments of children’s communication skills beyond the 3-month period.

Conclusions

Artificial intelligence (AI), particularly in the field of social robotics, has experienced rapid growth in the last 5 years, and the integration of SAR in classrooms becoming increasingly common among educators. These developments have also influenced the field of special education, where the number of studies exploring the use of SARs to develop communication and social skills continues to rise. This systematic review aims to provide a comparative analysis of studies conducted two different time periods on the use of the social robot NAO to enhance the communication skills of children with ASD. The focus on specific elements of the selected studies, such as the types of intervention, the role of the NAO, and its unique characteristics, highlights the positive impact of the social robot NAO on the communication skills of children with ASD. The contribution of this systematic review is significant, as it is both a pioneering study that has not been done before and reports a number of results that report on the contribution of the social robot NAO to the development of communication in children with ASD. In conclusion, the NAO social robot can be effectively utilized to enhance communication skills in children with (or without) ASD, provided that the intervention is conducted in specially designed spaces within the school environment.

Footnotes

Ethical considerations

The authors declared that this study did not require an ethics committee approval. This study did not include any personal information. The authors further declared that the highest ethical practices were followed during this study.

Funding

The authors received no financial support for the research, authorship, and/or publication of this article.

Declaration of conflicting interests

The authors declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

References

Alam

Mohanty

. Integrated constructive robotics in education (ICRE) model: a paradigmatic framework for transformative learning in educational ecosystem. Cogent Educ 2024; 11(1): 123–136.

Belanche

Casaló

Flavián

, et al. Service robot implementation: a theoretical framework and research agenda. Serv Ind J 2019; 40(3): 203–225.

Onnasch

Roesler

. Taxonomy to structure and analyze human–robot interaction. Int J Soc Robot 2021; 13(3): 833–849.

Roshidul

Shariff

ARBM

Blackmore

, et al. High adoption of behavior based robotics in the autonomous machines. J Inf Syst Technol 2010; 3(6): 30–41.

Kurtz

Kohen-Vacs . Humanoid robot as a tutor in a team-based training activity. Interact Learn Environ 2022; 32(1): 340–354.

Chita-Tegmark

Scheutz

. Assistive robots for the social management of Health: a framework for robot design and human–robot interaction research. Int J Soc Robot 2021; 13(3): 197–217.

Chiu

Hwang

. Roles, applications, and research designs of robots in science education: a systematic review and bibliometric analysis of journal publications from 1996 to 2020. Interact Learn Environ 2022; 32(5): 1834–1859.

Jung

Won

. Systematic review of research trends in robotics education for young children. Sustainability 2018; 10(4): 84–115.

Kofinas

Orfanoudakis

Lagoudakis

. Complete analytical forward and inverse kinematics for the NAO Humanoid Robot. J Intell Robot Syst 2015; 77(2): 251–264.

10.

Vaidis

Otis

MJD

. Toward a robot swarm protecting a group of migrants. Intell Serv Robot 2020; 13(3): 299–314.

11.

Cruz-Ramírez

García-Martínez

Olais-Govea

. NAO robots as context to teach numerical methods. Int J Interact Des Manuf 2022; 16(3): 1337–1356.

12.

Tahir

Dauwels

Thalmann

, et al. A user study of a humanoid robot as a social mediator for two-person conversations. Int J Soc Robot 2020; 12(2): 1031–1044.

13.

Díaz-Boladeras

Claveri Díaz

García-Sánchez

. Robots for inclusive classrooms: a scoping review. Universal Access Inf Soc 2023; 23(2): 23–37.

14.

Ahmad

Mubin

Orlando

. Adaptive social robot for sustaining social engagement during long-term children–robot interaction. Int J Hum Comput Interact 2017; 33(12): 943–962.

15.

Syriopoulou-Delli

Gkiolnta

. Review of assistive technology in the training of children with autism spectrum disorders. Int J Dev Disabil 2020; 68(2): 73–85.

16.

Georgieva-Tsaneva

Andreeva

Tsvetkova

, et al. Exploring the potential of social robots for speech and language therapy: a review and analysis of interactive scenarios. Machines 2023; 11(7): 693–734.

17.

World Health Organization . Global report on health equity for persons with disabilities. [Internet]. World Health Organization, 2022. [cited 2025 January 28]. Available from: https://www.who.int/publications/i/item/9789240063600

18.

Anderson-Chavarria

. The autism predicament: models of autism and their impact on autistic identity. Disabil Soc 2021; 37(8): 1321–1341.

19.

Alcorn

Ainger

Charisi

, et al. Educators’ views on using humanoid robots with autistic learners in special education settings in England. Front Robot AI 2019; 12(2): 92–116.

20.

Thom

Ravichandran

, et al. Clozapine for treatment refractory catatonia in individuals with autism spectrum disorder: a retrospective chart review study. Expert Rev Clin Pharmacol 2023; 16(9): 865–875.

21.

Alasmari

Alduais

Qasem

. A thematic review of autism spectrum disorder research in Saudi Arabia: insights into diagnosis, assessment, and language considerations. Appl Neuropsychol Child 2023; 14(2): 236–250.

22.

Watroba

Luttinen

Lappalainen

, et al. Effectiveness of school-based occupational therapy interventions on school skills and abilities among children with attention deficit hyperactivity and autism spectrum disorders: systematic review. J Occup Ther Sch Early Interv 2023; 17(3): 671–703.

23.

Amirova

Rakhymbayeva

Yadollahi

, et al. 10 years of human-NAO interaction research: a scoping review. Front Robot AI 2021; 12(3): 123–156.

24.

Robaczewski

Bouchard

, et al. Socially assistive robots: the specific case of the NAO. Int J Soc Robot 2021; 13(2): 795–831.

25.

Papakostas

Sidiropoulos

Papadopoulou

, et al. Social robots in special education: a systematic review. Electronics 2021; 10(12): 13–98.

26.

Meniado

. Digital language teaching 5.0: technologies, trends and competencies. RELC J 2023; 54(2): 461–473.

27.

Mazzei

Chiarello

Fantoni

. Analyzing social robotics research with natural language processing techniques. Cogn Comput 2021; 13(1): 308–321.

28.

Bamber

Schneider

. College students’ perceptions mindfulness-based interventions: a narrative review of the qualitative research. Curr Psychol 2022; 41(3): 667–680.

29.

Siddaway

Wood

Hedges

. How to do a systematic review: a best practice guide for conducting and reporting narrative reviews, meta-analyses, and meta-syntheses. Annu Rev Psycho 2019; 70(3): 747–770.

30.

Moher

Liberati

Tetzlaff

, et al. Preferred reporting items for systematic reviews and meta-analyses: the PRISMA statement. J Clin Epidemiol 2009; 62: 1006–1012.

31.

Wang

Chen

Liu

, et al. Reporting items for systematic reviews and meta-analyses of acupuncture: the PRISMA for acupuncture checklist. BMC Complement Altern Med 2019; 19(2): 208–224.

32.

CASP. CASP-Critical Appraisal Skills Programme . CASP-Critical Appraisal Skills Programme website. [Internet], 2018. Available from: https://casp-uk.net/

33.

Huskens

Verschuur

Gillesen

, et al. Promoting question-asking in school-aged children with autism spectrum disorders: effectiveness of a robot intervention compared to a human-trainer intervention. Dev Neurorehabilit 2013; 16(5): 345–356.

34.

Warren

Zheng

Swanson

, et al.

Can robotic interaction improve joint attention skills?

J Autism Dev Disord 2013; 45: 3726–3734.

35.

Anzalone

Tilmont

Boucenna

, et al. How children with autism spectrum disorder behave and explore the 4-dimensional (spatial 3d+ time) environment during a joint attention induction task with a robot. Res Autism Spectr Disord 2014; 8(7): 814–826.

36.

Bekele

Crittendon

Swanson

, et al. Pilot clinical application of an adaptive robotic system for young children with autism. Autism 2014; 18(5): 598–608.

37.

Barakova

Bajracharya

Willemsen

, et al. Long-term LEGO therapy with humanoid robot for children with ASD. Expert Syst 2014; 32(6): 698–709.

38.

Kaboski

Diehl

Beriont

, et al. Brief report: a pilot summer robotics camp to reduce social anxiety and improve social/vocational skills in adolescents with ASD. J Autism Dev Disord 2015; 45(1): 3862–3869.

39.

Aziz

Moganan

Ismail

, et al. Autistic children’s Kansei responses towards humanoid-robot as teaching mediator. Procedia Comput Sci 2015; 76(2): 488–493.

40.

Srinivasan

Eigsti

Neelly

, et al. The effects of embodied rhythm and robotic interventions on the spontaneous and responsive social attention patterns of children with autism spectrum disorder (ASD): a pilot randomized controlled trial. Res Autism Spectr Disord 2016; 27(10): 54–72.

41.

Chevalier

Martin

Isableu

, et al. Impact of sensory preferences of individuals with autism on the recognition of emotions expressed by two robots, an avatar, and a human. Auton Robots 2017; 41(1): 613–635.

42.

Wong

Lam

, et al. Using a social robot to teach gestural recognition and production in children with autism spectrum disorders. Disabil Rehabil Assist Technol 2018a; 13(6): 527–539.

43.

David

Costescu

Matu

, et al. Developing joint attention for children with autism in robot-enhanced therapy. Int J Soc Robot 2018; 10(1): 595–605.

44.

Desideri

Negrini

Malavasi

, et al. Using a humanoid robot as a complement to interventions for children with autism spectrum disorder: a pilot study. Adv Neurodev Disord 2018; 2(3): 273–285.

45.

Wong

MKY

, et al. Robot-based intervention may reduce delay in the production of intransitive gestures in Chinese-speaking preschoolers with autism spectrum disorder. Mol Autism 2018b; 9(2): 34–49.

46.

Taheri

Meghdari

Alemi

, et al. Human–robot interaction in autism treatment: a case study on three pairs of autistic children as twins, siblings, and classmates. Int J Soc Robot 2018; 10(3): 93–113.

47.

Chung

. Robotic intervention program for enhancement of social engagement among children with autism spectrum disorder. J Dev Phys Disabil 2019; 31(3): 419–434.

48.

Zhang

Song

Tan

, et al.

Could social robots facilitate children with autism spectrum disorders in learning distrust and deception?

Comput Hum Behav 2019; 98(2): 140–149.

49.

Anzalone

Xavier

Boucenna

, et al. Quantifying patterns of joint attention during human–robot interactions: an application for autism spectrum disorder assessment. Pattern Recognit Lett 2019; 11(8): 42–50.

50.

Qidwai

Kashem

SBA

Conor

. Humanoid robot as a teacher’s assistant: helping children with autism to learn social and academic skills. J Intell Robot Syst 2020; 98(3): 759–770.

51.

Marino

Chilà

Sfrazzetto

, et al. Outcomes of a robot-assisted social-emotional understanding intervention for young children with autism spectrum disorders. J Autism Dev Disord 2020; 50(12): 1973–1987.

52.

David

Costescu

Matu

, et al. Effects of a robot-enhanced intervention for children with ASD on teaching turn-taking skills. J Educ Comput Res 2020; 58(1): 29–62.

53.

Romero-García

Martínez-Tomás

Pozo

, et al. Q-CHAT-NAO: a robotic approach to autism screening in toddlers. J Biomed Inform 2021; 11(8): 103–117.

54.

Chung

. Robot-mediated social skill intervention programme for children with autism spectrum disorder: an ABA time-series study. Int J Soc Robot 2021; 13(2): 1095–1107.

55.

Kim

Hirokawa

Matsuda

, et al. Smiles as a signal of prosocial behaviors toward the robot in the therapeutic setting for children with autism spectrum disorder. Front Robot AI 2021; 12(8): 599–755.

56.

Taheri

Shariati

Heidari

, et al. Impacts of using a social robot to teach music to children with low-functioning autism. J Behav Robot 2021a; 12(1): 256–275.

57.

Barnes

Park

Howard

, et al. Child-robot interaction in a musical dance game: an exploratory comparison study between typically developing children and children with autism. Int J Hum Comput Interact 2021; 37(3): 249–266.

58.

Taheri

Meghdari

Mahoor

. A close look at the imitation performance of children with autism and typically developing children using a robotic system. Int J Soc Robot 2021b; 13(4): 1125–1147.

59.

Urdanivia Alarcon

Cano

Paucar

FHR

, et al. Exploring the effect of robot-based video interventions for children with autism spectrum disorder as an alternative to remote education. Electronics 2021; 10(21): 25–77.

60.

de Souza Jeronimo

de Albuquerque Wheler

de Oliveira

JPG

, et al. Comparing social robot embodiment for child musical education. J Intell Robot Syst 2022; 105(3): 28–43.

61.

Nagae

Lee

. Understanding emotions in children with developmental disabilities during robot therapy using EDA. Sensors 2022; 22(14): 51–72.

62.

van Straten

Peter

Kühne

, et al. On sharing and caring: investigating the effects of a robot’s self-disclosure and question- Asking on children’s robot perceptions and child-robot relationship formation. Comput Hum Behav 2022; 129(6): 107–135.

63.

Papadopoulou

Karageorgiou

Kechayas

, et al. Efficacy of a robot-assisted intervention in improving learning performance of elementary school children with specific learning disorders. Children 2022; 9(8): 11–55.

64.

Korneder

Louie

WYG

Pawluk

, et al. Robot-mediated interventions for teaching children with ASD: a new intraverbal skill. Assist Technol 2022; 34(6): 707–716.

65.

Telisheva

Amirova

Rakhymbayeva

, et al. The quantitative case-by-case analyses of the socio-emotional outcomes of children with ASD in robot-assisted autism therapy. Multimodal Technol Interact 2022; 6(6): 46–68.

66.

Baraka

Couto

Melo

, et al. “Sequencing matters”: investigating suitable action sequences in robot-assisted autism therapy. Front Robot AI 2022; 9(1): 78–102.

67.

Mutawa

Al Mudhahkah

Al-Huwais

, et al. Augmenting mobile app with NAO robot for autism education. Machines 2023; 11(8): 833–852.

68.

Al-Nafjan

Alhakbani

Alabdulkareem

. Measuring engagement in robot-assisted therapy for Autistic children. Behav Sci 2023; 13(2): 618–634.

69.

Amirova

Rakhymbayeva

Zhanatkyzy

, et al. Effects of parental involvement in robot-assisted autism therapy. J Autism Dev Disord 2023; 53(1): 438–455.

70.

Kostrubiec

Lajunta

Paubel

, et al.

Does the social robot nao facilitate cooperation in high functioning children with ASD?

Int J Soc Robot 2024; 16(2): 281–298.

71.

Lorenzo

Lorenzo-Lledó

. The use of artificial intelligence for detecting the duration of autistic students’ emotions in social interaction with the NAO robot: a case study. Int J Inf Technol 2024; 16(5): 625–631.

72.

Chung

Kuen-fung Sin

Dhk

. Effectiveness of robotic intervention on improving social development and participation of children with autism spectrum disorder – a randomized controlled trial. J Autism Dev Disord 2024; 55(2): 449–456.

73.

Lin

Peng

Cheng

. Using intelligent robots to detect body language and improve social development in children with autism spectrum disorder. Int J Intell Robot Appl 2024; 12(2): 123–138.

74.

Lorenzo

Lorenzo-Lledó

Gilabert-Cerdá

. Application of robotics in autistic students: a pilot study on attention in communication and social interaction. Technol Knowl Learn 2024b; 29(3): 757–780.

75.

Zhao

Zhang

Hou

, et al. NAO robot promoting sensorimotor synchronization in high-functioning autism spectrum disorder children. Int J Dev Disabil 2024; 12(3): 1–15.

76.

Alnafjan

Alghamdi

Alhakbani

, et al. Improving imitation skills in children with autism spectrum disorder using the NAO robot and a human action recognition. Diagnostics 2025; 15(1): 60–74.

77.

Cifuentes

Pinto

Céspedes

, et al. Social robots in therapy and care. Curr Robot Repo 2020; 12(1): 59–74.

78.

Podpečan

. Can you dance? A study of child-robot interaction and emotional response using the NAO robot. Multimodal Technol Interact 2023; 7(1): 1–14.

79.

Rizos

Foykas

Georgakopoulos

. Enhancing mathematics education for students with special educational needs through generative AI: a case study in Greece. Contemp Educ Technol 2024; 16(4): ep535.

80.

Slavković

Ćeranić

Desideri

. Exploring the possibilities of using assistive technology for people with multiple sclerosis. Technol Disabil 2025; 37(2): 168–177.

81.

Ma-thi

Truong-duc

Ngo

T-D

, et al. An assistive IOT smart control system for people with severe mobility disabilities. Technol Disabil 2024; 36(3): 85–98.

82.

Valagkouti

Troussas

Krouska

, et al. Emotion recognition in Human-Robot interaction using NAO Robot. Computers 2022; 11(5): 1–13.

83.

Alnajjar

Cappuccio

Renawi

, et al. Personalized robot interventions for autistic children: an automated methodology for attention assessment. Int J Soc Robot 2021; 24(13): 67–82.

84.

Poort

Jansen

Hofman

. Does the group matter? Effects of trust, cultural diversity, and group formation on engagement in group work in higher education. High Educ Res Dev 2020; 41(2): 511–526.

The use of the social robot NAO to enhance communication skills in children with autism spectrum disorder: A systematic review of studies from 2013–2019 to 2020–2025

Abstract

Keywords

Introduction

Socially Assistive Robots (SARs)/NAO

Interventions using NAO

Study aims

Materials and methods

Search strategy

Eligibility criteria

Study selection, data extraction, and quality assessment

Results

General characteristics of included studies

The types of interventions utilizing the NAO robot for the development of communication skills and their effectiveness

One-to-one (1:1) iterventions

Interventions one-to-one (1:1) with oral questions-prompts

One-to one (1:1) iterventions with interactive games and activities

Group-based interventions

Characteristics of the NAO (tone of voice, hand, head, and body gesture movements and facial expressions) most frequently used in interventions and to what extent they influence the development of communication skills

Tone of voice as a communicative feature in interventions

Hand, head, and body movements in interventions

Facial expressions in interventions

The role of social robot NAO in interventions to enhance communication skills: Mediator, interlocutor, imitator, and their combinations

NΑO as an interlocutor

NΑO as an imitator

NΑO as an imitator and interlocutor

NΑO as a mediator

NΑO as a mediator and an interlocutor

Discussion

Conclusions

Footnotes

Ethical considerations

Funding

Declaration of conflicting interests

References