POLSKI
Introduction
The aging population and the resulting dynamic increase in the number of people requiring long-term care represent one of the most serious challenges facing modern healthcare systems. It is estimated that by 2050, the number of people over 80 years of age will triple globally, significantly increasing the demand for care and nursing staff [1]. This situation is exacerbated by the shortage of qualified staff, nurses’ fatigue, and the growing number of patients with dementia who require individualized, patient, and often 24-hour care [2].
In the search for innovative solutions, social robotics is gaining increasing interest – especially the use of therapeutic and humanoid robots to support the daily care of the elderly. Particular attention is being paid to the Paro robot – a biomimetic seal equipped with tactile sensors and responding to sound, movement, and touch – and the Pepper humanoid robot, programmed to conduct conversations, recognize emotions, and support group activities [3–5].
Numerous studies indicate that interactions with robots such as Paro can contribute to improved mood, reduced symptoms of depression and anxiety, reduced agitation, and reduced aggression in people with dementia [6–10]. Social robots can also reduce the need for psychotropic medications and support the implementation of non-pharmacological therapies in care facilities [11 13]. Randomized studies have confirmed that short-term sessions with the Paro robot lead to a reduction in neuropsychiatric symptoms (BPSD) and improved quality of life in patients [8,14–16]. In turn, the Pepper robot works well as a moderator of active activities, supporting communication, physical exercise, and cognitive stimulation [17].
The perception of nursing staff is also important, as they increasingly see robots not only as tools that support therapy, but also as factors that facilitate establishing relationships with patients, building trust, and reducing the emotional burden associated with work [11,18,19]. At the same time, ethical, technological, and organizational questions arise: how can we ensure the safety of human-robot interactions? To what extent can robots replace human contact? Does implementing technology threaten to depersonalize care?
In light of these considerations, the aim of this article is to synthetically analyze the results of scientific research on the use of humanoid support robots in long-term care, with particular emphasis on their role in geriatric nursing.
This article is based on an analysis of scientific publications on the use of social robots – in particular the humanoid robot Pepper and the biomimetic therapeutic robot Paro – in long-term care for the elderly. The review is descriptive in nature.- synthetic and includes empirical studies published in peer-reviewed scientific journals between 2018 and 2024. The analyzed publications were selected based on their relevance to the topic and current relevance and contained data from studies conducted in nursing homes, day care centers, geriatric wards, and hospital facilities where people with dementia or other dementia syndromes resided.
The review included randomized controlled trials, pilot studies, and qualitative analyses. These studies varied in methodology, sample size, intervention length, and measurement tools, making quantitative meta-analysis impossible but allowing for consistent qualitative conclusions.
The selection of articles also took into account geographical diversity – the analysed studies were conducted in Australia, Japan, Canada, South Korea, Italy and Western European countries – which allowed for a broader perspective on the adaptation of technology in various healthcare systems. The work was both quantitative (e.g. assessment of psychological variables using standardized scales such as CSDD, NPI, GDS) and qualitative (semi-structured interviews, narrative analysis, interaction observations).
The analysis focused on five thematic areas: the impact of interactions with robots on emotions, mood, and depression; reduction of neuropsychiatric symptoms of dementia (BPSD); cognitive and social activation of older adults; acceptance and perception of robots by patients and staff; and identification of ethical and organizational challenges related to implementing robotics in care. This approach allowed for the capture of both the direct therapeutic effects and the broader organizational and cultural context in which robotic interventions operate.
The impact of interaction with robots on emotions, mood and depression
One of the most frequently researched and most promising areas of application of social robots in long-term care is their impact on the emotions and mental well-being of older adults, particularly those with dementia. Numerous studies conducted in nursing homes, geriatric wards, and day therapy centers indicate that regular interaction with a social robot—especially Paro, a biomimetic seal designed to support emotional connections—can contribute to significant mood improvement, reduce symptoms of depression, and lower levels of anxiety and agitation.
A randomized controlled trial by Pu et al. involving 415 people with dementia showed that daily Paro sessions for 10 weeks led to a significant reduction in depression levels as measured by the Cornell Scale for Depression in Dementia (CSDD). The intervention group also demonstrated a reduction in the use of psychotropic medications, including benzodiazepines and neuroleptics, compared to the group receiving standard care [13]. These results were also confirmed in the study by Bevilacqua et al., where a combined intervention with Paro and cognitive training not only improved the participants’ quality of life but also led to a reduction in anxiety and depressive symptoms [8].
The strong emotional impact of the Paro robot is also confirmed by qualitative research. In the study by Pu et al., participants described their experiences with the robot as „calming” and „comforting” – for many of them, Paro was a „companion” or even a „friend” who alleviated feelings of loneliness and anxiety [6]. Similar effects were observed by Hung et al. in a study conducted in a geriatric hospital – participants with dementia expressed positive emotions, spoke to the robot as if it were a living being, and its presence evoked memories and improved mood even in people with advanced cognitive impairment [20].
Quantitative data on mood improvement following interaction with the robot are also confirmed by Law et al., who conducted a study on adults with psoriasis, demonstrating that contact with Paro after a stressful task led to stress reduction, improved positive emotions, and decreased negative emotions [10]. Although this study focused on the general population, its results indicate the broad potential of the therapeutic robot as a tool for reducing emotional tension in high-risk groups.
Systematic reviews also confirm these observations. Leng and colleagues, in a meta-analysis of six randomized controlled trials in people with dementia, found that Paro statistically significantly reduces depressive and apathetic symptoms and has a positive effect on the overall mood of patients [7]. In turn, the review by Wang et al., which included only randomized controlled trials, emphasized that the greatest therapeutic effects in the area of emotions were achieved with individual, repeated interventions in which the robot was used for at least 15 minutes two or three times a week [16].
It is also worth emphasizing that social robots, in addition to directly influencing patient emotions, can indirectly support care by reducing stress-inducing behaviors in staff. Research by Jones et al. showed that nurses who observed improved mood and decreased arousal in residents after interactions with Paro reported greater job satisfaction and reduced mental strain [21].
Although most studies have focused on the Paro robot, a few studies have also assessed the performance of the humanoid robot Pepper. Carros et al. and Unbehaun et al. observed that this robot, when conducting group activities, quizzes, and physical activities, evoked joy, laughter, and liveliness in nursing home residents, and also led to greater social activity and interaction with other participants [4,22]. This effect, although more focused on social interaction than emotional connection, also contributed to improved mood and reduced negative emotions.
Data from numerous studies conducted in various countries and settings clearly indicate that interaction with social robots – particularly Paro – can be an effective, non-pharmacological intervention for improving the well-being of older adults with dementia. These robots support mental well-being, alleviate depressive symptoms, and constitute an important component of person centered care. Their effectiveness is particularly evident in settings where interpersonal contact is limited and the need for empathetic emotional support is high.
Reduction of neuropsychiatric symptoms of dementia (BPSD)
Accumulating scientific evidence indicates that the Paro robot can effectively support the reduction of BPSD symptoms. In a study conducted by Sato et al., interventions with Paro, implemented in patients with dementia and schizophrenia, led to calmness, reduced aggression, and improved patient-staff interactions. The presence of the robot promoted patient concentration and engagement in rehabilitation activities, while also reducing anxiety and emotional disorganization [15].
A study by Bevilacqua et al. indicates that Paro’s effectiveness in reducing neuropsychiatric symptoms is particularly pronounced in individuals with Alzheimer’s disease. In this group of patients, interactions with the robot most often led to reductions in anxiety, tension, and agitation. In contrast, individuals with Parkinson’s disease or multiple sclerosis showed milder responses, suggesting that its effectiveness depends on the patient’s neurological profile [8].
The results of the study presented by Kang et al. also confirm the importance of tailoring the Paro intervention to individual cognitive and emotional characteristics. In the analyzed groups, the robot contributed to reducing tension and behavioral symptoms, and also increased the sense of calm and relaxation in patients. Particularly significant changes were observed in the expression of positive emotions and a reduction in aggressive reactivity [23].
Additional support for the robot’s effectiveness in alleviating neuropsychiatric symptoms comes from physiological studies. Law and colleagues demonstrated that contact with Paro after a stressor resulted in a reduction of negative emotions and less transepidermal water loss, which is an indicator of improved emotional-physiological balance [23]. A study conducted by Geva et al.showed that only active interaction with the robot – reacting with movement and sound – led to a reduction in perceived pain and stress. This suggests that the effectiveness of Paro stems not only from its appearance, but primarily from its interactive communication capabilities [24].
Strong therapeutic effects of the robot were also noted in studies conducted in South Korea. Lee and colleagues conducted an intervention with elderly nursing home residents, which demonstrated a significant reduction in depression and stress levels. The reduction in emotional tension was visible after just four weeks of interaction with Paro, which may translate into a reduction in challenging behaviors, such as resistance to caregiving or shouting [25].
Also noteworthy is the report by Moyle et al., who examined the effect of the Paro robot on the responses of dementia patients during painful procedures. Participants demonstrated lower levels of anxiety and fewer pain symptoms in the presence of the robot, suggesting its potential to alleviate neurobehavioral responses associated with pain [26].
The collected results confirm that the Paro robot can be an effective tool supporting nurses in managing BPSD symptoms. Its operation contributes to reducing anxiety, agitation, and aggression, and also improves caregiver-patient relationships. Paro does not replace caregivers but rather empowers them, enabling individualized care and introducing an element of emotional safety in institutional settings.
Cognitive and social activation of older people
As dementia and brain aging progress, maintaining cognitive activity and stimulating social interactions becomes a growing challenge in caring for older adults. In this context, technology, and particularly social robots, are increasingly becoming a tool supporting cognitive and social activation.
The humanoid robot Pepper, designed for verbal interaction, conducting conversations and moderating group activities, is particularly useful in cognitive stimulation programs.
In a study by Unbehaun et al., this robot was used to facilitate classes with quizzes, word games, and memory training. Increased concentration levels, greater spontaneity, and improved collaborative atmosphere were observed among participants who had previously been reluctant to engage in the activities [22]. Similar conclusions were drawn from the work of Carros et al., where Pepper not only facilitated group activities but was also perceived as a „social animator”—facilitating barrier-breaking, enlivening the group, and encouraging collaborative action, even among those who were typically socially withdrawn [4].
In a study conducted by Blindheim et al. in a Norwegian nursing home, the impact of activities conducted by the Pepper robot on residents’ engagement in cognitive and motor activities was analyzed. The study found that the robot’s presence made participants more willing to participate in exercises, made more eye contact, and reported greater satisfaction with the activities. Thanks to its humanoid features and conversational capabilities, the robot broke the daily routine and motivated residents to be active [27].
Paro also performs similar functions, although in a slightly different dimension – emotional and sensory. Bechade and colleagues observed that contact with Paro can stimulate autobiographical memory processes. During interaction with the robot, participants talked about their pets, childhood, and family rituals, which activated cognitive resources and strengthened the identity of the elderly. This study also highlighted the value of humor and positive surprise as a form of activation – the robot was able to make participants laugh, thus provoking cognitive and social responses [17]. A systematic review by Wang et al., including randomized clinical trials, noted that although Paro does not clearly improve cognitive functions measured by neuropsychological scales, it may indirectly influence the maintenance of mental activity by encouraging contact, speech, and responsiveness to stimuli [16]. In a pilot study, Bevilacqua et al. combined cognitive therapy with interaction with Paro and observed improvements in executive functions and orientation. This comprehensive intervention had a positive impact not only on cognitive status but also on participants’ engagement in therapeutic activities [8].
Japanese studies have also observed that social robots can maintain daily rhythms and socially engage people with dementia. Miyagawa et al. tested the Pepper robot as a support tool for people with moderate dementia in the daytime environment. Participants were more willing to participate in activities led by the robot than by a human, and interactions with the robot built a sense of community and reduced tensions between group members [28].
Baisch and colleagues demonstrated that even in individuals initially reluctant to use technology, contact with a robot can stimulate reflection, arouse curiosity, and lead to cognitive activation— especially with longer exposure. Over time, participants became accustomed to the robot’s presence and began to show increasing interest in the interaction [12].
It is worth emphasizing that cognitive activation using robots can take various forms – from linguistic and memory stimulation, through supporting social orientation and navigation, to stimulating action through touch and sensory contact. The study by Inoue et al. shows that a robot can also be used as a supporting tool in crisis situations – for example, during episodes of disorientation or aggression – by simultaneously becoming a point of reference for the patient and a safe cognitive stimulus [11].
Both humanoid robots, such as Pepper, and biomimetic robots, such as Paro, have the potential to support the cognitive and social activities of older adults. Their use promotes communication, strengthens social contacts, and fosters engagement, which is an important element in slowing the progression of dementia and improving quality of life in long-term care.
Acceptance and perception of robots by patients and care staff
The effectiveness of social robots in long-term care for the elderly depends not only on their functionality and therapeutic impact, but also on how they are perceived by the users themselves – patients and nursing staff. from the patient’s perspective, the most frequently described experiences with robots— especially Paro—are feelings of warmth, peace, closeness, and „companionship.” In a study by Hung et al., people with dementia hospitalized in a psychiatric hospital perceived Paro as a „living creature” or „friend” who brought comfort in a stressful, institutionalized environment. Patients hugged the robot, spoke to it, and expressed emotions toward it, sometimes even attributing to it intentional qualities, such as the ability to understand or care [18].
Similar conclusions were drawn in a study by Pu et al., where participants described the robot as a source of comfort and emotional support. Respondents described Paro as a „good companion” and „someone who doesn’t judge,” which was particularly important for individuals with limited verbal communication skills. The robot served as a catalyst for emotions and memories, enabling expression that was often limited in the traditional caregiver-patient relationship [13].
It is worth noting the study by fattal et al., which analyzed the aesthetic and emotional perception of the humanoid robot Pepper. Participants residing in a care facility in france perceived the robot as pleasant, „young,” and arousing curiosity. Initial reactions of uncertainty and distance gave way to curiosity and joy, especially after several days of exposure. Pepper was treated more as a „present being” than as an object, which contributed to increased emotional and social engagement of the participants [29].
Interesting data were also provided by a study conduc by Baisch et al., which showed that older adults’ acceptance of the Paro robot increases with the duration of interaction. Although participants may initially demonstrate resistance or uncertainty, prolonged interaction leads to a change in attitude, often resulting in naming the robot, anthropomorphizing it, and treating it as a part of their everyday environment [12].
In a study by Inoue et al., analyzing the implementation of Paro in an acute care hospital setting, nurses noted that the robot could act as a mediator in crisis situations – reducing patients’ emotional stress, but also facilitating staff in establishing relationships with older adults. Importantly, Paro was perceived as a tool that builds empathy and understanding for the patient, allowing them to be viewed as a „person” rather than merely a clinical case [11].
However, not all reactions were unanimously positive. In Sato et al.’s study, in which the Pepper robot was used in the rehabilitation of people with dementia and schizophrenia, some patients experienced frustration when the robot did not respond adequately or asked the same questions. The robot was perceived as „nice but limited,” demonstrating the limits of the technology in building a complete relationship. Nevertheless, the overall reception of the intervention was positive, especially when staff acted as interpreters or mediators [15].
Nurses, as Inoue’s research shows, also found the presence of the robot helpful in de-escalating challenging behaviors. However, they also reported the need for training and defining hygiene procedures and equipment responsibility, which indicates the need to integrate the technology with existing care and teamwork systems [11].
Ultimately, as Lee et al.’s South Korean study shows, acceptance of Paro was high among both patients and caregivers. Participants rated the robot as pleasant, user-friendly, and supportive. High levels of positive emotion expressed toward the robot correlated with reduced stress levels and greater willingness to participate in therapeutic activities [25].
The perception and acceptance of social robots among older adults and healthcare professionals is complex but overwhelmingly positive. The way the technology is introduced, the support provided by staff, and the ability to personalize interactions are key factors. Robots like Paro and Pepper not only support patients emotionally but also facilitate relationship building for nurses, strengthen trust, and support the implementation of person-centered care.
Ethical and organizational challenges related to the implementation of robotics in care
Although the use of social robots in elderly care has gained recognition as an innovative way to support well-being and compensate for staff shortages, it is accompanied by significant ethical concerns and organizational challenges. On the one hand, this technology offers potential benefits in the areas of emotion, communication, and the reduction of neuropsychiatric symptoms; on the other, it raises questions about patient dignity, the limits of automation in care, and institutional accountability.
A study by Rashid et al. examined the impact of Paro not only on the well-being but also on the neurological activity of older adults. An increase in alpha and beta EEG wave power was recorded, which was considered a sign of relaxation and cognitive engagement. Although the effect was positive, the authors note that such studies, bordering on psychophysiology, can lead to instrumentalization of the older adult—treating them as an object of measurement, which may be controversial from the perspective of relationship-based nursing [30].
The work by Moyle and colleagues explored the dilemma of „taking away” a robot from a patient. Some participants, after several weeks of regular interactions, developed an emotional attachment to Paro. When the study ended, they felt sad or even angry about it being „taken away.” This raises the question of whether it is ethical to temporarily share a technology that evokes a bond similar to that established with a human or animal [26].
Miyagawa’s study in a Japanese nursing home also indicates that the very process of implementing a social robot – including defining disinfection procedures, safety, availability and responsibility for technical maintenance – requires extensive team consultations. Nurses reported the need to create guidelines for the “emotional management” of patients reacting to the robot with excessive attachment, as well as concerns about situations in which the robot would arouse fear or anxiety [28].
The analysis by Abdi et al. addressed the issue of unequal access to technology in healthcare facilities. The study demonstrated that robots can contribute to reducing the number of medical interventions, but only when their use is systematic and well-organized. In institutions with limited time and staff resources, a robot may become a symbol of „luxury,” available only to selected patients, which raises the issue of therapeutic inequality [31].
Organizational problems were also highlighted in the work by Geva et al. Comparing contact with the active and inactive Paro robot, the authors noted that only the active, interactive version of the robot produced measurable therapeutic effects. However, in real-world care settings, some facilities opt for simplified, non-interactive versions, or those lacking technical support, leading to frustration among staff and patients and potentially resulting in a loss of confidence in the entire intervention [32].
Another aspect concerns the linguistic and communication limitations of robots. A study published as the ICA protocol by Miyagawa noted that Pepper’s linguistic adaptation to the needs of people with hearing, perception, or speech impairments can be insufficient. There were instances in which the robot—despite its advanced technology— failed to recognize utterances, repeated errors, or generated unnatural messages. Such interactions can lead to patient irritation and undermine trust in the personnel associated with the system [28]. finally, the question of the robot’s identity and the boundary between assistance and illusion remains problematic. Bechade and colleagues point out that although robots can establish relationships, they are not conscious – and therefore lack intention, empathy in the human sense, or the capacity for deeper moral responsibility. This creates a tension between patient expectations and technological reality, requiring staff to be particularly vigilant to avoid crossing the line between care and the staging of care [17].
Implementing social robots in long-term care is a process that requires not only technological decisions but also ethical reflection, systemic institutional support, and appropriate staff training. Despite their therapeutic potential, robots are not „neutral” devices – their presence impacts relationships, values, and the structure of daily care.
Nurses, as the people closest to the patient, play a key role here, not only as operators of the technology, but also as its ethical guides.
Conclusions
Based on a review of scientific articles, it can be clearly concluded that social robots demonstrate significant utility in many areas of everyday nursing practice. Both emotionally and organizationally, they can provide a valuable complement to the care of older adults, especially those with dementia. Robots like Paro effectively support the reduction of neuropsychiatric symptoms of dementia (BPSD), such as aggression, anxiety, apathy, and agitation. Their operation contributes to improved mood and reduced pharmacotherapy use, which is consistent with current recommendations for non-pharmacological treatment. from a nursing perspective, the robot can be considered a tool supporting work with difficult patients, as well as a form of support in situations of emotional stress or disorientation. Both Paro and Pepper also serve an activating function, supporting cognitive processes, memory, and attention, and motivating patients to engage in social interactions. In caregiving environments, where patient withdrawal and passivity are often observed, the robot can act as a facilitator, conversation starter, or group activity guide. Such support can lead to increased engagement, improved quality of life, and a stronger patient caregiver relationship. from an organizational perspective, social robots can bring real benefits to nursing teams – they reduce the pressure of constant interpersonal contact, support the regulation of challenging behaviors, and offer the opportunity to temporarily relieve staff without sacrificing the value of care. However, it remains crucial to prepare institutions for the implementation of such technology, both in terms of infrastructure and team competencies. At the same time, significant ethical and organizational issues have emerged in the research. Questions about the boundaries of the human-robot relationship, the impact of technology on patient dignity, the risk of attachment, and unequal access to interventions require careful consideration. Social robots cannot replace human contact, but their well-targeted presence can enrich the care environment, especially where staffing and time resources are limited. The implications for nursing practice are clear: social robots can be an effective tool supporting care for older adults, provided they are implemented in an ethical, integrated, and responsible manner. In practice, this means developing standards for robot use in care facilities, incorporating the topic into nursing training programs, and developing staff digital competencies. Nurses should be active participants in the technology implementation process, acting not only as users but also as advocates for the quality and safety of care. The implementation of robotics does not replace care, but can enhance it – offering new resources that support both the patient and the nurse in everyday care activities.
