Abstract
Robots have the potential to perform various domestic tasks for older adults and support them to age in place. Previous studies investigating older adults’ preferences for robot assistance did not provide an alternative option for participants to choose. Such practice is incongruent with the concept of preference, which indicates a greater desire for something than another. It also discourages a direct comparison between different options. This study aims to identify the domestic tasks that older adults prefer robots to perform and the domestic tasks that they prefer humans to perform when an alternative option is available. This cross-sectional descriptive study assessed 385 domestic-living older adults’ preferences between human and robot assistance in 48 domestic tasks. Findings from a simple t-test confirmed that participants had a significant preference for human assistance in 13 domestic tasks and robot assistance in 33 other domestic tasks (p < .05). Compared with previous findings in 2014, this study identified an additional 18 domestic tasks that were preferred for robot assistance by older adults, implying that older adults are now more open to robot assistance. This study provides new insight into the development and choice of robots that meet the current preferences of older adults.
Introduction
Aging in place is defined as “the ability to live in one’s own home and community safely, independently and comfortably, regardless of age, income or level of capacity” (World Health Organization, 2015, p. 225). It is generally considered beneficial for older adults and has been one of the key goals for policymakers in the last two decades (World Health Organization, 2015). Most older adults wish to age in place (Bosch-Farré et al., 2020; Wu et al., 2011). Older adults should be able to reside safely, independently, and comfortably and continue living in an environment they are most familiar with (Bosch-Farré et al., 2020). However, aging commonly leads to declining functional abilities, imposing challenges in older adults’ daily living. Therefore, older adults generally need assistance to manage their living at home.
Workforce shortage in healthcare services, including elderly services, remains a challenge worldwide (Glarcher et al., 2025; Tamata & Mohammadnezhad, 2023). The supply of human resources cannot keep pace with the increasing demand. The increasing workforce challenges create a self-perpetuating cycle of burnout and attrition, further depleting the workforce. The cycle negatively affects the quality of care, adds strain to healthcare systems, and threatens the long-term sustainability of healthcare delivery (Glarcher et al., 2025). Innovations in healthcare, such as use of technologies, is identified as a way to cope with the workforce challenges (Glarcher et al., 2025). A scoping review indicates that robotic systems have long been recognized for their potential to ease workforce challenges, especially in healthcare (Abdi et al., 2018). Healthcare robots can be classified into three categories in terms of their key functions. Surgical robots assist medical staff in performing surgeries. Rehabilitation robots support individuals in body movement and recovery. Social robots interact with individuals and help with day-to-day activities (Bardaro et al., 2022). Among the three categories, social robots are more relevant to support aging in place.
Technological advancements have enabled social robots to perform a wider range of domestic tasks for older adults and support them in maintaining a higher level of independence, thereby giving them a higher chance to age in place. Social robots can be broadly classified into three categories according to their key functions. First, companion robots are designed to provide companionship and promote psychological well-being for their users (Abdi et al., 2018; Bardaro et al., 2022). For example, the Paro is a robotic seal that moves and makes sounds in response to touch and stimuli, providing comfort and emotional engagement. Another example is the Pepper, a humanoid robot equipped with a touch screen, which is capable of engaging in conversations with people, and recognizing emotions (Hung et al., 2021). Second, tele-operated robots facilitate remote interaction, enabling family members or caregivers to remotely manipulate the robot and communicate with older adults from a distance (Butler et al., 2015). For example, the Double is a telepresence robot that enables remote users to engage in video calls and interact with others via video and audio while moving around (G. Zhang & Hansen, 2022). Third, semi-autonomous assistive robots help with specific daily tasks autonomously but may need human guidance or assistance in complex or unfamiliar situations. These robots allow human-robot interaction (Graf et al., 2013; Mast et al., 2015). For example, the Roomba is a robotic vacuum cleaner that clean floors by itself but requires human set up and involvement for certain tasks such as emptying the dustbin.
The development of semi-autonomous robots carries specific implications. Robots are designed to support older adults with daily domestic tasks, enabling them to live independently at their own home for as long as possible. However, current robot technology does not support reliable fully autonomous operation in the varied environments of older adults’ homes. Thus, semi-autonomous robot control is developed. If the robot encounters a problem, it can reach out to the users or their remote family members for assistance and even contact professionals with advanced remote access to solve more complex problems (Mast et al., 2015). In other scenario, the robot can function in accordance with the commands or intention of the user (Z. Zhang et al., 2017). An issue arises regarding how robots can provide feedback to humans and how humans respond to this feedback. Using a semi-autonomous robotic vacuum cleaner as study intervention, a study found that robots provided feedback via various means such as visual, auditory, motion, and app-based feedback. Humans provided various responses such as having emotional reactions, building trust, modifying behavior, learning to interpret the feedback, and adapting the new behavior. These findings highlighted the importance of designing robots that trigger positive feelings and provide feedback that is coherent, appropriate to the situation, and non-disturbing to the user, in order to enhance user experience and engagement (Saplacan & Herstad, 2019). In addition, it is important to facilitate the older users to interpret the feedback from robots, leading to a better understanding of how to interact with the robot effectively. However, the consideration extends beyond simply providing support to older adults.
In the context of assistive robotics, scholars have expressed worries about the tendency to generalize the needs of older adults based on social factors, like aging demographics, instead of focusing on the unique and specific needs of individuals (Olatunji et al., 2024). Specific factors such as availability of physical space, ease of use, support system, digital confidence in handling the robots, and ethical issues in relation to privacy and security in handling personal information remain to be the challenges regarding deployment of robots for older adults (Olatunji et al., 2024). All these factors emphasize the necessity for a thorough investigation into the specific needs and preferences of older adults and how they envision robots supporting them.
While researchers have been investigating the effects of using robots in eldercare (Broekens et al., 2009), knowing older adults’ preferences before introducing robots is equally important. Such knowledge allows the introduction of robots to be personalized following the preferences of older adults and ensures greater acceptance in the future (Broadbent et al., 2009; Wu et al., 2011).
Studies have investigated older adults’ preferences for assistance in domestic tasks from robots (Broadbent et al., 2012; Chu et al., 2019; Ezer et al., 2009; Korchut et al., 2017; Mast et al., 2012; Smarr et al., 2014; Wu et al., 2011). These studies found that older adults preferred robots to assist in physically demanding domestic tasks, such as lifting heavy objects, cleaning, and laundry (Broadbent et al., 2012, 2013), and required managing information, such as reminding individuals and monitoring a condition (Broadbent et al., 2012; Smarr et al., 2014). In contrast, studies found that older adults preferred humans rather than robots to assist in highly personal domestic tasks that involve physical touch, such as bathing and dressing (Broadbent et al., 2012; Mast et al., 2012), shaving and performing hair care (Smarr et al., 2014). Older adults also preferred humans to assist in domestic tasks that required social interaction (Smarr et al., 2014).
While the reviewed studies provide insights into older adults’ preferences for robot assistance in domestic tasks, they investigated these preferences for robot assistance without offering an alternative option (i.e., human assistance) and failed to compare preferences for robot and human assistance (Broadbent et al., 2012; Mast et al., 2012). Concerning the definition of preference, which is defined as a greater desire for something than other things (Oxford Advanced Learner’s Dictionary, n.d.), a more desirable approach to investigate older adults’ preference for domestic assistance is to provide alternative sources of assistance (e.g., robot vs. human assistance) for participants to consider and compare the difference. Only one study addressed this point, but it had a small sample of 21 independent living older adults and was conducted a decade ago (Smarr et al., 2014). The small sample size and homogeneous nature of the sample limited its ability to generalize findings. Moreover, the previous findings may not fully reflect the present situation because older adults’ preference for assistance may have changed due to the rapid advancement in robotic technology and increased technological capability in the older population. Thus, a recent study that addresses previous studies’ limitations is indicated. The aims of this study are: (1) to describe older adults’ preferences for assistance in performing domestic tasks; and (2) to identify the domestic tasks that older adults prefer robots to perform and the domestic tasks they prefer humans to perform.
Methods
This study utilized data from the baseline findings of a larger cross-sectional descriptive study investigating the preferences for domestic assistance from robots in older adults with different characteristics. Another paper based on this major study, but with a different research focus, has been published (L. Y. K. Lee et al., 2024).
Population and Sample
This study was conducted in the older population who were living in a domestic household in Hong Kong. This cohort of older adults are aging in their own place. However, it does not mean they must have adequate domestic support. Among Hong Kong’s older residents, 13% lived alone and 27.3% lived solely with their spouse, who might be of old age as well (Census and Statistics Department, 2023). This phenomenon can be attributed by factors such as small housing apartments that cannot accommodate large families and desire for nuclear families among the young generation (L. Y. K. Lee et al., 2023). Although statistical data also showed that 49.9% of older adults were living with children (Census and Statistics Department, 2023), their children may not always contribute to domestic duties because of long working hours and high demands in the job market. In this situation, regarding the division of responsibilities in Chinese families, older parents typically manage daily household chores. They often experience challenges in daily living and are well-positioned to identify the types of assistance that could help them continue living in their homes.
This study recruited 385 older adults through convenience sampling. Inclusion criteria were (1) be 65 years old or above; (2) living in a domestic household; and (3) without any experience in using robots. Exclusion criteria included (1) unable to communicate in Cantonese (the major language in Hong Kong) and (2) has been diagnosed with cognitive impairment. Participants were recruited in public areas in the three major regions in Hong Kong (i.e., Hong Kong Island, Kowloon, and New Territories) from 9 am to 9 pm and on weekdays and weekends. This practice increased the chance of obtaining a heterogeneous sample in non-probability sampling. According to the Cochran Formula, a sample size of 385 was adequate to achieve 95% confidence and a 5% margin of error for a local population of 1,417,900 older adults aged 65 years or above (Census and Statistics Department, 2021).
Data Collection
The Assistance Preference Checklist was used to assess participants’ preferences for assistance in performing domestic tasks (L. Y. K. Lee et al., 2022). This checklist consists of 48 domestic task items under six categories, including personal care, leisure activities, health, chores, information management, and manipulating objects. These daily living tasks are considered by older adults essential to maintain their homes and fulfill their general needs. Participants were asked to imagine if they required assistance with various domestic tasks and if a robot could perform the task at a human level. Participants were instructed to indicate their preference for assistance on a five-point Likert scale (1 = only a human, 2 = prefer a human, 3 = no preference, 4 = prefer a robot, 5 = only a robot; Smarr et al., 2014). Permission has been granted by the author to use the checklist.
The checklist was originally developed in English and translated into Chinese, where the psychometric properties of the Chinese version were tested (L. Y. K. Lee et al., 2022). The Chinese version demonstrates good content validity with an index of 1.00 at the scale and item levels. It demonstrates satisfactory internal consistency with a Cronbach’s alpha of .95 for the whole scale. The Cronbach’s alpha values for the sub-scales were .84 (personal care), .62 (leisure activities), .59 (health), .93 (chores), .72 (information management), and .89 (manipulating objects). Several questions were included to collect data on participants’ personal characteristics (age, gender, marital status, functional dependency level).
Researchers approached participants in public areas and invited them to join the research. Upon obtaining participants’ consent, the researchers conducted face-to-face interviews and collected data at location where they engaged with the participants.
Procedures and Ethical Approval
The researchers approached older adults and invited them to join the research. Information on the objectives and nature of the research was presented to the older adults. They were informed of their involvement and assured of their rights as a research participant. All participants gave informed verbal consent. This research received ethical approval from the School of Nursing and Health Studies of Hong Kong Metropolitan University (Ref: 2021/04). The School has changed its name to the School of Nursing and Health Sciences in 2024.
Data Analysis
Participants’ personal characteristics were described using descriptive statistics. Their preferences were presented in mean and 95% confidence interval. One sample t-test was used to determine whether participants had a significant preference for robot or human assistance. The mean of each domestic task was compared again to “3,” which indicated no preference. A p-value of below .05 was considered statistically significant.
Results
Participants’ Characteristics
A total of 385 participants were recruited, and their personal characteristics are summarized in Table 1. Their average age was 74.5. Among them, 54.0% were male, 57.9% were married, and 51.2% were functionally independent.
Personal Characteristics (n = 385).
Preferences for Human Assistance Versus Robot Assistance
Participants’ preferences for assistance on individual domestic tasks, in terms of mean and 95% confidence interval, are indicated by the error bars in Figure 1. Within each domestic category, domestic tasks are sequenced from the highest mean preference for human assistance to the highest mean preference for robot assistance. Participants reported the highest mean preference for human assistance in most domestic tasks under personal care, followed by the domestic tasks under leisure activities and health. In contrast, they reported the highest mean preference for robot assistance in domestic tasks under manipulating objects, followed by the domestic tasks under information management and chores.
Preference for human versus robot assistance by domestic tasks (error bars represent a 95% confidence interval of the mean).
One sample t-test confirmed that participants’ preferences for human assistance in 13 domestic tasks were significantly different compared to no preference (no preference = 3; Table 2). The mean ranged from 2.42 to 2.86. Overall, participants preferred humans to assist in domestic tasks, mostly in personal care (e.g., feeding, shaving, and bathing), leisure activities (e.g., being entertained, calling family/friends, and entertaining guests), and health (e.g., deciding what medication to take, taking medicine, and exercising).
Domestic Tasks for Which Older Adults Preferred Receiving Assistance From Humans Compared to No Preference (n = 385).
Probability of type I error for a one sample t-test comparing each task mean to no preference (=3).
p < .01, ***p < .001.
One sample t-test also confirmed that participants’ preference for robot assistance in 33 other domestic tasks were significantly different compared to no preference (no preference = 3; Table 3), with a mean range from 3.25 to 3.80. Participants preferred robots to assist in domestic tasks, mostly in chores (e.g., cleaning windows, sweeping/scrubbing/mopping, and cleaning kitchen), information management (e.g., monitoring home or warning about dangers, being reminded of appointments, and getting information on weather or news), and manipulating objects (e.g., picking up or moving heavy objects, fetching objects from the floor or other room, and finding or delivering items).
Domestic Tasks for Which Older Adults Preferred Receiving Assistance From Robots Compared to No Preference (n = 385).
Probability of type I error for a one sample t-test comparing each task mean to no preference (=3).
p < .05, ***p < .0.
Discussion
This study compares older adults’ preferences for domestic assistance between humans and robots. Such practice enables a direct comparison between different options, which is uncommon in the literature. Moreover, this study has the strength of recruiting a sample with an adequate sample size, which is essential for making statistical inferences. We compare our findings with those of Smarr and her team in 2014 because it is one of the few studies investigating older adults’ preferences for domestic assistance on a comprehensive list of domestic tasks (Smarr et al., 2014). Moreover, the two studies adopted the same data collection tool. We also refer to the findings of other studies to support our arguments.
The authors acknowledge that some of the cited papers are older, which is notable for a topic related to technology. We found it challenging to locate recent studies on preferences for domestic assistance. This may be due to limited research activity in this specific area. This study aims to fill this knowledge gap by investigating older adults’ preferences in recent years, thereby providing an opportunity for comparison with earlier findings. Since the findings of this study demonstrated difference with those studies published a decade ago, this study provides new insights and perspectives that are not addressed in the earlier literature. We believe this contribution is significant in advancing understanding in the field.
Domestic Tasks That Older Adults Preferred to be Assisted by Humans
Participants reported a higher preference for humans to assist in most tasks related to personal care and leisure activities. This finding is understandable because robots are perceived as least useful for highly personal tasks that involve extensive human-robot interaction (Broadbent et al., 2012; Ezer et al., 2009; Mast et al., 2012; Smarr et al., 2014).
In the health category, deciding what medication to take, taking medicine, and exercising were domestic tasks that older adults had a significant preference for human over robot assistance. In particular, deciding what medication to take moved from the ninth to the first preferred domestic task to be assisted by humans, compared with the previous findings by Smarr and her team in 2014 (Smarr et al., 2014). Humans are perceived to be more reliable than robots in making health decisions because robots are not liable for the adverse consequences of any inappropriate decision that they make. Similarly, another study indicated that older adults seek human effort to make medical decisions (Pasquale, 2020). In this study, taking medicine and exercising were ranked as the 12th and 13th domestic tasks significantly preferred to be assisted by humans. They were not found in Smarr et al.’s list, neither for human nor robot assistance (Smarr et al., 2014).
The acknowledgment of these three domestic tasks for human assistance and the advancement of deciding what medication to take from the ninth to the first rank for human assistance suggest that older adults are now more health conscious and desire humans to support them in health-related tasks. Hong Kong’s population is well known for its long life expectancy, with 82.9 years for males and 88.0 years for females in 2020 (Center for Health Protection, 2021). Having more concern for health might contribute to the longevity of the population. Moreover, the present findings reflect the preferences of a heterogeneous sample of older adults regarding functional dependency. In contrast to Smarr et al.’s independent living sample (Smarr et al., 2014), the present sample contained 48.9% of the participants who were either partially or wholly dependent. Their functional dependency puts them in actual need of a reliable assistant to assist in day-to-day health issues, which are of paramount importance. As indicated by Hong Kong adults in earlier studies, having a stable health condition is essential to positively impact all other aspects of aging (L. Y. K. Lee & Fan, 2008, 2014). Findings highlight the significant role of human manpower in delivering health-related domestic care, which is less likely to be replaced by robots.
In the personal care category, eating/feeding, shaving, bathing, washing/combing hair, and brushing teeth were six out of seven domestic tasks under this category significantly preferred to be assisted by humans over robots. These tasks are characterized as highly personal. Physical contact and interactions between the caregiver and care recipient are involved. The findings are expected because humans can be gentler and more flexible than robots in delivering personal care (Mast et al., 2012). Using feeding as an example, older adults are overwhelmed by the large size of the robot and feel threatened when the thick arm of the robot approaches them during feeding (Park et al., 2019). Additionally, older adults perceive that feeding includes human actions such as tilting and retracting with a spoon (Park et al., 2019). They likely prefer having human assistance rather than depending on robots for this activity.
In the leisure activities category, being entertained, calling family/friends, and entertaining guests were domestic tasks under significantly preferred to be assisted by humans over robots. Performing these tasks requires instance interaction and good communication. Humans are perceived to be more competent in developing constructive and meaningful communication with older adults. In contrast, robots are often perceived as less competent when it comes to real-life communication, particularly for tasks requiring emotional, perceptual, and cognitive abilities. Thus, older adults prefer humans over robots in assisting the abovementioned tasks related to leisure activities (K. Lee et al., 2008). Another explanation is that older adults perceive a robot as an assistant rather than a friend. Leisure activities expected to be performed with a friend are not expected to be performed with an assistant (Smarr et al., 2014).
In the chores category, preparing meals/cooking was the only domestic task under this category that was preferred to be assisted by humans over robots. Older adults commonly have individual preferences for meal preparation. They prefer certain food to be cooked in a certain way (L. Y. Lee et al., 2024). They also prefer variations in the meals. As robots are programed to perform in a standardized way and may not be good at meeting individual preferences, robots are less preferred to assist in preparing meals or cooking.
Domestic Tasks That Older Adults Preferred to be Assisted by Robots
Participants reported a higher preference for robots to assist in most tasks related to chores, manipulating objects, and information management. The present findings concur with other works in the US (Ezer et al., 2009; Olatunji et al., 2024; Smarr et al., 2014), New Zealand (Broadbent et al., 2012), and France (Wu et al., 2011). These studies revealed that robots are preferred when carrying out functional tasks with little human-robot interaction (Ezer et al., 2009; Olatunji et al., 2024).
In the chores category, findings indicated that robot assistance was significantly preferred for 20 out of 21 domestic tasks. Chores are characterized as boring or even unpleasant but need to be done regularly. Older adults can be relieved from performing the chores if robots are available. While maintaining lawn or raking leaves was indicated as the rank one task to be assisted by robots (mean = 3.81) in independent older adults in the US (Smarr et al., 2014), this task was only ranked at 21 (mean = 3.50) by the older adults in this study. Comparison of findings between studies suggests that differences in living environment and practice between countries/regions can influence older adults’ preferences. Hong Kong is well known for its small average living space per person. In 2022, the average living space per person was 13.6 m2 (Housing Bureau, 2022). Older adults in Hong Kong commonly live in small flats or rooms, which are not conducive to growing large plants. Thus, maintaining lawns or raking leaves is not a common practice.
In the information management category, robot assistance was significantly preferred for tasks such as monitoring the home/warning about dangers, getting information on weather/news, being reminded of appointments, and being reminded of daily activities. These tasks were ranked at the seventh, 16th, 19th, and 23rd positions in the list, respectively. Robots have the strength to provide up-to-date, accurate, and comprehensive information (Iio et al., 2020). Additionally, they are perceived as effective at sending accurate reminders based on a pre-set schedule (Suwa et al., 2020) and alarming for unusual situations (Wu et al., 2011). They can be programed to perform tasks in a timely and accurate way. Human errors can be avoided. With these strengths, robots can manage information, provide instant and accurate reminders, and alert older adults whenever necessary.
In the manipulating objects category, picking up/moving heavy objects received the highest mean score and was ranked as the most preferred domestic task to be assisted by robots. Other domestic tasks in this category, including fetching objects from the floor or other room, finding/delivering items, opening and closing doors/drawers, and reaching for objects, also obtained high mean scores. These tasks are mostly mechanical and possibly physically demanding. Robots have the advantage of having more strength and durability in performing mechanical and physically demanding tasks (Suwa et al., 2020). With the assistance of robots, a more efficient and safer outcome can be achieved (Gould, 2019). Findings coincided with previous ones (Smarr et al., 2014).
Older Adults Are Becoming More Open to Robot Assistance
Older adults’ preferences change over time, suggesting that older adults are now becoming more open to robot assistance. This conclusion is drawn from several observations from the study findings. While the mean of preference for human assistance in those domestic tasks in the previous study ranged from 1.90 to 2.18 (Smarr et al., 2014), the mean of preference for human assistance in those domestic tasks in this study ranged from 2.42 to 2.86 (Table 2). A higher mean value implies a lesser preference for human assistance.
While Smarr and her team revealed 15 domestic tasks reported a significant preference for robot assistance (Smarr et al., 2014), this study revealed that 33 domestic tasks obtained such findings. The additional 18 domestic tasks belong to various domestic categories. Eleven domestic tasks belong to the chores category (i.e., doing laundry, taking our trash/recyclables, painting, repairing plumbing, keeping the refrigerator clean/stocked, sorting mail, shredding, throwing away junk mail, watering plants, grocery shopping, washing dishes by hand, gardening/pruning, and setting the table), two domestic tasks belong to the information management category (i.e., monitoring home/warning about dangers and getting information on weather/news), two domestic tasks belong to the health category (i.e., being reminded to take medicine, calling doctors/911), one domestic task belong to the personal care category (i.e., walking), one domestic task belongs to the leisure activities category (i.e., getting information on hobbies/topics of interest), and one domestic task belong to the manipulating objects category (i.e., opening and closing doors/drawers). These changes, which happened within a decade, suggest that older adults are becoming more open to robot assistance.
Several possible reasons are identified. The rapid advancement of technology may positively influence the preference for robots for domestic assistance. Over the past decade, computer technology and digital devices have become more common in the older population. Increased exposure to technology, such as smartphones, tablets, and smart devices, has made older adults more comfortable with choosing technology for assistance. Older adults have probably benefited from computer technology and consider it a possible way to improve their daily living (Mitzner et al., 2008).
At the same time, advancements in robot technology are being made. New robots are being introduced to the market and becoming more capable and user-friendly (Oxford Economics, 2019). The media have highlighted their positive and success stories. Users also report positive experiences of living with robot (Gasteiger et al., 2022). These testimonials may contribute to older adults’ increased openness to embracing robot assistance.
The cost of robot technology has decreased over time, making it more accessible to a wider range of individuals, including older adults. Countries or regions have set up or are encouraged by policymakers to set up funding to support older adults using gerontechnology (Genge et al., 2023; Mitseva et al., 2012). When the affordability of robots increases, using robots for assistance becomes a viable option.
Lastly, the population has been aging in many areas of the world, and the shortage of caregivers is a global challenge. When the supply of caregivers cannot meet the demand, older adults are more likely to accept robots as an alternative (L. Y. K. Lee et al., 2024). Although this study was conducted in an older Chinese population that expects to receive filial care from their children, discrepancies usually happen between expectation and reality. Due to practical considerations, older adults becoming increasingly aware that they cannot expect care from their children. They, especially the young-old individuals, consider assistive technologies to help them manage daily living and remain in the community as long as possible (Bai et al., 2020).
Methodological Limitations and Research Recommendations
This study recruited older adults who had no experience with robots to reduce the potential influence of prior robotic experience on their preferences. Literature indicates that older adults’ preference for robot assistance can be related to their experience of what robots can and cannot do (Broadbent et al., 2009; Smarr et al., 2014). Older adults with more experience with robots may perceive domestic robots as easier to use and more useful than those with less robotic experience (Ezer et al., 2009). However, older adults without experience with robots may provide responses influenced by their individual perceptions, lack of experience, or misconceptions (Broadbent et al., 2009). We initially looked at the views of older adults who have no experience with robots. Our findings concur with other work which showed that older adults with little or no exposure to robots were generally open to robot assistance (Olatunji et al., 2024). Future studies should consider the views of older adults who do have experience with robots in order to obtain a comprehensive understanding.
While the Assistant Preference Checklist includes a comprehensive list of domestic tasks, there is no evidence to support that all tasks can currently performed by robots. This study aims to identify older adults’ preferences, which can inform robotic developers about the most desired functionalities and priorities. Our ultimate aim is to influence the direction of robotic development in accordance with older adults’ preferences, paving the way for advancements in the field. Future studies are suggested to evaluate the feasibility and effectiveness of using robots to carry out the domestic tasks that older adults prefer to be assisted by robots.
In addition, this study was limited by assuming preference equates to actual use. It is aware that actual adoption and continuous use of robots do not solely depend on preference. Without considering factors such as facilitators and barriers to use, design of robots, support system, users’ experience, and impact on daily living, the present study might result in incomplete insights into why older adults use or do not use robots. A more comprehensive investigation could inform effective recommendations for robot design and implementation. Additionally, it would be desirable to test the robots in the actual places where they will be used (Gasteiger et al., 2022).
Lastly, this study considered older adults in Hong Kong as a homogeneous group. However, their needs likely differ based on various personal characteristics such as functional ability, knowledge about robots, living arrangements, and availability of support systems. For example, older adults living alone may lack the necessary support systems for successful adoption of robotic technologies, which could be provided by family or younger generations. A related study by our research group compared the difference in preferences between older adults with different functional dependence levels and highlighted significant difference in preferences between groups (L. Y. K. Lee et al., 2024). Recognizing the diversity within the older adult population, future research should focus on comparing preferences among older adults with a specific or combinations of personal characteristics, as well as exploring how specific personal characteristics affect the acceptance, and use of domestic robots. In addition, longitudinal studies could be valuable to provide insights into how these preferences evolve over time. Although Chinese culture emphasizes respect for one’s parent and parents typically expect care from their children or family caregivers (De Mente, 2009), the contextual situation in Hong Kong, as described in the study sample, is not conducive to family caregiving. Mixed-method studies could deepen our understanding of the cultural and contextual factors influencing adoption of domestic robots in daily living among older adults.
Recommendations for Robotics Development and Domestic Care
Our findings can inform future robotics development and domestic care. Following the findings that older adults are now more open to robot assistance, this study strongly recommends introducing robots in domestic care to reference recent findings and cover more domestic tasks. Moreover, preference should be evaluated immediately following the introduction of robots for instant feedback and frequently over an extended period (i.e., within a few years) following actual interactions with robots because of potential changes in preferences and declines in capabilities of older adults. When the robots are better developed and older adults have more confidence in them, they can be used to assist in more domestic tasks or become all-around domestic helpers.
Despite presenting an overall pattern in preferences for assistance in the six domestic categories, findings also revealed different preferences for assistance in a few domestic tasks within the same category. Given that older adults can discriminate their preference, assessing their preference for assistance in individual domestic tasks before the introduction of robot in domestic settings is essential.
Consistently, some domestic tasks such as feeding and bathing are preferred to be assisted by humans. It is understood that humans have strengths over robots, especially in carrying out tasks that involve touch and instant interaction. Combining human and robot efforts may be a desirable model for assisting these domestic tasks to reduce the demand for human effort and improve efficiency (L. Y. K. Lee et al., 2024). It would be desirable to explore the various challenges associated with integrating robots into models of care. This could include deployment, feasibility, usability, acceptance, and the impact on caregiving dynamics. Efforts are also required to guide development of ethical frameworks regarding the responsibility and accountability of human in conjunction with robots in domestic support, rights of vulnerable groups, equitable arrangement in various contexts, fairness in considering the views of various stakeholders, alignment with societal values, and supporting sustainable and socially responsible practices. The ultimate aim is to create responsible robots and guide responsible behavior that benefit the older adults and the society in the long run.
This study investigated older adults’ preferences for robot assistance on domestic tasks that require a single and lower level of capability (e.g., washing dishes or doing laundry). It is a good idea to investigate older adults’ preferences for robot assistance on domestic tasks requiring more and even at a higher level of capabilities (e.g., organizing an apartment). Those findings will certainly complement the present understanding.
Older adults are willing to engage in issues that matter to them. Recognizing the importance of addressing their needs and preferences, co-designing robots with older adults is a valuable practice (Gasteiger et al., 2022). This participatory approach can help bridge the gap between technology development and real-world application.
Lastly, given that robot assistance is preferred for domestic support, devoting resources to support the introduction of robots in domestic care through relevant policies and proactive planning should be accomplished in a timely manner.
Conclusions
Appropriate assistance is essential to actualizing aging in place (World Health Organization, 2004). Efforts have been made to introduce robots to assist older adults in performing daily activities in a home setting, compensate for their age-related declines, and support them to age in place. Social robots interact with individuals and help with day-to-day activities. They can be further classified into companion robots, tele-operated robots, and semi-autonomous robots, each designed to perform specific functions and meet the distinct needs of older adults. Assessing older adults’ preference toward domestic assistance is essential for developing and selecting appropriate robots. Older adults prefer humans to assist in highly personal and flexible domestic tasks that require instant interaction and communication. In contrast, they prefer robots to assist in domestic tasks that are standardized, mechanical, physically demanding, and information-related. Older adults’ responses toward receiving robot assistance are much more favorable than previous findings from a US study in 2014 and other studies in the field. Findings suggest that older adults are now more open to robot assistance in domestic tasks. They prefer robot assistance for more domestic tasks in various domestic categories. This study provides new insight into the current practice of using robots to strengthen domestic support. Thus, when considering robot assistance in domestic tasks, it should reference recent findings and can be applied in many domestic tasks in different categories. Frequent assessment of older adults’ preferences is necessary because they are prone to change. While older adults consistently prefer humans to assist in certain domestic tasks, future research is suggested to develop a model which combine human and robot efforts. Also, it would be desirable for future research to explore the various challenges and ethical issues associated with integrating robotic support into the home environments. Lastly, co-designing robots with older adults would close the gap between technology development and real-world use.
Footnotes
Acknowledgements
The authors thank the participants for their participation. Deep appreciation is also extended to Prof. Cory-Ann Smarr for granting permission to use the Assistance Preference Checklist.
Ethical Approval
This research obtained ethical approval from the School of Nursing and Health Studies of Hong Kong Metropolitan University (Ref: 2021/04). The School has changed its name to the School of Nursing and Health Sciences in 2024.
Funding
The author(s) received no financial support for the research, authorship, and/or publication of this article.
Declaration of Conflicting Interests
The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.