Skip to main content
NCBI home page
Wiley Open Access Collection logoLink to Wiley Open Access Collection
. 2026 Aug 10;28(3):e70410. doi: 10.1111/nhs.70410

Exploring Emotional Attachment to Socially Assistive Robots: A Qualitative Study Among Older Adults

Jixia Cao 1, Can Yang 1, Xingru Yang 1, Junjin An 1, Jinchang Yu 1, Jiahui Chen 1, Ying He 1,✉
PMCID: PMC13456949  PMID: 42575723

ABSTRACT

Socially assistive robots (SARs) show potential to alleviate loneliness among older adults but raise ethical risks regarding emotional attachment. While existing studies largely focus on normative frameworks, empirical insights into older adults' subjective experiences in specific cultural settings remain limited. Grounded in China's care realities, this qualitative study explored older adults' anticipated emotional attachment to SARs and the associated ethical concerns through semi‐structured, scenario‐based interviews. Data were analyzed using Braun and Clarke's six‐phase thematic analysis. Three themes emerged: (1) anticipated and imagined emotional attachment to SARs; (2) perceived catalysts of anticipated emotional attachment, driven by loneliness, limited family support, and the projection of human‐like qualities; and (3) anticipated consequences of emotional attachment, including expected emotional support alongside concerns about family ties and autonomy. Chinese older adults expressed ambivalent expectations regarding emotional attachment to SARs. Their accounts reflected unmet emotional and caregiving needs within family‐based care norms while raising concerns about intergenerational relationships and autonomy. Future SAR implementation should ensure that robots complement rather than replace human care and preserve relational autonomy.

Keywords: emotional attachment, ethics, older adults, qualitative research, socially assistive robots

Key Points

  • Older adults' anticipated attachment to socially assistive robots reflected unmet emotional and caregiving needs rather than technology use alone.

  • Older adults interpreted anticipated attachment to robots through family‐based expectations of care and concerns about intergenerational relationships.

  • Nurses should attend to the unmet emotional and caregiving needs underlying anticipated attachment and ensure that robots complement rather than replace human care.

1. Introduction

The global population aged 65 years and above is projected to double by 2050 (United Nations 2020). This demographic trend is associated with increased functional decline and strained care resources (Lill 2020; Matthews et al. 2023). Loneliness among older adults presents a significant risk for physical and psychological distress (Chen, Moyle, et al. 2020; Chen et al. 2022). Traditional care models frequently struggle to provide consistent companionship, making it increasingly difficult to meet older adults' emotional needs.

Socially assistive robots (SARs), defined as robots that deliver emotional support and social interaction (Rabbitt et al. 2015), have shown potential for functional applications such as cognitive impairment care and daily living assistance (Luperto et al. 2021; Nakamura et al. 2021; Persson et al. 2022). Beyond functional assistance, SARs are increasingly discussed as tools to support older adults' emotional well‐being and alleviate loneliness (Søraa et al. 2022; Follmann et al. 2021; Pirhonen et al. 2020; Lee and Joo 2025). As attention on SARs grows, ethical concerns in human‐robot interaction (HRI) have also been raised, such as autonomy, privacy and security, and trust risks (Tamir 2025; Velmurugan et al. 2025).

In addition to these concerns, HRI may also give rise to issues of emotional attachment. This concept of attachment was originally proposed by Bowlby to describe the emotional bond between an infant and its primary caregiver (Bowlby 1969). HRI researchers have extended these frameworks to examine how humans form emotional bonds with non‐human agents (Szondy and Fazekas 2024). In care for older adults, previous studies have examined how older adults perceive the emotional roles of robots, finding that robots are often viewed as family members (Randall et al. 2019). Other scholars have identified different forms of human‐machine attachment and discussed the potential ethical risks (Law et al. 2022; Rabb et al. 2021). For example, critics have pointed out that emotional computing in robots lacks “emotional authenticity,” which can easily lead older adults to become entangled in one‐sided pseudo‐emotional relationships (Liu 2025). However, most existing studies have primarily focused on normative frameworks, that is, the ethical principles that should guide human‐robot relationships. Older adults' subjective experiences of emotional attachment to SARs, particularly within specific cultural settings, remain underexplored.

Understanding emotional attachment to SARs requires attention to the realities and norms of social care. Findings from the CARESSES project suggest that culturally appropriate robotic care may enhance older adults' engagement and emotional well‐being (Papadopoulos et al. 2022, 2020). Research by I. Papadopoulos et al. also indicates that cultural factors are significantly associated with the acceptance of SARs among healthcare and social care professionals in multiple non‐Western and Western countries (Papadopoulos et al. 2023). Studies conducted in several non‐Western contexts have further suggested that cultural norms and caregiving expectations may influence perceptions and acceptance of SARs. At the theoretical level, technology acceptance frameworks suggest that cultural values may shape expectations toward care technologies and influence how human‐robot relationships are understood in different social contexts (Dharmansyah et al. 2026). These perspectives indicate that cultural context may be an important dimension in understanding emotional responses to SARs.

In China, the older adult care service sector faces challenges including an imperfect pension system, regional economic disparities, and a continued reliance on home care (Shi et al. 2025). Many older adults live in empty‐nest households, especially in rural areas, which may pose risks to their physical and mental health (Li et al. 2022; Su et al. 2012; Wu 2021). Although policies actively promote smart aging industries (Chen, Lou, et al. 2020; Hou et al. 2022), the practical deployment of SARs remains constrained by high costs and technological immaturity (Luo et al. 2024). The social compensation hypothesis suggests that individuals experiencing limited social connection may attribute social meaning to non‐human agents in response to unmet belongingness needs (Epley et al. 2008). As SARs have not yet been widely adopted in care settings for older adults in China, most older adults lack direct experience with such technologies. It therefore remains unclear how unmet emotional and caregiving needs may shape older adults' expectations of SARs within the Chinese care context. At the same time, empirical research grounded in China's local care context has paid limited attention to older adults' anticipated emotional attachment to SARs.

Against this background, this study aimed to explore older adults' anticipated emotional attachment to socially assistive robots within the Chinese sociocultural context. Specifically, using scenario‐based qualitative interviews, we sought to examine how older adults understand and imagine such attachment, identify the perceived factors that may give rise to it, and explore its anticipated consequences. We also examined the ethical concerns older adults associated with human‐robot emotional attachment, particularly in relation to family relationships and individual autonomy. By doing so, this study provides empirical evidence to inform the culturally appropriate and ethically responsible development of care technologies for older adults.

2. Methods

This study was reported in accordance with the Consolidated Criteria for Reporting Qualitative Research (COREQ) checklist. We employed a qualitative descriptive approach to explore older adults' perceptions and anticipated emotional experiences regarding SARs within their specific sociocultural context. To address the challenge that SARs are not yet widely integrated into care for older adults in China, we utilized a semi‐structured interview format integrated with scenario‐based techniques. Therefore, we used a scenario‐based approach commonly applied in prospective qualitative research. Scenario‐based approaches are considered valuable in prospective qualitative research (Kosow and Gaßner 2008; van Notten et al. 2003). The approach combines fictional situations with open‐ended questions, encouraging participants to reflect on their anticipated emotional and behavioral responses (Bradfield et al. 2005; Jaidin 2018). It has been effectively applied in healthcare, education, and technology (Nesi and Alsop 2021).

2.1. Ethical Statement

This study received approval from an institutional Ethics Committee (Approval No. 2024‐755). All participants provided written and verbal informed consent prior to participation, in accordance with the approved study protocol and ethical requirements. To protect confidentiality, personal information was anonymized using pseudonyms, and participants were assigned unique identifiers (N1‐N18). Audio recordings and transcripts were securely stored and accessible only to the research team. Participants were free to pause or terminate interviews at any time, and any withdrawn data were permanently deleted.

2.2. Participant Recruitment

Participants were recruited using maximum variation sampling to seek variation in gender, urban–rural residence, and educational background. We had no prior relationship with participants. Recruitment was conducted through hospitals, community centers, universities for older adults, and affiliated townships, with assistance from institutional staff and community workers. The sample included both community‐active older adults and those living at home. Inclusion criteria were adults aged 65 years and above. Exclusion criteria included severe cognitive impairment, significant difficulty in verbal expression, diagnosed severe mental illness (e.g., schizophrenia or major depression), or acute serious illness.

In accordance with the SAGER guidelines, the term “gender” refers to participants' self‐identified social identity (man/woman). Sex was not clinically assessed, and no participant reported a gender identity differing from their sex assigned at birth. Six invited eligible individuals declined participation due to lack of interest or time constraints. During the interviews, several participants spontaneously mentioned their familiarity with digital technologies, such as using smartphones (N1, N4, N8, N9), owning robotic vacuum cleaners (N11), or referring to robots as intelligent household appliances (N7). None reported prior direct experience with socially assistive robots. These spontaneous accounts offered contextual insight into participants' general technological familiarity. The final sample comprised 18 older adults (Table 1).

TABLE 1.

Characteristics of interview participants.

ID Gender Age (years) Residence a Education level b
N1 M 66 Urban Senior High School
N2 F 68 Urban Junior High School
N3 F 66 Rural Illiterate
N4 M 65 Urban Junior High School
N5 F 66 Urban Senior High School
N6 F 71 Urban Junior High School
N7 F 68 Urban Bachelor's Degree
N8 M 70 Rural Primary School
N9 F 68 Rural Primary School
N10 M 78 Urban Senior High School
N11 M 65 Urban Master's Degree
N12 F 81 Rural Illiterate
N13 F 70 Urban Primary School
N14 M 78 Rural Primary School
N15 M 66 Urban Junior College Diploma
N16 M 65 Urban Senior High School
N17 F 65 Urban Illiterate
N18 M 67 Rural Senior High School
Open in a new tab
a

Residence: Urban = residing in municipal districts; Rural = townships/villages.

b

Education Level: Illiterate = no formal schooling; Primary School = elementary education (typically ≤ 6 years); Junior High School = junior secondary education; Senior High School = senior secondary education; Associate's Degree = junior college diploma; Bachelor's Degree = undergraduate degree; Master's Degree = postgraduate degree.

2.3. Data Collection

2.3.1. Interview Tools

Scenario‐based materials and a semi‐structured interview guide were used as the primary interview tools. The scenario‐based approach is an established qualitative methodology for eliciting beliefs and normative expectations in prospective technology research (Nivette et al. 2024). This design accommodates the reality that participants lacked prior direct contact with SARs. Consequently, the results should be understood as indicative of expected attitudes, rather than experience‐based assessments derived from HRI.

The scenario was centered around a relatable fictional character, Mr. Wang (a 65‐year‐old older adult) and his socially assistive robot “LELE” (Table 2). The scenario content was informed by existing literature on socially assistive robots in care for older adults and common care situations reported in previous studies, such as companionship, daily assistance, health monitoring, and emergency support (Fan et al. 2026; Góngora Alonso et al. 2019). These situations reflect typical applications of SARs in older adult care. Table 2 presents the scenario material used in the interview. The complete scenario‐based interview guide, including the core questions and follow‐up probes, is provided in Supporting Information S1.

TABLE 2.

Scenario materials.

Imagine a future where socially assistive robots are integral to daily life, especially in living and care settings for older adults. 65‐year‐old Mr. Wang has a SAR named “LELE” because his family can't be with him often. “LELE” is a powerful robot that reminds Mr. Wang to take his medication, gives him dietary advice, and keeps him company. For example, when Mr. Wang laments the quietness of his home during dinner, “LELE” mentions his favorite braised pork, evoking warm memories of his mother's cooking. At first, Mr. Wang did not have high expectations for “LELE”, but over time, through repeated daily interactions, “LELE” gradually became a close companion. After reading this scenario, participants were invited to reflect on their own potential relationships with such a robot.
Open in a new tab

To ensure theoretical depth, the interview guide was primarily informed by attachment theory (Bowlby 1969).

  1. Safe Haven, defined as the tendency to actively seek protection and comfort from an attachment figure during times of fear or distress, led us to explore whether older adults might seek robots as an emotional refuge during vulnerable moments. We were therefore interested in questions such as “Under what specific circumstances do you think you would develop attachment to the robot?”, followed by more specific probes, for example, “If you were at home alone and felt unwell or frightened, how would you want a robot to help you?”.

  2. Secure Base, defined as the attachment figure acting as a reliable foundation that enables the individual to safely explore the external environment, led us to examine whether SARs might be perceived as affecting older adults' daily social lives. This directed our attention to questions such as “Would using a socially assistive robot change your daily interactions with family or friends?”

  3. Separation Anxiety, defined as the manifestation of anxiety and distress when separated from the attachment figure, pointed our inquiry toward whether attachment to the robot might influence meaningful human connections. We explored this through questions like “If an older adult relies completely on a robot, do you think there would be some consequences?”, followed by more specific probes, for example, “Would you worry that relying on a robot might affect the relationship between you and your family members, or between you and other people close to you, such as friends or neighbors?” These questions were designed to explore potential concerns associated with attachment to SARs, including possible implications for family relationships and social connectedness. This approach aimed to ensure that the scenario‐based interviews were grounded in both attachment theory and the cultural context.

After developing the preliminary interview guide, four older adults who met the inclusion criteria were recruited for pilot interviews. The pilot interviews revealed that some early questions were overly abstract. For example, broad questions about participants' views on human‐robot relationships often elicited brief or uncertain responses. These questions were subsequently revised into more concrete scenario‐based prompts, such as asking participants how they might feel if Mr. Wang received daily companionship or assistance from the robot. This adjustment helped participants engage more easily with the imagined situations while preserving the open‐ended nature of the interviews. The pilot phase helped refine the wording, sequence, and probing strategy of the interview guide, ensuring that participants could articulate their anticipated experiences and concerns without being directed toward particular viewpoints.

2.3.2. Interview Process

Interviews were conducted independently by two trained researchers who had undergone preparation in qualitative interviewing techniques. To ensure informed participation, each session commenced with an explanation of the study's purpose, the researchers' roles, and the absence of any conflict of interest. To facilitate engagement with the topic, participants were shown introductory visual materials depicting socially assistive robots before receiving a printed copy of the scenario materials (Table 2). Further information about these materials is provided in Supporting Information S2. All interviews were held in private settings, such as hospital consultation rooms, community centers, or classrooms at universities for older adults, lasting approximately 30–50 min. A research assistant was occasionally present to manage technical equipment, such as audio recording, without interacting with the participants. Researchers adhered to the semi‐structured guide while flexibly adjusting the language, question sequence, and tone to ensure clarity and participant comfort (e.g., using simplified expressions or clarifying probes). A consistent opening script was employed across all sessions to maintain comparability. Data collection continued until thematic saturation was achieved. As no new codes emerged after the 15th interview, and the subsequent three interviews (16–18) confirmed thematic stability, the research team determined that further data collection would not yield additional insights. Consequently, the study was concluded with a total sample of 18 participants (Table 1).

2.4. Data Analysis

Data were analyzed using Clarke and Braun's six‐phase thematic analysis (Clarke and Braun 2013). Two researchers independently coded all 18 transcripts using NVivo 12 to promote analytic rigor. In Phase 1 (familiarization), the researchers repeatedly read the transcripts, which comprised approximately 65 000 words, to gain an overall understanding of participants' anticipated attitudes toward SARs. In Phase 2 (generating initial codes), line‐by‐line inductive coding was conducted, resulting in 38 initial codes. In Phase 3 (searching for themes), related codes were compared and grouped into preliminary subthemes. For example, codes such as “loneliness,” “empty nest,” and “lack of companionship” were grouped under the subtheme “Loneliness as a Perceived Pathway to Attachment.” In Phase 4 (reviewing themes), the two researchers independently examined the relationships among codes and subthemes and compared their coding structures during regular meetings. Discrepancies, such as whether human‐like qualities should be classified as a catalyst of attachment or as a separate theme, were resolved through discussion and by revisiting the original transcripts. In Phase 5 (defining and naming themes), the subthemes were refined, clearly defined, and consolidated into three overarching themes: Anticipated and imagined emotional attachment to SARs; Perceived catalysts of anticipated emotional attachment; Anticipated consequences of emotional attachment. In Phase 6 (producing the report), the final themes were reviewed in relation to the research objectives and supported by representative quotations. Inter‐coder agreement (Cohen's κ) exceeded 0.70, indicating acceptable consistency. The resulting subthemes were then examined in relation to the research objectives and mapped onto broader conceptual categories to identify and refine overarching themes. No new codes or themes emerged after the fifteenth interview, and subsequent interviews yielded no substantially new insights. Based on established criteria for thematic saturation, data saturation was considered achieved (Mack et al. 2005).

2.5. Rigor and Researcher Reflexivity

Methodological rigor was enhanced through strategies addressing credibility, dependability, and reflexivity. We maintained a reflexive stance to enhance the trustworthiness of our findings. The interviewers were two female nursing graduate students. Their healthcare backgrounds helped build trust with the older adults, but might also have caused a “Hawthorne effect” where participants expressed more acceptance of robots to please us. To reduce this bias, we used neutral questions and encouraged participants to share their honest worries or negative views without judgment. To help ensure that our interpretation remained grounded in the participants' expressed perspectives and anticipated concerns rather than our own clinical assumptions, we continuously compared newly emerging themes with the original interview records. This process helped to ensure that every theme was strongly supported by the raw data.

3. Results

Three themes and eight subthemes (Table 3) were identified through thematic analysis.

TABLE 3.

Themes with verbatim extracts on emotional attachment to socially assistive robots.

Themes Subthemes Verbatim extract (selected)
Anticipated and imagined emotional attachment to SARs Welcoming or skeptical anticipations

“Living with a robot every day, of course I will develop feelings for it.” (N8)

“Robots' emotions are different from human emotions; robots only exhibit emotions through human input, so the emotions they generate are still artificial.” (N5)
Expecting that anticipated attachment would grow over time “Its communication brings joy over time, and after getting used to it, attachment may develop.” (N4)
Envisioning anticipated attachment in specific caregiving moments “I don't need it now, but I might grow attached if I live alone at home in the future. If my spouse isn't around and I faint from low blood sugar, the robot could bring me sugar; having a robot that could help would make a big difference.” (N17)
Perceived catalysts of anticipated emotional attachment Loneliness as a perceived pathway to anticipated attachment “Empty‐nesters feel lonely, so relying on robots is helpful. They can chat and assist with tasks.” (N15)
Insufficient family presence as a perceived condition for anticipated attachment “Older adults who live alone need robots, because children are often unavailable, while robots can provide consistent companionship, acting as a ‘Lao Ban’ (companion in old age).” (N2)
Projection of human‐like qualities as a perceived basis for connection “If SARs have a gentle tone, I will feel attached.” (N17)
Anticipated consequences of emotional attachment Expected emotional support alongside concerns about family ties

“My son juggles child rearing and caring for me. A robot companion would enhance our communication.” (N10)

“I might prefer my children not to visit if the robot already manages my daily needs, their visits annoy me, maybe I just need children's financial support.” (N16)
Anticipated concerns about autonomy “Relying too much on the robot, letting it do everything without thinking, might cause ‘Nao zi sheng xiu’ (my brain to get rusty).” (N9)
Open in a new tab

3.1. Anticipated and Imagined Emotional Attachment to SARs

Participants expressed a range of beliefs and wishes about what emotional attachment to SARs might look like. Three subthemes were identified within this theme.

3.1.1. Welcoming or Skeptical Anticipations

Across their accounts, participants expressed an anticipated tension between emotional openness and functional skepticism. Several participants envisioned a natural acceptance of the attachment, perceiving it as a possible outcome of daily companionship:

Living with a robot every day, of course I will develop feelings for it. (N8, Male, 70, Rural, Primary School)

In contrast, others drew a clear boundary between human and artificial interaction, expressing doubts about the possibility of establishing genuine bonds with non‐human entities:

I need a real person to talk to; these robots don't make sense. (N2, Female, 68, Urban, Junior High School)

Robots' emotions are different from human emotions; robots only exhibit emotions through human input, so the emotions they generate are still artificial. (N5, Female, 66, Urban, Senior High School)

These participants anticipated interactions with SARs as simulated responses contingent on human programming rather than as authentic emotional exchanges.

3.1.2. Expecting That Anticipated Attachment Would Grow Over Time

Some participants imagined that potential attachment, if it were to develop, would emerge gradually over time rather than instantaneously. For instance, one participant anticipated that daily companionship would lead to a sense of habituation and longing:

After spending a long time with it every day, I got used to it and missed it. (N8, Male, 70, Rural, Primary School)

Another participant similarly speculated that emotional attachment could arise over time:

Its communication brings joy over time, and after getting used to it, attachment may develop. (N4, Male, 65, Urban, Junior High School)

These accounts reflected participants' belief that attachment would not develop as an immediate reaction, but rather as a gradual process they imagined would unfold through repeated and sustained HRI.

3.1.3. Envisioning Anticipated Attachment in Specific Caregiving Moments

Beyond temporal considerations, participants also perceived that emotional attachment might be situationally conditioned, particularly emerging in imagined moments of vulnerability or caregiving need. In such scenarios, robots were envisioned as sources of immediate and reliable support.

One participant described a hypothetical future situation in which living alone might increase emotional attachment to a robot during health‐related emergencies:

I don't need it now, but I might grow attached if I live alone at home in the future. If my spouse isn't around and I faint from low blood sugar, the robot could bring me sugar; having a robot that could help would make a big difference. (N17, Female, 65, Urban, Illiterate)

These accounts suggest that participants' beliefs and wishes regarding emotional attachment were primarily underpinned by an expressed need for dependable support during vulnerable moments.

3.2. Perceived Catalysts of Anticipated Emotional Attachment

Participants attributed anticipated emotional attachment to three perceived catalysts: feelings of loneliness, limited family support, and the projection of human‐like qualities. These factors reflected participants' beliefs about what might foster such bonds.

3.2.1. Loneliness as a Perceived Pathway to Anticipated Attachment

Participants speculated that loneliness would be an important perceived catalyst for developing emotional attachment to SARs. In hypothetical scenarios, robots were perceived as a potential source of companionship that could partially fill the emotional void associated with social isolation.

For lonely older adults, it is great because it keeps them company and chats with them. (N2, Female, 68, Urban, Junior High School)

Similarly, another participant envisioned that the constant presence of a robot would be particularly helpful for empty‐nesters to mitigate their sense of isolation:

Empty‐nesters feel lonely, so relying on robots is helpful. They can chat and assist with tasks. (N15, Male, 66, Urban, Junior College Diploma)

These reflections suggest that, in participants' accounts, the anticipated emotional attachment to robots was closely linked to feelings of loneliness, particularly in situations where social interaction was perceived as limited.

3.2.2. Insufficient Family Presence as a Perceived Condition for Anticipated Attachment

Participants also anticipated that a lack of consistent family presence would be an important perceived condition for emotional attachment to SARs. In the context of children being unavailable due to work or distance, robots were perceived as a stable alternative for companionship. As one participant expressed:

Older adults who live alone need robots, because children are often unavailable, while robots can provide consistent companionship, acting as a ‘Lao Ban’ (companion in old age). (N2, Female, 68, Urban, Junior High School)

Such reflections suggest that anticipated emotional attachment was envisioned as more likely to develop in contexts where traditional family‐based companionship was perceived as insufficient or disrupted.

3.2.3. Projection of Human‐Like Qualities as a Perceived Basis for Connection

Participants further speculated that attributing certain human qualities, such as gentleness, to the robot could facilitate emotional resonance and foster a sense of attachment.

Robots will always be gentle and even‐tempered. (N9, Female, 68, Rural, Primary School)

If SARs have a gentle tone, I will feel attached. (N17, Female, 65, Urban, Illiterate)

These accounts suggest that participants attributed human qualities, such as gentleness and emotional responsiveness, to the robot, interpreting these qualities as fostering anticipated emotional attachment.

3.3. Anticipated Consequences of Emotional Attachment

Participants also shared their wishes and concerns about how emotional attachment to SARs might affect their lives.

3.3.1. Expected Emotional Support Alongside Concerns About Family Ties

When asked about the anticipated impacts of attachment, participants' accounts clustered around two dimensions. On the one hand, some participants anticipated that emotional attachment to robots might support family communication. As one participant emphasized:

My son juggles child rearing and caring for me. A robot companion would enhance our communication. (N10, Male, 78, Urban, Senior High School)

On the other hand, participants also anticipated that attachment to SARs for emotional comfort might reduce their expectations for emotional closeness with their children. As one participant shared:

With a robot comforting me, I might not fixate on my children. (N17, Female, 65, Urban, Illiterate)

Another participant expressed this concern more directly:

I might prefer my children not to visit if the robot already manages my daily needs, their visits annoy me, maybe I just need children's financial support. (N16, Male, 65, Urban, Senior High School)

These narratives suggest that participants imagined emotional attachment to SARs as potentially reshaping intergenerational relationships, either by alleviating perceived family strain or by reducing expectations for emotional closeness with children.

3.3.2. Anticipated Concerns About Autonomy

Participants anticipated that emotional attachment to SARs might lead them to rely more on robotic assistance in daily activities, raising concerns about becoming less actively involved in doing things themselves.

One participant stated:

I can still cook independently, but might eventually rely on the robot to handle meal preparation. (N5, Female, 66, Urban, Senior High School)

Furthermore, some participants anticipated that prolonged attachment to robots might reduce their active mental engagement.

Relying too much on the robot, letting it do everything without thinking, might cause ‘Nao zi sheng xiu’ (my brain to get rusty). (N9, Female, 68, Rural, Primary School)

This metaphor reflected a concern that relying too much on robots might lead older adults to think less and participate less in daily activities. Participants worried that this reduced involvement could gradually weaken their sense of autonomy and self‐management in future care situations.

4. Discussion

This study explored Chinese older adults' anticipated emotional attachment to socially assistive robots through scenario‐based qualitative interviews, yielding three interrelated themes. Participants held mixed orientations toward the prospect of human‐robot bonding: some imagined attachment developing naturally through daily companionship, while others questioned whether such bonds could be genuinely meaningful. Their anticipated attachment was primarily attributed to pervasive loneliness, the perceived absence of consistent family support, and the human‐like qualities they projected onto robots. In envisioning the consequences of such attachment, participants anticipated both emotional benefits, particularly relief from isolation, and ethical concerns, including the potential erosion of family obligations and threats to their sense of relational autonomy.

This study found that older adults expressed ambivalent attitudes toward anticipated emotional attachment to SARs, characterized by both acceptance and skepticism. Some participants expected attachment to develop through daily companionship. This reflects a similar expectation that emotional bonds may emerge through sustained companionship (Kühne and Jeglinski geb. Mende 2022; Martens and Hildebrand 2021). Others emphasized that SARs' emotional responses were pre‐programmed. This aligns with the view that simulated systems cannot truly understand human emotions and with the concept of “false intimacy” (Turkle 2017). HRI may therefore create an artificial sense of closeness (Arnd‐Caddigan 2015). Furthermore, some participants anticipated that emotional attachment to SARs would develop gradually, describing “time spent together” as essential. This finding is consistent with previous research suggesting that “interaction duration fosters emotional connection” (Forlizzi and Disalvo 2006). This expectation should be interpreted cautiously. Studies of real‐world interactions with physical robots have shown that users' initial enthusiasm may be influenced by the “novelty effect” and may decline as technical limitations become apparent (Helmi et al. 2026; Leite et al. 2013). Therefore, in this scenario‐based study, such accounts should not be understood as evidence that sustained attachment would necessarily occur in actual SARs use. Rather, they may reflect older adults' wishes for stable companionship, emotional reassurance, and continuous care support in later life. Future field‐based or longitudinal studies are needed to examine whether and how anticipated attachment changes after direct interaction with SARs.

Our findings also suggested that participants' anticipated attachment was closely tied to specific caregiving situations. When envisioning potential future challenges (such as living alone or sudden illness), older adults tended to view SARs as reliable sources of timely support. This expectation should not be understood as an objective assessment of current SARs capabilities. Rather, it reflects unmet needs for security, timely support, and caregiving assistance. This pattern may be understood in relation to the concept of “secondary attachment” (Rabb et al. 2021). When primary interpersonal attachment resources, such as care from relatives, are unavailable, individuals may project their unmet need for security onto non‐human entities, forming a compensatory bond. In this study, expectations toward SARs in situations such as sudden illness appeared to arise when primary caregiving resources were perceived as unavailable, a pattern consistent with the notion of secondary attachment. In future nursing practice, caregivers should treat these expectations as clues for identifying unmet safety and home‐care needs, thereby providing more precise interventions and support.

Our findings suggest that participants' beliefs and wishes regarding emotional attachment were closely tied to loneliness. In many family‐oriented societies like China, loneliness reflects broader tensions between traditional filial expectations and contemporary social realities (Chen et al. 2014). Participants suggested that interacting with SARs might alleviate loneliness. Previous research has often focused on the direct effects of robotic functions on loneliness. However, our findings suggest that the perceived emotional meaning of the relationship may also matter, alongside the robot's functional capabilities. From the perspective of attachment theory, the mere presence of an attachment figure can provide individuals with emotional regulation and a sense of security (Bowlby 1969). In this study, participants seemed to expect that an emotionally supportive relationship with a robot could offer psychological comfort. In addition to loneliness, participants also viewed human‐like qualities, such as “kindness”, as important for anticipated emotional attachment. Previous research has shown that older adults project human characteristics onto robots to gain a sense of familiarity (Wan and Chen 2021). Our study further suggests that, compared to realistic physical features, older adults in this region place greater emphasis on the emotional connection established during interaction. This is consistent with recent findings that older adults value the sense of warmth conveyed by companion robots more than their human‐like physical features (Li and Wang 2025). It also echoes the ongoing focus in the field of HRI on robots' social presence and the quality of the relationship (Conde et al. 2024; Kadylak et al. 2023). Therefore, within a local care context that values family warmth, future SARs in nursing practice should place greater emphasis on the emotional warmth of interactions, thereby responding more precisely to older adults' unmet need for companionship.

Our findings suggest that older adults' anticipated emotional attachment to SARs may reshape their perceptions of intergenerational relationships. Previous research has shown that when family companionship is limited, older adults may project familial roles onto AIs, viewing them as emotional substitutes (Moon et al. 2020; Song et al. 2026). Our study suggests that the anticipated substitution effect may extend beyond psychological compensation. Some older adults imagined that if SARs could provide sufficient emotional support, their expectations for children's companionship and emotional closeness might gradually decrease. In these imagined situations, the emotional role of children could become less central, while their material support might be perceived as sufficient. The ethical meaning of this possible shift needs to be understood within the cultural contexts. In some Western discussions of aging, reduced dependence on children is often discussed in relation to independence and autonomy (Marnfeldt and Wilber 2025). However, in Confucian cultural contexts, such as China, caregiving is not merely the fulfillment of functional needs but also a core ethical bond that sustains family responsibilities, intergenerational obligations, and the practice of filial piety (Delman 2011). From this perspective, participants' concerns echo the ethical concern raised that AI robotic companions may reshape familial bonds and traditional norms of filial responsibility (Muyskens et al. 2024). For nursing practice, this suggests that the introduction of SARs should be accompanied by attention to how technological companionship may interact with older adults' perceptions of their role and value within family relationships.

In addition to concerns about family relationships, participants also worried that their anticipated attachment to SARs might undermine their sense of autonomy. While HRI research often highlights the cognitive benefits of robotic assistance (Ruan and Cheung 2025; Varrasi et al. 2025), participants in our study emphasized a different aspect of the experience, expressing concerns that over‐attachment to robotic assistance might make them think less, do less by themselves, and participate less in daily activities. As one older adult put it, “Over‐attachment may lead to ‘nao‐zi‐sheng‐xiu’” (N9). Such concerns extend beyond individual functioning and relate to whether older adults can maintain agency and self‐management in everyday life. This concern can be understood through the lens of relational autonomy. Relational autonomy suggests that autonomy is shaped within social relationships rather than in isolation (Mackenzie and Stoljar 2000). In this study, participants' concerns about becoming overly reliant on SARs suggest a desire to stay involved in daily activities and decision‐making. These concerns do not indicate a rejection of assistance. Instead, they reflect the importance of maintaining agency in daily life. Participants' views on autonomy may also reflect the local care context. Previous studies in Western care settings have often discussed autonomy in relation to self‐determination, control, and privacy in technology use (Berridge et al. 2022; Bölenius et al. 2023). In China's family‐oriented care context, participants mainly expressed concerns about becoming passive or overly dependent on robotic assistance. This suggests that autonomy‐related concerns may be linked to expectations about daily functioning and care. This highlights the importance of supporting older adults' continued participation in daily activities, family interactions, and care‐related decision‐making when introducing SARs into care practice.

Taken together, these findings suggest that anticipated emotional attachment to SARs should not be understood solely as a response to technological characteristics, but also as a reflection of older adults' unmet emotional and caregiving needs. Participants interpreted potential relationships with SARs through existing family‐based expectations of care, while simultaneously expressing concerns about the possible implications for intergenerational relationships, family role identity, and active participation in care. In the future, nursing practice should ensure that technological support addresses unmet care needs while preserving older adults' relational autonomy and social connectedness.

5. Limitations and Future Research

This study has several limitations. First, most participants lacked direct experience with SARs, which remain uncommon in care for older adults in China. Scenario‐based interviews were therefore used to elicit anticipated responses to imagined situations. While this approach enabled us to explore older adults' expectations and concerns, the findings likely reflect anticipatory expectations rather than experience‐based evaluations of HRI and may not fully correspond to real‐world behavior. Participants' limited familiarity with SARs may also have contributed to idealized expectations, particularly regarding robotic assistance during acute health events. Rather than being treated merely as a methodological limitation, such misunderstandings can be seen as a finding in their own right: they reveal not only what older adults do not know about current technology but also what they hope technology could do for them. Second, participants were recruited from a single region in central China, and their experiences may differ from those of older adults in other regions. Furthermore, additional contextual details could have enriched our interpretation. Although several participants spontaneously mentioned their familiarity with digital technologies or smart household devices, these factors were not assessed using structured measures or used for subgroup comparison. We also did not systematically assess socioeconomic background or level of functional independence. Future studies could combine scenario‐based interviews with field trials involving actual SARs to examine how anticipated attachment evolves when users directly experience the technology. Multi‐center studies involving participants from diverse regions are also recommended.

6. Conclusion

This study explored Chinese older adults' beliefs and wishes regarding anticipated emotional attachment to SARs through scenario‐based interviews. The findings suggest that such attachment was imagined in relation to unmet emotional and caregiving needs, especially loneliness, limited family support, and expectations for timely support in vulnerable situations. Participants viewed SARs as potential sources of companionship and reassurance, but also expressed concerns that emotional reliance on robots might reshape intergenerational relationships, weaken family role identity, or reduce active participation in care. These findings indicate that anticipated attachment is embedded in China's family‐based care context rather than arising from technology alone. Nursing practice should therefore address the care needs underlying anticipated attachment and ensure that SARs complement, rather than replace, human care while preserving relational autonomy and social connectedness.

7. Relevance for Clinical Practice

The findings suggest that integrating SARs into care for older adults is not merely a technological enhancement but a relational intervention that warrants nursing attention.

  1. Assessment: Prior to the introduction of SARs, nurses may consider assessing older adults' levels of loneliness and family context, as these factors appeared to shape expectations of emotional support and anticipated attachment in this study.

  2. Education: The findings suggest a need for nurses to support communication with families regarding the role of SARs as supplementary tools rather than replacements for familial care, which may help address older adults' concerns about the perceived weakening of family care.

  3. Monitoring: For older adults living alone, nurses may remain attentive to changes in social engagement patterns, including increased emotional attachment to SARs, to help ensure that HRI complements rather than substitutes for existing social connections.

Author Contributions

Jiahui Chen: investigation. Ying He: conceptualization, writing – review and editing, supervision, resources, funding acquisition. Junjin An: funding acquisition. Jinchang Yu: data curation. Jixia Cao: conceptualization, methodology, formal analysis, writing – original draft, writing – review and editing, project administration. Can Yang: investigation, formal analysis, validation. Xingru Yang: data curation.

Funding

This work was supported by: National Social Science Foundation of China (grant no. 22BZX039); Hunan Normal University Student Innovation and Entrepreneurship Training Program (grant no. 2024266).

Ethics Statement

This study received approval from the Hunan Normal University Biomedical Research Ethics Committee (approval no. 2024‐755). Written and verbal informed consent was obtained from all participants prior to the interviews. Participants were informed of their right to withdraw from the study or request data deletion at any time.

Conflicts of Interest

The authors declare no conflicts of interest.

Supporting information

Data S1: nhs70410‐sup‐0001‐Supinfo1.docx.

NHS-28-e70410-s002.docx (57.6KB, docx)

Data S2: nhs70410‐sup‐0002‐Supinfo2.docx.

NHS-28-e70410-s001.docx (503.4KB, docx)

Acknowledgments

We sincerely thank all participants for sharing their experiences, as well as the staff at participating institutions (hospitals, community centers, and universities for older adults) for their support in participant recruitment. The authors acknowledge the use of ChatGPT (OpenAI, accessed in 2026) to assist with English language editing, including improving sentence structure, readability, coherence, and the overall organization of the manuscript, particularly in the Discussion section. ChatGPT was also consulted for alternative suggestions regarding the naming of qualitative themes; however, all final theme names were independently determined by the authors based on the coding results and interpretation of the data. No AI tools were used for study design, data collection, qualitative coding, data analysis, interpretation of findings, or development of the study conclusions. All AI‐assisted suggestions were carefully reviewed, verified, and revised or rejected as appropriate by the authors. The authors take full responsibility for the accuracy, integrity, and originality of the manuscript.

Data Availability Statement

The data generated and analyzed during this study are not publicly available in order to protect participant confidentiality. De‐identified data may be obtained from the corresponding author, Ying He (yinghe@hunnu.edu.cn), upon reasonable request and in compliance with ethical and privacy guidelines.

References

  1. Arnd‐Caddigan, M. 2015. “Sherry Turkle: Alone Together: Why we Expect More From Technology and Less From Each Other.” Clinical Social Work Journal 43, no. 2: 247–248. 10.1007/s10615-014-0511-4. [DOI] [Google Scholar]
  2. Berridge, C. , Zhou Y., Lazar A., et al. 2022. “Control Matters in Elder Care Technology: Evidence and Direction for Designing It in.” DIS (Des Interact Syst Conf) 2022: 1831–1848. 10.1145/3532106.3533471. [DOI] [PMC free article] [PubMed] [Google Scholar]
  3. Bölenius, K. , Lämås K., and Edvardsson D.. 2023. “Older Adults' Experiences of Self‐Determination When Needing Homecare Services—An Interview Study.” BMC Geriatrics 23: 824. 10.1186/s12877-023-04533-6. [DOI] [PMC free article] [PubMed] [Google Scholar]
  4. Bowlby, J. 1969. Attachment and Loss. Vol. 3. 1991st ed. Hogarth Press: Institute of Psycho‐Analysis. [Google Scholar]
  5. Bradfield, R. , Wright G., Burt G., Cairns G., and Van Der Heijden K.. 2005. “The Origins and Evolution of Scenario Techniques in Long Range Business Planning.” Futures 37, no. 8: 795–812. 10.1016/j.futures.2005.01.003. [DOI] [Google Scholar]
  6. Chen, K. , Lou V. W., Tan K. C., Wai M. Y., and Chan L. L.. 2020. “Effects of a Humanoid Companion Robot on Dementia Symptoms and Caregiver Distress for Residents in Long‐Term Care.” Journal of the American Medical Directors Association 21, no. 11: 1724–1728.e1723. 10.1016/j.jamda.2020.05.036. [DOI] [PubMed] [Google Scholar]
  7. Chen, S.‐C. , Moyle W., Jones C., and Petsky H.. 2020. “A Social Robot Intervention on Depression, Loneliness, and Quality of Life for Taiwanese Older Adults in Long‐Term Care.” International Psychogeriatrics 32, no. 8: 981–991. 10.1017/s1041610220000459. [DOI] [PubMed] [Google Scholar]
  8. Chen, Y. , Hicks A., and While A. E.. 2014. “Loneliness and Social Support of Older People Living Alone in a County of Shanghai, China.” Health & Social Care in the Community 22, no. 4: 429–438. 10.1111/hsc.12099. [DOI] [PubMed] [Google Scholar]
  9. Chen, Z. , Jiang W., Xu J., Jiang L., and Wang G.. 2022. “Ethical, Legal, and Social Implications of Genomics in China: A Scoping Review and Implications for Precision Public Health.” China CDC Weekly 4, no. 32: 706. 10.46234/ccdcw2022.147. [DOI] [PMC free article] [PubMed] [Google Scholar]
  10. Clarke, V. , and Braun V.. 2013. Successful Qualitative Research: A Practical Guide for Beginners. Sage. [Google Scholar]
  11. Conde, M. , Mikhailova V., and Döring N.. 2024. “‘I Have the Feeling That the Person Is Here’: Older Adults' Attitudes, Usage Intentions, and Requirements for a Telepresence Robot.” International Journal of Social Robotics 16, no. 7: 1619–1639. 10.1007/s12369-024-01143-z. [DOI] [Google Scholar]
  12. Delman, J. 2011. “The Individualization of Chinese Society by Yunxiang Yan.” Journal of Asian Studies 70: 219–221. 10.2307/41302237. [DOI] [Google Scholar]
  13. Dharmansyah, D. , Rahayuwati L., Pramukti I., and Mutyara K.. 2026. “Cultural Determinants of Assistive Health Technology Acceptance in Asian Older Adults in Low‐ and Middle‐Income Countries: A Mixed‐Methods Systematic Review With Evidence Predominantly From East Asia.” Patient Preference and Adherence 20: 606459. 10.2147/PPA.S606459. [DOI] [PMC free article] [PubMed] [Google Scholar]
  14. Epley, N. , Akalis S., Waytz A., and Cacioppo J. T.. 2008. “Creating Social Connection Through Inferential Reproduction: Loneliness and Perceived Agency in Gadgets, Gods, and Greyhounds.” Psychological Science 19, no. 2: 114–120. 10.1111/j.1467-9280.2008.02056.x. [DOI] [PubMed] [Google Scholar]
  15. Fan, B. , Lou V. W., and Huang Y.. 2026. “Social Robots in Elderly Care: A Systematic Review of the Effectiveness on Formal and Informal Caregivers.” International Journal of Social Robotics 18: 60. 10.1007/s12369-026-01396-w. [DOI] [Google Scholar]
  16. Follmann, A. , Schollemann F., Arnolds A., et al. 2021. “Reducing Loneliness in Stationary Geriatric Care With Robots and Virtual Encounters‐A Contribution to the COVID‐19 Pandemic.” International Journal of Environmental Research and Public Health 18, no. 9: 4846. 10.3390/ijerph18094846. [DOI] [PMC free article] [PubMed] [Google Scholar]
  17. Forlizzi, J. , and Disalvo C.. 2006. “Service Robots in the Domestic Environment: A Study of the Roomba Vacuum in the Home.” (Vol. 2006). 10.1145/1121241.1121286. [DOI]
  18. Góngora Alonso, S. , Hamrioui S., de la Torre Díez I., Motta Cruz E., López‐Coronado M., and Franco M.. 2019. “Social Robots for People With Aging and Dementia: A Systematic Review of Literature.” Telemedicine and e‐Health 25, no. 7: 533–540. 10.1089/tmj.2018.0051. [DOI] [PubMed] [Google Scholar]
  19. Helmi, A. , Morales Mayoral R., Wang T.‐H., Logan S. W., and Fitter N. T.. 2026. “Go Go Gadget GoBot: Leaning Into the Novelty Effect for Reengagement With Modular Robotic Hardware.” International Journal of Social Robotics 18, no. 2: 37. 10.1007/s12369-025-01344-0. [DOI] [Google Scholar]
  20. Hou, H. , Wei H., Wang Y., Yu J., and Qiu G.. 2022. “Construction Path of Smart Health Care Platform for the Elderly in China.” Chinese Journal of Engineering Science 24: 170. [Google Scholar]
  21. Jaidin, J. H. 2018. “Scenario‐Based Interview: An Alternative Approach to Interviewing Children?” Pacific Early Childhood Education Research Association 12: 23–37. [Google Scholar]
  22. Kadylak, T. , Bayles M. A., and Rogers W. A.. 2023. “Are Friendly Robots Trusted More? An Analysis of Robot Sociability and Trust.” Robotics 12, no. 6: 162. [Google Scholar]
  23. Kosow, H. , and Gaßner R.. 2008. “Methods of Future and Scenario Analysis. Overview, Assessment, and Selection Criteria.”
  24. Kühne, K. , and Jeglinski geb. Mende M. A.. 2022. “Tamagotchi on Our Couch: Are Social Robots Perceived as Pets?”. 10.13140/RG.2.2.23236.30089. [DOI]
  25. Law, T. , Chita‐Tegmark M., Rabb N., and Scheutz M.. 2022. “Examining Attachment to Robots: Benefits, Challenges, and Alternatives.” ACM Transactions on Human‐Robot Interaction 11: 1–18. [Google Scholar]
  26. Lee, J. , and Joo C.. 2025. “Can Social Robots Reduce the Loneliness of Older Adults? ‐Interaction and Communication Between Social Robots and Older Adults.” Geriatric Nursing 64: 103359. 10.1016/j.gerinurse.2025.04.032. [DOI] [PubMed] [Google Scholar]
  27. Leite, I. , Martinho C., and Paiva A.. 2013. “Social Robots for Long‐Term Interaction: A Survey.” International Journal of Social Robotics 5, no. 2: 291–308. 10.1007/s12369-013-0178-y. [DOI] [Google Scholar]
  28. Li, Q. , and Wang L.. 2025. “Evaluating the Physical Appeal of Elderly Companion Robots: A Study of Cuteness Versus Anthropomorphism.” IEEE Access 13: 70626–70639. 10.1109/ACCESS.2025.3558896. [DOI] [Google Scholar]
  29. Li, X. D. , Cao H. J., Xie S. Y., et al. 2022. “Urban‐Rural Differences in the Effect of Empty‐Nest on Mental Health and Behaviors of Chinese Older Population.” Psychology, Health & Medicine 27, no. 6: 1334–1346. 10.1080/13548506.2021.1874439. [DOI] [PubMed] [Google Scholar]
  30. Lill, L. 2020. “Staff Shortages in Swedish Elderly Care ‐ Reflections on Gender and Diversity Politics.” International Journal of Migration, Health and Social Care 16: 269–278. [Google Scholar]
  31. Liu, Z. X. 2025. “A Perspective on the Emotional Authenticity of Anthropomorphism in Nursing Robots.” Medicine, Health Care, and Philosophy 46, no. 17: 33–38. 10.12014/j.issn.1002-0772.2025.17.07. [DOI] [Google Scholar]
  32. Luo, C. H. , Yuan R., Mao B., Liu Q., Wang W. H., and He Y.. 2024. “Technology Acceptance of Socially Assistive Robots Among Older Adults and the Factors Influencing It: A Meta‐Analysis [Review].” Journal of Applied Gerontology 43, no. 2: 115–128. 10.1177/07334648231202669. [DOI] [PubMed] [Google Scholar]
  33. Luperto, M. , Romeo M., Monroy J., et al. 2021. “What is My Robot Doing? Remote Supervision to Support Robots for Older Adults Independent Living: A Field Study.” 2021 European Conference on Mobile Robots (ECMR).
  34. Mack, N. , Woodsong C., Macqueen K., Guest G., and Namey E.. 2005. Qualitative Research Methods: A Data Collector's Field Guide. Family Health International. [Google Scholar]
  35. Mackenzie, C. , and Stoljar N., eds. 2000. Relational Autonomy: Feminist Perspectives on Autonomy, Agency, and the Social Self. Oxford University Press. 10.1093/oso/9780195123333.001.0001. [DOI] [Google Scholar]
  36. Marnfeldt, K. , and Wilber K.. 2025. “The Safety–Autonomy Grid: A Flexible Framework for Navigating Protection and Independence for Older Adults.” Gerontologist 65, no. 6: gnaf111. 10.1093/geront/gnaf111. [DOI] [PMC free article] [PubMed] [Google Scholar]
  37. Martens, R. , and Hildebrand C.. 2021. “Dementia Care, Robot Pets, and Aliefs.” Bioethics 35, no. 9: 870–876. 10.1111/bioe.12952. [DOI] [PubMed] [Google Scholar]
  38. Matthews, N. R. , Porter G. J., Varghese M., et al. 2023. “Health and Socioeconomic Resource Provision for Older People in South Asian Countries: Bangladesh, India, Nepal, Pakistan and Sri Lanka Evidence From NEESAMA.” Global Health Action 16, no. 1: 2110198. 10.1080/16549716.2022.2110198. [DOI] [PMC free article] [PubMed] [Google Scholar]
  39. Moon, B. , Kwak S., Kang D., Lee H., and Choi J.. 2020. “The Effects of Internet of Robotic Things on In‐Home Social Family Relationships.” 10.1109/RO-MAN47096.2020.9223345. [DOI]
  40. Muyskens, K. , Ma Y., and Dunn M.. 2024. “Can an AI‐Carebot Be Filial? Reflections From Confucian Ethics.” Nursing Ethics 31, no. 6: 999–1009. 10.1177/09697330241238332. [DOI] [PubMed] [Google Scholar]
  41. Nakamura, M. , Okajima K., Matsumoto Y., Tanaka T., Iijima K., and Nihei M.. 2021. “Effectiveness of Continuous Grip Strength Measurement Using Social Assistive Robots on Older Adults at Home.” Journal of Robotics and Mechatronics 33: 719–729. [Google Scholar]
  42. Nesi, H. , and Alsop S.. 2021. “Stories and Scenarios: Lecturers' Use of Fantastic Hypothetical Events.” Journal of English for Academic Purposes 53: 101022. 10.1016/j.jeap.2021.101022. [DOI] [Google Scholar]
  43. Nivette, A. , Nägel C., and Stan A.. 2024. “The Use of Experimental Vignettes in Studying Police Procedural Justice: A Systematic Review.” Journal of Experimental Criminology 20, no. 1: 151–186. 10.1007/s11292-022-09529-7. [DOI] [Google Scholar]
  44. Papadopoulos, C. , Castro N., Nigath A., et al. 2022. “The CARESSES Randomised Controlled Trial: Exploring the Health‐Related Impact of Culturally Competent Artificial Intelligence Embedded Into Socially Assistive Robots and Tested in Older Adult Care Homes.” International Journal of Social Robotics 14, no. 1: 245–256. 10.1007/s12369-021-00781-x. [DOI] [PMC free article] [PubMed] [Google Scholar]
  45. Papadopoulos, C. , Hill T., Battistuzzi L., et al. 2020. “The CARESSES Study Protocol: Testing and Evaluating Culturally Competent Socially Assistive Robots Among Older Adults Residing in Long Term Care Homes Through a Controlled Experimental Trial.” Archives of Public Health 78, no. 1: 26. 10.1186/s13690-020-00409-y. [DOI] [PMC free article] [PubMed] [Google Scholar]
  46. Papadopoulos, I. , Wright S., Koulouglioti C., et al. 2023. “Socially Assistive Robots in Health and Social Care: Acceptance and Cultural Factors. Results From an Exploratory International Online Survey.” Japan Journal of Nursing Science 20, no. 2: e12523. 10.1111/jjns.12523. [DOI] [PubMed] [Google Scholar]
  47. Persson, M. , David R., and Iversen C.. 2022. “Caregivers' Use of Robots and Their Effect on Work Environment ‐ a Scoping Review.” Journal of Technology in Human Services 40, no. 3: 251–277. 10.1080/15228835.2021.2000554. [DOI] [Google Scholar]
  48. Pirhonen, J. , Tiilikainen E., Pekkarinen S., Lemivaara M., and Melkas H.. 2020. “Can Robots Tackle Late‐Life Loneliness? Scanning of Future Opportunities and Challenges in Assisted Living Facilities.” Futures 124: 102640. 10.1016/j.futures.2020.102640. [DOI] [PMC free article] [PubMed] [Google Scholar]
  49. Rabb, N. , Law T., Chita‐Tegmark M., and Scheutz M.. 2021. “An Attachment Framework for HRI.” International Journal of Social Robotics 14, no. 2: 539–559. 10.1007/s12369-021-00802-9. [DOI] [PMC free article] [PubMed] [Google Scholar]
  50. Rabbitt, S. M. , Kazdin A. E., and Scassellati B.. 2015. “Integrating Socially Assistive Robotics Into Mental Healthcare Interventions: Applications and Recommendations for Expanded Use.” Clinical Psychology Review 35: 35–46. 10.1016/j.cpr.2014.07.001. [DOI] [PubMed] [Google Scholar]
  51. Randall, N. , Bennett C. C., Šabanović S., et al. 2019. “More Than Just Friends: In‐Home Use and Design Recommendations for Sensing Socially Assistive Robots (SARs) by Older Adults With Depression.” Paladyn, Journal of Behavioral Robotics 10, no. 1: 237–255. 10.1515/pjbr-2019-0020. [DOI] [Google Scholar]
  52. Ruan, Y. X. , and Cheung M. C.. 2025. “Effectiveness of Group Interventions With Socially‐Assistive Robots for Older Adults: A Systematic Review.” Journal of Nursing Scholarship 57, no. 6: 919–940. 10.1111/jnu.70035. [DOI] [PMC free article] [PubMed] [Google Scholar]
  53. Shi, J. , Zhang N., Wu K., and Wang Z.. 2025. “Application Status, Challenges, and Development Prospects of Smart Technologies in Home‐Based Elder Care.” Electronics 14, no. 12: 2463. [Google Scholar]
  54. Song, T. , Sun B., Li J., et al. 2026. “Understanding Older Adults' Experiences of Support, Concerns, and Risks From Kinship‐Role AI‐Generated Influencers.” [Preprint]. arXiv. 10.48550/arXiv.2602.22993. [DOI]
  55. Søraa, R. A. , Tøndel G., Kharas M. W., and Serrano J. A.. 2022. “What Do Older Adults Want From Social Robots? A Qualitative Research Approach to Human‐Robot Interaction (HRI) Studies.” International Journal of Social Robotics 15, no. 3: 411–424. 10.1007/s12369-022-00914-w. [DOI] [Google Scholar]
  56. Su, D. , Wu X. N., Zhang Y. X., et al. 2012. “Depression and Social Support Between China's Rural and Urban Empty‐Nest Elderly.” Archives of Gerontology and Geriatrics 55, no. 3: 564–569. 10.1016/j.archger.2012.06.006. [DOI] [PubMed] [Google Scholar]
  57. Szondy, M. , and Fazekas P.. 2024. “Attachment to Robots and Therapeutic Efficiency in Mental Health.” Frontiers in Psychology 15: 1347177. [DOI] [PMC free article] [PubMed] [Google Scholar]
  58. Tamir, S. 2025. “Ethical Governance of AI‐Based Humanoid Carebots: The Case for Ethics of Techno‐Care.” Medicine, Health Care, and Philosophy 29: 75–87. 10.1007/s11019-025-10305-3. [DOI] [PMC free article] [PubMed] [Google Scholar]
  59. Turkle, S. 2017. Alone Together: Why we Expect More From Technology and Less From Each Other. Third ed. Basic Books. [Google Scholar]
  60. United Nations, Department of Economic and Social Affairs, Population Division . 2020. “World Population Ageing 2020 Highlights: Living Arrangements of Older Persons.”
  61. van Notten, P. W. F. , Rotmans J., van Asselt M. B. A., and Rothman D. S.. 2003. “An Updated Scenario Typology.” Futures 35, no. 5: 423–443. 10.1016/S0016-3287(02)00090-3. [DOI] [Google Scholar]
  62. Varrasi, S. , Vagnetti R., Camp N., et al. 2025. “How HRI Can Influence Task Performance and Perceived Cognitive Load at Different Support Conditions.” Information 16, no. 5: 374. 10.3390/info16050374. [DOI] [Google Scholar]
  63. Velmurugan, V. , Holthaus P., Amirabdollahian F., and Dragone M.. 2025. “Clinicians' Perspectives on Safety, Ethical, and Legal Considerations for Home‐Based Physical Rehabilitation Robots: Highlighting Key Themes.” IEEE Robotics and Automation Magazine 32, no. 4: 77–83. 10.1109/MRA.2025.3619330. [DOI] [Google Scholar]
  64. Wan, E. W. , and Chen R. P.. 2021. “Anthropomorphism and Object Attachment.” Current Opinion in Psychology 39: 88–93. 10.1016/j.copsyc.2020.08.009. [DOI] [PubMed] [Google Scholar]
  65. Wu, H. 2021. “The Empty Nests of Elderly Only‐Child Parents in Urban China—An Analysis Based on Survey Data From Five Provinces.” China Population and Development Studies 4, no. 4: 362–383. 10.1007/s42379-020-00064-4. [DOI] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

Data S1: nhs70410‐sup‐0001‐Supinfo1.docx.

NHS-28-e70410-s002.docx (57.6KB, docx)

Data S2: nhs70410‐sup‐0002‐Supinfo2.docx.

NHS-28-e70410-s001.docx (503.4KB, docx)

Data Availability Statement

The data generated and analyzed during this study are not publicly available in order to protect participant confidentiality. De‐identified data may be obtained from the corresponding author, Ying He (yinghe@hunnu.edu.cn), upon reasonable request and in compliance with ethical and privacy guidelines.

Articles from Nursing & Health Sciences are provided here courtesy of Wiley

RESOURCES