Attitudes toward social robots among parents of children with neurodevelopmental disorders: A questionnaire survey on a remote island in Japan

Abstract

Aim

This pilot study was performed to clarify parental acceptance of social robots to support children with neurodevelopmental disorders (NDDs) on a remote island in Japan.

Methods

Cross‐sectional analysis was performed to compare Negative Attitudes toward Robots Scale (NARS) scores of parents of children with NDDs cared for in welfare institutions and health care professionals, as well as the association between parental NARS scores, characteristics such as their own gender and age, and their children's gender and school grade. Between‐group comparisons were performed using the Kruskal–Wallis exact test.

Results

A total of 54 parents and 75 health care professionals participated in the study. Parents showed significantly lower NARS scores than health care professionals (median total scale, 33.5 vs. 37, respectively, p = 0.040, ε² = 0.033). Among parents with a single child with NDDs, those whose children were in the lower grades of elementary school showed significantly lower NARS scores compared with those whose children were in others (median total scale, 31 vs. 36, respectively, p = 0.015, ε² = 0.14).

Conclusion

As a pilot survey for the implementation of social robots, parents of children with NDDs had lower negative attitudes toward social robots than professionals. Parental acceptance of social robots may differ depending on the children's school grade. These findings may provide preliminary insights that could inform health care providers considering the introduction of advanced technology to address the limited availability and accessibility of medical resources in rural areas, including remote islands.

INTRODUCTION

The number of children with neurodevelopmental disorders (NDDs) is increasing worldwide. ¹ These children rely on services provided by various professionals including those in child welfare, disability services, mental health, and education. Interprofessional collaboration among these services is essential for providing comprehensive support. ² Despite this recognized need for coordinated care, regional barriers exist. Providing appropriate medical care for this population remains a challenge, particularly in remote island settings, due to the limited availability and accessibility of medical resources. While support groups for individuals with NDDs are relatively abundant in urban areas, their presence is limited in rural regions, reflecting a significant regional disparity. ³ This is especially critical in pediatric specialty care, which often requires access to specialized professionals and resources.

Telemedicine enables communication between patients and doctors regardless of distance, and telemedicine with social robots can potentially address gaps in child psychiatric services on remote islands. Telemedicine has been shown to be clinically useful in the diagnosis and treatment of patients with NDDs. ⁴ However, it is difficult to build trusting relationships with patients due to limited nonverbal communication, such as eye contact, as well as physical distance. ⁵ Social robots can enable interaction with patients residing in remote or underserved areas, ⁶ , ⁷ and their use is expected to help attenuate the negative effects of telemedicine. The use of social robots in therapy for individuals with NDDs, especially autism spectrum disorder (ASD), is a unique approach that has recently attracted considerable public attention. ⁸ A systematic review regarding the use of social robots in the care of individuals with ASD reported that social robots can improve the quality of services provided by clinicians. ⁹ Telemedicine with social robots has the potential to provide effective treatment for children with NDDs while addressing the limitations of child psychiatric services in remote communities.

Despite the recognized need for technology‐based support, the use of information and communication technology (ICT) in health care remains limited in rural areas of Japan. ¹⁰ This limitation raises even greater concern about acceptance in remote island settings, given that previous studies have reported that some people consider social robot unacceptable replacements for therapists for children with NDDs. ¹¹ , ¹² To date, no evidence has been reported regarding the acceptability of social robots for supporting children with NDDs on remote islands. Under the hypothesis that parents on remote islands might exhibit negative attitudes toward social robots, we conducted a preliminary investigation—before launching a support project—of their acceptance among parents, who are the most familiar with the children.

METHODS

Design, setting, participants, and data collection

This was a pilot survey for the implementation of telemedicine using social robots for children with NDDs. The Negative Attitudes toward Robots Scale (NARS) was measured in 2023 on a remote island in Goto City, Japan. Participants included parents of children with NDDs receiving care in welfare institutions (e.g., child development support and after‐school daycare service centers), as well as professionals such as hospital nurses, staff members at welfare institutions, staff members from the child support division in the Goto City Welfare and Health Department, and staff members from the Goto City Board of Education. In Japan, child development support centers and after‐school daycare service centers are community‐based outpatient facilities that provide daily life and educational support to children with NDDs who live at home. ¹³ In addition, cross‐sectional analysis of the association between parental NARS, their characteristics such as their own gender and age, and their children's gender and school grade, was conducted.

The survey was conducted in an open setting. After the purpose of the study was explained, participants were asked to complete paper questionnaires voluntarily and anonymously. As participation was not tracked and the total number of individuals who viewed the questionnaire was unknown, the response rate could not be determined. Our study was approved by the Institutional Review Board of Nagasaki University Institutional Ethics Committee (approval number 23063004) and was conducted in accordance with the Declaration of Helsinki.

Instrument

The Japanese version of NARS was used to assess the acceptability of social robots. ¹⁴ NARS is a widely accepted and cited measure of attitudes toward robots, even when applied in the field of medical and nursing care. ¹⁵ , ¹⁶ , ¹⁷ The 14 NARS items are classified into three subscales: Subscale 1, “negative attitude toward interaction with robots” (six items); Subscale 2, “negative attitude toward the social influence of robots” (five items); and Subscale 3, “negative attitude toward emotional interaction with robots” (three items) (Table S1). Each item was scored on a five‐point scale (from 1, strongly disagree, to 5, strongly agree), and individual scores on each subscale were calculated by summing the scores of all items included in the subscale, with some items reverse‐coded. Therefore, the minimum and maximum scores were 6 and 30 for Subscale 1, 5 and 25 for Subscale 2, and 3 and 15 for Subscale 3, respectively. Higher scores reflect more negative attitudes toward robots, whereas lower scores reflect less negative attitudes. The survey responses were confidential, and identifiable personal information was not collected.

Statistical analysis

First, attitudes toward social robots were compared between parents and professionals after exclusion of participants with missing values on the NARS. Second, we focused on parents and explored factors influencing their attitudes after excluding those with two or more children with NDDs because it was unclear which child influenced the parental attitudes, as well as those with missing values for their child's characteristics. Due to the small sample size and non‐normal data distribution, the Kruskal–Wallis exact test was used for group comparisons, and effect sizes (ε²) were calculated to assess the magnitude of the observed effects. The internal consistency of the NARS was assessed using Cronbach's α for the total scale and each subscale.

The Kruskal–Wallis test was performed using R version 4.5.1 for Windows (R Foundation, Vienna, Austria), and Cronbach's α was calculated using SPSS Statistics version 29 (IBM, Tokyo, Japan). In all analyses, p < 0.05 was taken to indicate statistical significance.

RESULTS

Parental attitudes toward robots compared with those of professionals

The study consisted of 129 participants after the exclusion of two participants with missing values on the NARS. Cronbach's α was 0.87 for the total scale, 0.85 for Subscale 1, 0.77 for Subscale 2, and 0.77 for Subscale 3. Table 1 shows the characteristics of the participants. The study population consisted of 54 parents of children enrolled in welfare institutions, along with 75 professionals, including 22 nurses working in hospitals, 11 staff members from welfare institutions, 19 staff members from a division related to support for children in the Goto City Welfare and Health Department, and 23 staff members from the Goto City Board of Education.

Table 1.

Participant characteristics (n = 129).

	Parents, n = 54	Professionals, n = 75
Gender
Men, n (%)	10 (18.5)	18 (24.0)
Women, n (%)	44 (81.5)	57 (76.0)
Age group
15–39 years, n (%)	37 (68.5)	31 (41.3)
40–69 years, n (%)	17 (31.5)	44 (58.7)

Table 2 shows a comparison of participants' NARS scores. Total, Subscale 1, and Subscale 2 NARS scores of parents were significantly lower than those of professionals (median total scale, 33.5 vs. 37, respectively, p = 0.040, ε² = 0.033; Subscale 1, 12 vs. 13, respectively, p = 0.024, ε² = 0.040; and Subscale 2, 13 vs. 15, respectively, p = 0.039, ε² = 0.033) (Table 2). There were no associations between NARS score and either age or gender.

Table 2.

Comparison of participants' Negative Attitudes toward Robots Scale (NARS) total scale and subscale scores.

	Relationship with children				Gender				Age group
	Parents	Professionals			Men	Women			15–39 yrs	40–69 yrs
	n = 54	n = 75	p value	ε²	n = 28	n = 101	p value	ε²	n = 68	n = 61	p value	ε²
Subscale 1	12 (7–14)	13 (10–16.5)	0.024	0.040	11 (7–13)	12 (10–16)	0.053	0.029	12 (7.5–15)	13 (10–17)	0.054	0.029
Subscale 2	13 (11–16)	15 (12–17)	0.039	0.033	15 (11–17)	14 (12–17)	0.45	0.005	13 (11–16)	14 (12–17)	0.16	0.016
Subscale 3	9 (7–10)	9 (7.5–10)	1.0	<0.001	9 (8–10)	8 (7–10)	0.23	0.011	8.5 (7–10.5)	9 (7–10)	0.84	<0.001
Total scale	33.5 (27–40)	37 (32.5–41.5)	0.040	0.033	34 (29–38.5)	36 (30–41)	0.47	0.004	34 (27.5–40.5)	37 (32–41)	0.14	0.017

Note: Higher scores indicate a more negative attitude. Figures represent the median (Quartiles 1–3) or exact p‐value for the Kruskal–Wallis test.

Abbreviation: yrs, years.

Factors influencing the attitudes of parents with a single child with NDDs toward social robots

To better understand the factors associated with parental attitudes, we focused on a subpopulation of parents with a single child with NDDs (Table 3). The second analysis included 41 parents and excluded 13 parents who had two or more children with NDDs or had missing values regarding their child's characteristics. None were under 20 or over 59 years old.

Table 3.

Characteristics of parents who have a single child with neurodevelopmental disorders (n = 41).

	n (%)
Parent's gender
Men	6 (14.6)
Women	35 (85.4)
Parent's age group
20–39 years	30 (73.2)
40–59 years	11 (27.8)
Gender of children with NDDs
Boys	32 (78.0)
Girls	9 (22.0)
School grade of child with NDDs
Pre‐elementary school	19 (46.3)
Lower grade of elementary school	15 (36.6)
Upper grade of elementary school and over	7 (17.1)

Abbreviation: NDDs, neurodevelopmental disorders.

Table 4 shows a comparison of scores of NARS among parents of children with NDDs based on the characteristics of the parents themselves and their children. Significant differences were found on the total scale and Subscale 3 depending on the child's school grade (Table 4). NARS scores of parents of children in lower grades of elementary school were significantly lower than those in other categories (median total scale, 31 vs. 36, respectively, p = 0.015, ε² = 0.14; median Subscale 2, 12 vs. 14, respectively, p = 0.041, ε² = 0.10; median Subscale 3, 7 vs. 10, respectively, p = 0.001, ε² = 0.24). There were no associations between NARS scores and parental gender, parental age, or the gender of the children with NDDs.

Table 4.

Comparison of Negative Attitudes toward Robots Scale (NARS) total and subscale scores in parents who have a single child with neurodevelopmental disorders.

	Parent's gender				Parent's age group				Gender of children with NDDs				School grade of children with NDDs
	Men	Women			20–39 yrs	40–59 yrs			Boys	Girls			Pre‐ES	LG	UG
	n = 6	n = 35	p value	ε²	n = 30	n = 11	p value	ε²	n = 32	n = 9	p value	ε²	n = 19	n = 15	n = 7	p value	ε²
Subscale 1	9.5 (6–12)	12 (8–14)	0.51	0.012	11.5 (8–14)	12 (6–15.5)	0.92	<0.001	12 (7.5–14.5)	11 (7–12)	0.38	0.020	12 (8.5–14.5)	10 (6.5–12)	12 (11–13)	0.35	0.054
Subscale 2	14.5 (9–16)	13 (10–16)	0.59	0.008	13 (11–16)	14 (8.5–20)	0.46	0.014	14 (10–17.5)	12 (9–13)	0.11	0.065	14 (11–16.5)	12 (7–14)	14 (13–20)	0.093	0.12
Subscale 3	9.5 (9–11)	9 (7–10)	0.31	0.027	9 (7–10)	9 (6–10)	0.74	0.003	9 (7–10)	8 (7–10)	0.74	0.003	9 (8–10)	7 (5.5–9)	11 (10–11.5)	0.002	0.29
Total scale	34.5 (28–37)	34 (27–41)	0.86	<0.001	33 (28–37)	36 (20.5–44.5)	0.57	0.009	35 (28.5–41)	31 (27–35)	0.15	0.054	35 (30–41)	31 (20–34.5)	37 (35–44.5)	0.035	0.16

Note: Higher scores indicate a more negative attitude. Figures represent the median (Quartiles 1–3) or exact p‐value for the Kruskal–Wallis test.

Abbreviations: ES, elementary school; LG, lower grades of elementary school; UG, upper grades of elementary school and over; yrs, years.

DISCUSSION

Our study, conducted as a pilot survey for the implementation of social robots, found that parents of children with NDDs had lower negative attitudes toward social robots than professionals. Some previous studies could explain why the parents were receptive to the robots in our study. Parents of children with NDDs were found to perceive robots as more human‐like than parents of typically developing children, and to have higher expectations for the use of telemedicine than health care professionals. ¹⁸ , ¹⁹ These characteristics may have influenced parental attitudes toward the use of robots in the present study. A previous European study showed that parents of children with NDDs accepted the use of robots with lower levels of acceptability. ¹² This had to be considered carefully by developers and implementers of robotics. On the other hand, the cultural background and the time period of the previous European study published in 2015 differ from those of the present study. Cultural background and nationality have been identified as factors influencing attitudes toward and acceptance of social robots, ²⁰ , ²¹ and such attitudes may also change over time. ²² Our results indicated high acceptability of robots among parents of children with NDDs in contemporary Japan. These findings may help to reduce concerns of developers and implementers of robotics. On the other hand, professionals showed more negative attitudes toward robots than parents. They are familiar with the challenges of current robotics technology and often have concerns regarding the practical applicability of robots in daily support and the use of telemedicine. ¹⁹ , ²³ In addition, some professionals may perceive robots as potentially replacing aspects of their professional roles, which could contribute to their skepticism. ²⁴ These factors may explain why they did not exhibit a very positive attitude toward robots.

Our study also found that parents of children with NDDs in lower grades of elementary school had lower negative attitudes toward social robots. To our knowledge, this is the first study to show an association between school grades of children with NDDs and their parents' attitudes toward social robots. Some previous studies have shown that children in lower grades of elementary school spent more time interacting with social robots and had greater adherence to social robots than children in other grades. ²⁵ , ²⁶ Most studies of educational robots for children with NDDs have focused on lower elementary school grades. ²⁷ Our results may reflect the expectations of parents with children at an age when positive effects of robotic interventions are especially anticipated. Our results showed no significant differences between parents of boys and girls regarding attitudes toward social robots. As there were a limited number of parents with girls in our study, further research is needed to verify differences between parents of boys and girls.

Our study was unique in that it focused on the acceptability of parents living on a remote island in Japan, where the prevalence of information and ICT use for health care are low, ¹⁰ and also investigated the factors influencing parental attitudes toward social robots. Most previous social studies investigating attitudes and expectations toward using robots in the education and care of children with NDDs have targeted therapists, psychologists, educators, and students in relevant fields. ²³ , ²⁸ , ²⁹ , ³⁰ This study provided evidence regarding the factors influencing the acceptance of robots among parents, who are the primary caregivers and closest figures to children.

This study had some limitations. First, the inclusion of professionals as a comparison group may have introduced selection bias, which could affect the generalizability of the findings. Second, the analyses were conducted within a small sample and restricted to univariable models. Potential confounding factors, such as familiarity with ICT and socioeconomic status (e.g., occupation), could not be adjusted for. Third, multiple statistical tests were performed across the NARS subscales and participant characteristics, as this was an exploratory study. Given these limitations, future confirmatory hypothesis‐testing studies with larger sample sizes and multivariable analyses are required to validate the observed associations. Fourth, this study was conducted as part of a pilot survey for a broader project aiming to introduce social robots to remote islands to support children with NDDs. Because we assessed attitudes in a real‐world context, participants' awareness of our intention to implement social robots may have biased their responses to the survey. To evaluate attitudes more neutrally and confirm validity, studies are needed in settings where robot use is not explicitly assumed. Finally, this was a study in a single region. Therefore, further studies are necessary to replicate our results in other geographical or cultural contexts.

In conclusion, parents of children with NDDs had lower negative attitudes toward social robots than professionals. In particular, parents of children in lower elementary school grades tended to show relatively low negative attitudes. These findings may provide preliminary insights that could inform health care providers considering the introduction of advanced technology to address the limited availability and accessibility of medical resources in rural areas, including remote islands.

AUTHOR CONTRIBUTIONS

All authors contributed to the study design. Kengo Maeda, Kazuhiko Arima, Jun Miyata, and Yasuhiro Nagata interpreted the data and performed statistical analyses. Kengo Maeda and Yasuhiro Nagata wrote the manuscript. All authors contributed to the draft revisions and approved the final manuscript.

CONFLICT OF INTEREST STATEMENT

J.M., F.N., and T.M. are affiliated with an endowed department (Department of Island and Community Medicine) at Nagasaki University Graduate School of Biomedical Science, which is supported by Nagasaki Prefecture and Goto City.

ETHICS APPROVAL STATEMENT

Our study plan was approved by the Institutional Review Board of Nagasaki University Institutional Ethics Committee (approval number 23063004). This study conformed to the tenets of the Declaration of Helsinki.

PATIENT CONSENT STATEMENT

All participants were informed of the purpose of the study and that their participation was anonymous and voluntary. All participants provided written informed consent to participate in the study.

CLINICAL TRIAL REGISTRATION

N/A.

Supporting information

Supporting Information.

Maeda K, Kawashiri S‐Y, Arima K, Miyata J, Niri T, Honda Y, et al. Attitudes toward social robots among parents of children with neurodevelopmental disorders: a questionnaire survey on a remote island in Japan. Psychiatry Clin Neurosci Rep. 2025;4:e70223. 10.1002/pcn5.70223

DATA AVAILABILITY STATEMENT

The data that support the findings of this study are available from the corresponding author upon reasonable request.