You are currently viewing a new version of our website. To view the old version click .
MDPIMDPI
Search all articles
Healthcare
Download PDF
  • Systematic Review
  • Open Access

20 February 2025

AI Applications to Reduce Loneliness Among Older Adults: A Systematic Review of Effectiveness and Technologies

,
,
and
1
Division of Computational and Data Sciences, McKelvey School of Engineering, Washington University in St. Louis, St. Louis, MO 63130, USA
2
Department of Surgery, Division of Public Health, Washington University in St. Louis, St. Louis, MO 63130, USA
3
School of Social Work, University of Michigan, Ann Arbor, MI 48109, USA
4
Silver School of Social Work, New York University, New York, NY 10003, USA
Healthcare2025, 13(5), 446;https://doi.org/10.3390/healthcare13050446
This article belongs to the Special Issue Quality of Life and Mental Health of People with Disabilities and Chronic Illnesses in the Digital Era
Version Notes
Order Reprints
Review Reports

Abstract

Background/Objectives: Loneliness among older adults is a prevalent issue, significantly impacting their quality of life and increasing the risk of physical and mental health complications. The application of artificial intelligence (AI) technologies in behavioral interventions offers a promising avenue to overcome challenges in designing and implementing interventions to reduce loneliness by enabling personalized and scalable solutions. This study systematically reviews the AI-enabled interventions in addressing loneliness among older adults, focusing on the effectiveness and underlying technologies used. Methods: A systematic search was conducted across eight electronic databases, including PubMed and Web of Science, for studies published up to 31 January 2024. Inclusion criteria were experimental studies involving AI applications to mitigate loneliness among adults aged 55 and older. Data on participant demographics, intervention characteristics, AI methodologies, and effectiveness outcomes were extracted and synthesized. Results: Nine studies were included, comprising six randomized controlled trials and three pre–post designs. The most frequently implemented AI technologies included speech recognition (n = 6) and emotion recognition and simulation (n = 5). Intervention types varied, with six studies employing social robots, two utilizing personal voice assistants, and one using a digital human facilitator. Six studies reported significant reductions in loneliness, particularly those utilizing social robots, which demonstrated emotional engagement and personalized interactions. Three studies reported non-significant effects, often due to shorter intervention durations or limited interaction frequencies. Conclusions: AI-driven interventions show promise in reducing loneliness among older adults. Future research should focus on long-term, culturally competent solutions that integrate quantitative and qualitative findings to optimize intervention design and scalability.

1. Introduction

Loneliness is a subjective feeling of being alone, arising from the gap between an individual’s desired and actual social connections [1]. It affects people of all ages, including older adults. According to the World Health Organization (WHO), 20–34% of older adults report loneliness in countries such as the United States and across Europe [2,3]. Strong evidence links loneliness to increased risks of physical and mental health problems in older adults, resulting in reduced quality of life and life expectancy [4,5,6,7]. The health impact of chronic loneliness is comparable to smoking 15 cigarettes a day [6]. Recognizing its profound public health implications, researchers and policymakers have intensified efforts to develop and test interventions to reduce loneliness and mitigate its effects [8].
Current strategies for reducing loneliness typically fall into four categories: (1) improving social skills, (2) enhancing social support, (3) increasing opportunities for social contact, and (4) addressing maladaptive social cognition [9]. However, these interventions have shown mixed effectiveness, often facing challenges in scalability due to limited resources and inconsistent outcomes across diverse populations [9,10]. These limitations highlight the need for innovative and optimized approaches to reduce loneliness among older adults more effectively.
Applying artificial intelligence (AI) technologies in behavioral interventions offers a promising avenue to overcome these challenges by enabling personalized and scalable solutions [11]. AI, first conceived in 1956, initially relied on symbolic AI, which used logical rules and representations to model human intelligence and problem-solving [12]. Over the decades, AI has evolved significantly, advancing into modern approaches such as machine learning (ML), which develops rules from training data [13], and deep learning (DL), a subset of ML that leverages artificial neural networks to model complex patterns in large-scale data [14,15]. Reinforcement learning (RL), another ML subdomain, integrates concepts from psychology and engineering to enable autonomous learning in dynamic environments [16,17]. Robotics, including human–computer interaction (HCI), further extends AI’s applications by addressing physical and emotional needs through interactive technologies [18]. Robotics focuses on designing and building machines that can perform tasks traditionally requiring human intervention [19], and HCI explores the interfaces and interactions between humans and computers to improve usability and accessibility [20]. These capabilities provide scalable solutions to support independent living and enhance the quality of life and mental health of individuals with disabilities and chronic illnesses, including older adults [21]. For this population, AI has enabled improved monitoring of social isolation, the development of companion robots tailored to emotional and social needs, and personalized interventions that foster engagement and psychological well-being [22]. Increasingly, these applications are being leveraged to address loneliness among older adults.
AI-based behavioral interventions take many forms, including social technologies, intergenerational programs, support groups, pet companions, recreational activities, psychological therapies, physical exercises, and assistive technologies [23,24,25,26]. For example, Badal et al. employed natural language processing (NLP) and machine learning models to predict loneliness in older adults using transcribed speech data with high sensitivity and accuracy [3]. Beyond prediction, AI can actively reduce loneliness through innovative tools. Examples include social robots that provide companionship, virtual assistants that simulate conversations and facilitate connections with loved ones, wearable devices that deliver timely interventions, and algorithms that enhance the personalization of interventions [27]. For instance, early social robots like Paro utilized symbolic AI to simulate human-like interactions. Over time, robots have evolved to incorporate more advanced AI capabilities, enabling a broader range of functions and enhanced adaptability to individual needs. These approaches effectively address common access barriers by eliminating the need for travel, offering 24/7 availability, and supporting multiple languages through NLP capabilities. They are generally more cost-effective and may reduce the stigma associated with seeking help from human providers.
Emerging innovations reported in the media further illustrate AI’s potential in reducing loneliness. For instance, an AI-powered board game developed by Johns Hopkins University students combines elements of NLP and interactive gameplay to engage older adults in retirement communities. Modeled after the board game “Guess Who?”, this prototype allows users to converse with an AI opponent while playing, fostering companionship and mental stimulation [28]. Advancements like these highlight the versatility of AI in addressing loneliness through novel, engaging, and scalable approaches.
This study presents a systematic review of AI applications for addressing loneliness among older adults. This review is timely, as AI is rapidly evolving and has the potential to enhance the personalization and scalability of interventions for loneliness among older adults. However, the lack of a comprehensive review leaves a critical gap in understanding how AI has been applied in this space. By mapping the current landscape, this study aims to identify gaps and provide actionable insights to guide future research and development in this promising field.

2. Materials and Methods

2.1. Review Protocol and Registration

This systematic review was not registered on a public platform. However, the review protocol was developed using the Cochrane Handbook for Systematic Reviews of Interventions as the methodological framework and reported following the Preferred Reporting Items for Systematic Review and Meta-Analysis Protocols (PRISMA-P) guidelines [29].

2.2. Study Selection Criteria

Studies meeting all the following criteria were included in the review: (1) Study designs—experimental studies (e.g., randomized controlled trials, pre–post interventions, and cross-over trials) that explicitly tested an AI-enabled intervention aimed at reducing loneliness; (2) Use of AI—The intervention incorporated AI techniques, including symbolic AI, machine learning (ML), deep learning (DL), or reinforcement learning (RL), to intervene in loneliness-related outcomes; (3) Study subjects—Adults aged 55 and older; (4) Outcomes—The primary or secondary goal was to reduce loneliness, including related constructs such as social isolation; (5) Article type—original, empirical, peer-reviewed journal publications; (6) Time window of search—studies published from the inception of electronic bibliographic databases to 31 January 2024; and (7) Language—articles written in English.
Studies were excluded from the review if they met any of the following criteria: (1) studies did not explicitly examine loneliness as an outcome; (2) AI techniques were mentioned but not actively integrated into the intervention (e.g., descriptive discussions of AI without implementation); (3) articles were written in a language other than English; and (4) letters, editorials, study or review protocols, case reports, review articles, or conference abstracts.

2.3. Search Strategy

A keyword search was performed in eight electronic bibliographic databases: PubMed/MEDLINE, EBSCO, Academic Search Complete, APA PsycArticles, APA PsycInfo, CINAHL Plus, Web of Science, and Cochrane Library. The search algorithm (Appendix A) includes terms related to AI (e.g., “machine learning”, “neural network”), loneliness (e.g., “social isolation”, “social disconnection”, “emotional isolation”), and older adults (e.g., “aged”, “seniors”, “elderly”) to identify relevant study titles and abstracts in the databases. Two co-authors independently screened titles and abstracts identified from the keyword search, retrieved potentially eligible articles, and evaluated their full texts. Cohen’s kappa (κ = 0.67) was used to assess the interrater agreement between two co-authors. Discrepancies were resolved through discussion. Additionally, relevant articles published after the database search were identified through manual review and included in the study to ensure up-to-date coverage of the topic.

2.4. Data Extraction and Synthesis

A standardized data extraction form was used to collect the following methodological and outcome variables from each included study: author(s), year of publication, country/region, study design, overall sample size, arm-specific sample sizes, participants’ age range, participants’ mean/median age, participants’ sex distribution, chronic conditions, AI methodology category, AI implementation fields, intervention type, intervention setting, intervention frequency, intervention duration, nature of intervention, outcome measures, other outcome measures, and intervention effectiveness (i.e., outcome-specific treatment effect estimates).

2.5. Study Quality Assessment

The Grading of Recommendations, Assessment, Development, and Evaluations (GRADE) framework was used to assess the quality of each study. This framework evaluates evidence based on factors such as risk of bias, imprecision, inconsistency, indirectness, and publication bias, assigning studies to one of four levels: very low, low, moderate, or high quality. Randomized controlled trials typically begin at high quality, while observational studies start at low quality. The evidence level is adjusted during evaluation based on the presence or absence of these factors, ensuring a transparent and systematic approach to assessing study quality.

3. Results

3.1. Identification of Studies

Figure 1 shows the PRISMA flow diagram. A total of 223 articles were identified through keyword and reference searches, including one article published after the database search. The articles were retrieved from the following databases: PubMed (102), Web of Science (43), Cochrane Library (16), and EBSCO (including Academic Search Complete, APA PsycArticles, APA PsycInfo, CINAHL Plus, and MEDLINE) (61). After removing duplicates, 148 unique articles underwent title and abstract screening, from which 108 were excluded. The full texts of the remaining 40 articles were reviewed against the study selection criteria. Of these, 31 articles were excluded due to not meeting the inclusion criteria for study design, outcomes, interventions, or article type. No articles were excluded due to the inability to access full text. Therefore, nine studies were included in the systematic review.
Figure 1. PRISMA flow diagram.

3.2. Characteristics of Study Participants

Table 1 summarizes the participants’ characteristics in the nine included studies. Studies were published between 2008 and 2024, with seven published after 2020. Participants were recruited from six countries/regions, including the US (n = 4), New Zealand (n = 2), Taiwan (n = 1), Korea (n = 1), England (n = 1), and Japan (n = 1). The mean and median sample sizes are 32 and 33 participants ranging from 15 to 64. Participants’ age ranged from 50 to 100 years. Males constituted 24.96% of participants on average. Chronic conditions among participants varied across studies, including mild depression, cognitive impairment or dementia, and physical disabilities. Some studies excluded participants with severe cognitive impairment or Alzheimer’s disease, severe loneliness, psychiatric conditions, or recent use of psychiatric drugs.
Table 1. Characteristics of the studies included in the review.

3.3. Characteristics of Interventions

Table 2 summarizes the intervention characteristics of the included studies. Two study designs were adopted: RCTs (n = 6) and single-arm pre–post studies (n = 3). Among the RCTs, four studies randomized participants to two arms (a treatment arm and a control arm), and two studies randomized participants to three arms (one treatment arm and two control arms). The AI methodologies used in these studies included Symbolic AI (n = 4), ML (n = 5), and DL (n = 5). AI implementation fields covered a wide range of technologies, such as speech recognition (n = 6), emotion recognition and simulation (n = 5), learning and adaptation (n = 4), text-to-speech (TTS) (n = 4), NLP (n = 4), computer vision (CV) (n = 3), vision recognition (n = 2), autonomous decision-making (n = 2), sound localization (n = 2), and sensory processing and response (n = 2). Intervention types varied, including social robots (n = 6), personal voice assistants (n = 2), and digital human facilitators (n = 1), which are virtual agents designed to simulate human-like appearance and interaction using advanced technologies such as speech synthesis, emotion simulation, and human-like gestures [35].
Table 2. Intervention characteristics of the studies included in the review.
Most interventions were conducted exclusively in long-term care facilities (LTCFs) (n = 6). The remaining studies were implemented in other settings, including senior centers, independent living facilities, and mixed settings such as the community, retirement communities, and nursing homes. The intensity of interventions ranged from 15 min per day to up to 18 h over two weeks, with overall durations spanning from one week to three months. The average duration of the interventions was 1.78 months. The arm-specific sample sizes varied, with the smallest group having 10 participants and the largest group having 33 participants.

3.4. Outcome Measures

Table 3 reports the nature of interventions, primary outcome measures, other outcome measures, and intervention effectiveness from the nine included studies. Most studies measured loneliness by the UCLA loneliness scale or its variants [39]. Other outcome measures include scales for attachment to pets, depression (e.g., GDS-SF), quality of life (e.g., WHO-QOL-OLD), cognitive function, perceived stress, psychological well-being, anthropomorphic interactions, and multidimensional scale of perceived social support.
Table 3. Intervention effectiveness of the studies included in the review.
Intervention effectiveness varied among the studies. Six studies [30,31,33,37,38] reported significant reductions in loneliness using various forms of the UCLA Loneliness Scale, particularly those utilizing social robots. Social robots, such as Paro [40] and PIO [34], stood out due to their ability to foster emotional engagement, provide companionship, and adapt to participants’ needs. For example, Lim observed significant improvements in loneliness and cognitive function with robot-led storytelling and gymnastics sessions [34]. Similarly, Robinson et al. highlighted the emotional benefits of interaction with Paro, noting marked loneliness reductions compared to the control group [37].
Three studies [32,35,36] reported non-significant results, possibly attributed to shorter intervention durations, smaller sample sizes, or limited interaction frequencies. Fields et al. used a participatory arts approach combined with social robots, which yielded only minor changes in loneliness, possibly due to the single-session intervention design [32]. Loveys et al. employed a digital human facilitator but observed no significant impact on loneliness, likely due to the brief one-week intervention duration [35]. Papadopoulos et al. investigated culturally competent AI, which integrates cultural knowledge bases to enable socially assistive robots to adapt their interactions to the cultural backgrounds, values, and preferences of users [36]. Despite the advanced tailoring of interactions, the study found no significant reductions in loneliness among older adults, underscoring the need for further refinement and personalization in AI-based interventions to address diverse cultural and individual needs effectively.

3.5. Study Quality Evaluation Results

We assessed the evidence/quality of the studies included in the review using the GRADE framework. As shown in Table 1, four studies were rated as “high” quality and the other five “moderate”. Two primary reasons for a “moderate” rating concern a non-randomized study design (pre–post study) and incomplete reporting of participant characteristics and outcomes. Specifically, the studies rated as “moderate” often lacked detailed reporting on mean age, standard deviation, and percentages of males. These gaps in reporting reduce confidence in the results due to potential biases and limited clarity in the study populations.

4. Discussion

This review evaluated the research landscape concerning the application of AI in interventions aimed at reducing loneliness in older adults. Among the nine studies included, six reported statistically significant reductions in loneliness, particularly those using social robots. These robots were noted for their ability to simulate human-like interactions, engage users emotionally, and provide consistent companionship. Some studies also suggested that personalized activities, such as storytelling or memory exercises, may have contributed to improved emotional well-being and reduced loneliness. In contrast, three studies reported non-significant effects, which may have been influenced by factors such as shorter intervention durations or less frequent interactions. For example, interventions lasting only a single session or a brief one-week period may not provide sufficient time for AI-assisted companionship to establish meaningful connections. While the review offers insights into potential patterns, it does not allow for definitive conclusions about why some interventions were more effective than others. Nonetheless, the findings suggest that interventions combining AI-assisted companionship with physical or cognitive activities and incorporating longer durations and frequent, meaningful interactions may hold promise for addressing loneliness in older adults.
Consistent with the literature, our findings support the benefits of social robots. Park et al. highlighted the potential of socially assistive robots such as Pepper, ElliQ, and Hyodol in enhancing emotional well-being and reducing loneliness among older adults [41]. These robots not only provided companionship but also facilitated recreational activities like games and simple conversations, which were particularly beneficial for older adults living alone [41]. Similarly, Shah et al. reviewed digital technology interventions (DTIs) such as videoconferencing tools, sensor-based systems, and social apps, finding mixed evidence regarding their long-term effectiveness [42]. While DTIs showed promise in addressing loneliness, the evidence for sustained effects was limited, particularly in studies with shorter durations and smaller sample sizes [42]. Hoang et al. also emphasized that interventions implemented in long-term care settings demonstrated substantial potential in reducing loneliness, though heterogeneity in intervention types and methodologies posed challenges to generalizability [23]. In addition, previous research on animatronic pets has shown their potential in reducing loneliness among older adults. Tkatch et al. found that animatronic pets not only decreased loneliness but also improved psychological outcomes such as resilience and optimism, particularly for individuals with limited ability to care for live pets [43]. These findings highlight the potential for AI-driven technologies to meet diverse needs while minimizing the challenges of traditional pet ownership.
Building upon the success of animatronic pets, humanoid robots represent a more advanced evolution in AI-driven companionship. These robots are designed to mimic human appearance and behavior, enabling more intuitive and engaging interactions [44]. Equipped with speech recognition, natural language processing, and emotion recognition capabilities, humanoid robots can engage in meaningful dialogues, perform tasks such as medication reminders or item retrieval, and provide emotional support by recognizing and responding to human emotions [45]. Their ability to offer personalized companionship makes them especially promising in addressing loneliness. Moreover, their decreasing costs—such as Tesla’s projected $20,000–$30,000 for Optimus or Unitree Robotics’ G1 at $16,000—are making them increasingly accessible [46]. As these technologies continue to advance, humanoid robots are expected to integrate seamlessly into older adults’ lives, further enhancing their potential as tools for fostering companionship and mitigating loneliness.
A critical aspect to consider is the underlying AI technologies employed in these interventions. Speech recognition was the most frequently used technology, enabling conversational engagement and emotional support. For example, Amazon Echo devices utilized speech recognition to facilitate daily interactions, addressing loneliness by providing companionship [33]. Text-to-speech was also employed in Amazon Echo devices, complementing NLP capabilities to create more natural conversational exchanges [33]. Computer vision (CV) was another key technology, aiding robots like PIO in recognizing facial expressions and body language to adapt their responses and foster emotional engagement [34]. Additionally, symbolic AI were integral to social robots like Paro, allowing them to simulate human-like interactions and provide therapeutic support [37]. These technologies collectively enhance the ability of AI interventions to address loneliness by tailoring interactions to individual needs and preferences, highlighting the importance of leveraging advanced AI capabilities in future developments. Moreover, Wu et al. explored the use of conversational agents equipped with advanced NLP and contextual understanding, demonstrating that these agents can facilitate meaningful dialogue with older adults and provide emotional support comparable to human interactions [47]. Similarly, a recent exploratory study found that ChatGPT-3.5 provided emotional support and companionship to older adults, with participants describing it as engaging and easy to use, suggesting its potential as a tool for mitigating loneliness [48]. This suggests that even non-robotic AI technologies may have a role in addressing loneliness.
Our findings echo those from qualitative studies not included in this review. For instance, a qualitative analysis of older adults’ experiences with virtual social interactions via robots during the COVID-19 pandemic found that participants valued the companionship and emotional connection provided by AI robots, describing them as “comforting” and “engaging [49]”. Similarly, Wang et al. highlighted the ability of AI robots to foster a sense of routine and purpose through daily interactions, particularly for those living in isolation [50]. These studies emphasize the importance of emotional engagement and context-specific adaptability in AI interventions. While current qualitative findings highlight the potential benefits of AI-driven companionship, they rely on self-reported experiences, which may introduce variability and subjectivity. Integrating more robust quantitative analyses, such as confidence intervals and statistical effect sizes, would help confirm their impact and generalizability. Existing studies have already demonstrated promise in specific contexts. For example, AI-driven interventions have shown particular promise during solitary leisure activities, such as reading or watching TV, where they can provide conversational engagement and emotional support [51,52]. Additionally, AI robots can guide users through relaxation exercises, cognitive games, or physical activities in therapeutic settings, fostering a sense of connection and well-being [53].
This review was conducted following Cochrane guidance to ensure methodological rigor and systematicity. However, one limitation of our review is the small sample sizes in included studies and the heterogeneity in the duration and intensity of the intervention, which may compromise our findings’ external validity. Another limitation is that our review did not include results from qualitative studies, such as thematic analyses of interview transcripts from participants after using AI robots. These insights could provide valuable context and a deeper understanding of participants’ experiences and perceptions. The overall low percentage of male study participants may also introduce a potential gender bias. Male participants may respond differently to AI interventions due to varying social behaviors, preferences for interaction styles, or different acceptance levels toward technology. This could influence the study outcomes and underscores the need for future research to explore gender-specific responses to AI-based loneliness interventions. Selection bias in the included studies may affect our results, as participants who agreed to enroll may be more open to social engagement [54]. Additionally, some included studies provided limited participant information, such as cognitive impairment status, which might affect the study outcomes. Moreover, studies that did not explicitly use the terms “artificial intelligence” or other AI-specific terminology were unlikely to have been picked up by our search algorithm. For instance, programs focused on conversational agents, chatbot-assisted interventions, and virtual reality may exist but may not have been captured due to the specificity of the search terms. While this review included two studies involving the Paro robot, other Paro-related studies may not have been captured for similar reasons or because they did not meet our inclusion criteria.
This systematic review underscores the promising role of AI in addressing loneliness among older adults, paving the way for further innovation in AI-driven interventions to improve social and emotional well-being. As AI technologies continue to evolve, it is essential to integrate findings from both quantitative and qualitative studies to develop interventions that are not only effective but also deeply personalized and contextually relevant. Future research should focus on scalable models, such as cloud-based platforms that can deliver AI-driven companionship to many users simultaneously, and culturally competent approaches, such as designing AI robots with region-specific languages and customs to resonate with diverse populations. Long-term solutions, including year-round interactive programs that combine AI interventions with human support, should also be explored to ensure that AI interventions are both equitable and impactful in reducing loneliness among older adults.

5. Conclusions

AI-driven interventions show promise in mitigating loneliness among older adults by enabling personalized, convenient, and cost-effective support. Social robots in particular—enhanced by emotional simulation, speech recognition, and adaptive capabilities—demonstrate the greatest potential for fostering meaningful engagement. However, further exploration using larger samples, longer intervention periods, and more frequent interactions is needed to refine efficacy and address the complex emotional needs of diverse older populations. Culturally tailored solutions and comprehensive evaluation methods that integrate qualitative and quantitative data may optimize engagement and outcomes. Ultimately, sustainable and scalable AI-based strategies, aligned with ethical and human-centered design principles, are critical for the long-term reduction of loneliness in older adults.

Author Contributions

Conceptualization, Y.Y. and R.A.; methodology, Y.Y. and R.A.; formal analysis, Y.Y. and C.W.; investigation, Y.Y.; resources, Y.Y.; data curation, Y.Y. and C.W.; writing—original draft preparation, Y.Y. and C.W.; writing—review and editing, Y.Y., X.X. and R.A.; visualization, Y.Y.; supervision, R.A.; project administration, Y.Y. All authors have read and agreed to the published version of the manuscript.

Funding

This research received no external funding.

Institutional Review Board Statement

Not applicable.

Data Availability Statement

No new data were created.

Conflicts of Interest

The authors declare no conflicts of interest.

Appendix A

Table A1. Search Algorithm (Use Cochrane Library as an example).
Table A1. Search Algorithm (Use Cochrane Library as an example).
Search Algorithm
((“artificial intelligence” OR “computational intelligence” OR “machine intelligence” OR “computer reasoning” OR “machine learning” OR “deep learning” OR “neural network” OR “neural networks” OR “reinforcement learning”) AND (“aged” OR “geriatrics” OR “geroscience” OR “elderly” OR “older adults” OR “seniors” OR “aging population” OR “senior citizens” OR “aging” OR “old age” OR “advanced age”) AND (“loneliness” OR “social isolation” OR “emotional isolation” OR “solitude” OR “social disconnection” OR “lack of social support”)):ti,ab,kw.

References

  1. Russell, D.; Peplau, L.A.; Cutrona, C.E. The Revised UCLA Loneliness Scale: Concurrent and Discriminant Validity Evidence. J. Pers. Soc. Psychol. 1980, 39, 472–480. [Google Scholar] [CrossRef] [PubMed]
  2. Social Isolation and Loneliness. Available online: https://www.who.int/teams/social-determinants-of-health/demographic-change-and-healthy-ageing/social-isolation-and-loneliness (accessed on 26 August 2024).
  3. Badal, V.D.; Graham, S.A.; Depp, C.A.; Shinkawa, K.; Yamada, Y.; Palinkas, L.A.; Kim, H.-C.; Jeste, D.V.; Lee, E.E. Prediction of Loneliness in Older Adults Using Natural Language Processing: Exploring Sex Differences in Speech. Am. J. Geriatr. Psychiatry 2021, 29, 853–866. [Google Scholar] [CrossRef] [PubMed]
  4. Fakoya, O.A.; McCorry, N.K.; Donnelly, M. Loneliness and Social Isolation Interventions for Older Adults: A Scoping Review of Reviews. BMC Public Health 2020, 20, 129. [Google Scholar] [CrossRef] [PubMed]
  5. Ong, A.D.; Uchino, B.N.; Wethington, E. Loneliness and Health in Older Adults: A Mini-Review and Synthesis. Gerontology 2016, 62, 443–449. [Google Scholar] [CrossRef] [PubMed]
  6. Holt-Lunstad, J.; Smith, T.B.; Baker, M.; Harris, T.; Stephenson, D. Loneliness and Social Isolation as Risk Factors for Mortality: A Meta-Analytic Review. Perspect. Psychol. Sci. 2015, 10, 227–237. [Google Scholar] [CrossRef]
  7. Social Isolation and Loneliness in Older Adults: Opportunities for the Health Care System; National Academies Press: Washington, DC, USA, 2020; ISBN 978-0-309-67100-2.
  8. Jentoft, E.E.; Sandset, T.; Haldar, M. Problematizing Loneliness as a Public Health Issue: An Analysis of Policy in the United Kingdom. Crit. Policy Stud. 2024. [Google Scholar] [CrossRef]
  9. Masi, C.M.; Chen, H.-Y.; Hawkley, L.C.; Cacioppo, J.T. A Meta-Analysis of Interventions to Reduce Loneliness. Personal. Soc. Psychol. Rev. 2011, 15, 219–266. [Google Scholar] [CrossRef]
  10. Chua, C.M.S.; Chua, J.Y.X.; Shorey, S. Effectiveness of Home-Based Interventions in Improving Loneliness and Social Connectedness among Older Adults: A Systematic Review and Meta-Analysis. Aging Ment. Health 2024, 28, 1–10. [Google Scholar] [CrossRef]
  11. Patil, U.; Braun, K.L. Interventions for Loneliness in Older Adults: A Systematic Review of Reviews. Front. Public Health 2024, 12, 1427605. [Google Scholar] [CrossRef]
  12. Bringsjord, S.; Govindarajulu, N.S. Artificial Intelligence. In The Stanford Encyclopedia of Philosophy; Zalta, E.N., Nodelman, U., Eds.; Metaphysics Research Lab, Stanford University: Stanford, CA, USA, 2024. [Google Scholar]
  13. Deep Learning with Python, Second Edition. Available online: https://www.manning.com/books/deep-learning-with-python-second-edition (accessed on 15 January 2025).
  14. Alzubaidi, L.; Zhang, J.; Humaidi, A.J.; Al-Dujaili, A.; Duan, Y.; Al-Shamma, O.; Santamaría, J.; Fadhel, M.A.; Al-Amidie, M.; Farhan, L. Review of Deep Learning: Concepts, CNN Architectures, Challenges, Applications, Future Directions. J. Big Data 2021, 8, 53. [Google Scholar] [CrossRef]
  15. Chauhan, N.K.; Singh, K. A Review on Conventional Machine Learning vs. Deep Learning. In Proceedings of the 2018 International Conference on Computing, Power and Communication Technologies (GUCON), Greater Noida, India, 28–29 September 2018. [Google Scholar]
  16. Botvinick, M.; Ritter, S.; Wang, J.X.; Kurth-Nelson, Z.; Blundell, C.; Hassabis, D. Reinforcement Learning, Fast and Slow. Trends Cogn. Sci. 2019, 23, 408–422. [Google Scholar] [CrossRef] [PubMed]
  17. Reinforcement Learning for Control: Performance, Stability, and Deep Approximators—ScienceDirect. Available online: https://www.sciencedirect.com/science/article/pii/S1367578818301184 (accessed on 15 January 2025).
  18. Jacobs, C.; Johnson, H.; Rennie, T.; Lambert, J.; Joiner, R. Human-Computer Interaction and Artificial Intelligence: Advancing Care Through Extended Mind Theory. Cureus 2024, 16, e74968. [Google Scholar] [CrossRef] [PubMed]
  19. Soori, M.; Arezoo, B.; Dastres, R. Artificial Intelligence, Machine Learning and Deep Learning in Advanced Robotics, a Review. Cogn. Robot. 2023, 3, 54–70. [Google Scholar] [CrossRef]
  20. Human Computer Interaction—An Overview|ScienceDirect Topics. Available online: https://www.sciencedirect.com/topics/computer-science/human-computer-interaction (accessed on 19 January 2025).
  21. Abadir, P.M.; Chellappa, R.; Choudhry, N.; Demiris, G.; Ganesan, D.; Karlawish, J.; Li, R.M.; Moore, J.H.; Walston, J.D. The Promise of AI and Technology to Improve Quality of Life and Care for Older Adults. Nat. Aging 2023, 3, 629–631. [Google Scholar] [CrossRef]
  22. Use of Social Robots in Mental Health and Well-Being Research: Systematic Review. Available online: https://www.researchgate.net/publication/334664525_Use_of_Social_Robots_in_Mental_Health_and_Well-Being_Research_Systematic_Review (accessed on 17 December 2024).
  23. Hoang, P.; King, J.A.; Moore, S.; Moore, K.; Reich, K.; Sidhu, H.; Tan, C.V.; Whaley, C.; McMillan, J. Interventions Associated with Reduced Loneliness and Social Isolation in Older Adults. JAMA Netw. Open 2022, 5, e2236676. [Google Scholar] [CrossRef]
  24. Computational Linguistics Based Text Emotion Analysis Using Enhanced Beetle Antenna Search with Deep Learning During COVID-19 Pandemic—PMC. Available online: https://pmc.ncbi.nlm.nih.gov/articles/PMC10773760/ (accessed on 13 February 2025).
  25. Balasubramani, J.; Surendran, R. Utilizing Hybrid-Deep Learning for Autism Spectrum Disorder Detection in Children via Facial Emotion Recognition. In Proceedings of the 2024 2nd International Conference on Self Sustainable Artificial Intelligence Systems (ICSSAS), Erode, India, 23–25 October 2024; pp. 487–492. [Google Scholar]
  26. Balasubramani, J.; Surendran, R. An Innovative Gated Graph Recurrent Neural Network in Toddler Autism Prediction. In Proceedings of the 2024 2nd International Conference on Self Sustainable Artificial Intelligence Systems (ICSSAS), Erode, India, 23–25 October 2024; pp. 171–176. [Google Scholar]
  27. Broadbent, E.; Loveys, K.; Ilan, G.; Chen, G.; Chilukuri, M.M.; Boardman, S.G.; Doraiswamy, P.M.; Skuler, D. ElliQ, an AI-Driven Social Robot to Alleviate Loneliness: Progress and Lessons Learned. JAR Life 2024, 13, 22–28. [Google Scholar] [CrossRef]
  28. Apr 28, C.G./P.; 2022 AI-Powered Board Game Can Help Reduce Social Isolation in Older Adults. Available online: https://hub.jhu.edu/2022/04/28/ai-board-game-with-virtual-opponent-reduces-social-isolation/ (accessed on 17 January 2025).
  29. Cochrane Handbook for Systematic Reviews of Interventions. Available online: https://training.cochrane.org/handbook/current (accessed on 17 December 2024).
  30. Banks, M.R.; Willoughby, L.M.; Banks, W.A. Animal-Assisted Therapy and Loneliness in Nursing Homes: Use of Robotic versus Living Dogs. J. Am. Med. Dir. Assoc. 2008, 9, 173–177. [Google Scholar] [CrossRef]
  31. Chen, S.-C.; Moyle, W.; Jones, C.; Petsky, H. A Social Robot Intervention on Depression, Loneliness, and Quality of Life for Taiwanese Older Adults in Long-Term Care. Int. Psychogeriatr. 2020, 32, 981–991. [Google Scholar] [CrossRef]
  32. Fields, N.; Xu, L.; Greer, J.; Murphy, E. Shall I Compare Thee…to a Robot? An Exploratory Pilot Study Using Participatory Arts and Social Robotics to Improve Psychological Well-Being in Later Life. Aging Ment. Health 2021, 25, 575–584. [Google Scholar] [CrossRef]
  33. Jones, V.K.; Hanus, M.; Yan, C.; Shade, M.Y.; Blaskewicz Boron, J.; Maschieri Bicudo, R. Reducing Loneliness Among Aging Adults: The Roles of Personal Voice Assistants and Anthropomorphic Interactions. Front. Public Health 2021, 9, 750736. [Google Scholar] [CrossRef]
  34. Lim, J. Effects of a Cognitive-Based Intervention Program Using Social Robot PIO on Cognitive Function, Depression, Loneliness, and Quality of Life of Older Adults Living Alone. Front. Public Health 2023, 11, 1097485. [Google Scholar] [CrossRef] [PubMed]
  35. Loveys, K.; Sagar, M.; Pickering, I.; Broadbent, E. A Digital Human for Delivering a Remote Loneliness and Stress Intervention to At-Risk Younger and Older Adults During the COVID-19 Pandemic: Randomized Pilot Trial. JMIR Ment. Health 2021, 8, e31586. [Google Scholar] [CrossRef] [PubMed]
  36. Papadopoulos, C.; Castro, N.; Nigath, A.; Davidson, R.; Faulkes, N.; Menicatti, R.; Khaliq, A.A.; Recchiuto, C.; Battistuzzi, L.; Randhawa, G.; et al. 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. Int. J. Soc. Robot. 2022, 14, 245–256. [Google Scholar] [CrossRef]
  37. Robinson, H.; MacDonald, B.; Kerse, N.; Broadbent, E. The Psychosocial Effects of a Companion Robot: A Randomized Controlled Trial. J. Am. Med. Dir. Assoc. 2013, 14, 661–667. [Google Scholar] [CrossRef]
  38. Jones, V.K.; Yan, C.; Shade, M.Y.; Boron, J.B.; Yan, Z.; Heselton, H.J.; Johnson, K.; Dube, V. Reducing Loneliness and Improving Social Support among Older Adults through Different Modalities of Personal Voice Assistants. Geriatrics 2024, 9, 22. [Google Scholar] [CrossRef]
  39. Russell, D.W. UCLA Loneliness Scale (Version 3): Reliability, Validity, and Factor Structure. J. Pers. Assess. 1996, 66, 20–40. [Google Scholar] [CrossRef]
  40. PARO Therapeutic Robot. Available online: http://www.parorobots.com/ (accessed on 17 December 2024).
  41. Effectiveness of Artificial Intelligence Robot Interventions on Psychological Health in Community-Dwelling Older Adults: A Systematic Review. Available online: https://jkgn.org/journal/view.php?doi=10.17079/jkgn.2024.00353 (accessed on 17 January 2025).
  42. Shah, S.G.S.; Nogueras, D.; van Woerden, H.C.; Kiparoglou, V. Evaluation of the Effectiveness of Digital Technology Interventions to Reduce Loneliness in Older Adults: Systematic Review and Meta-Analysis. J. Med. Internet Res. 2021, 23, e24712. [Google Scholar] [CrossRef]
  43. Tkatch, R.; Wu, L.; MacLeod, S.; Ungar, R.; Albright, L.; Russell, D.; Murphy, J.; Schaeffer, J.; Yeh, C.S. Reducing Loneliness and Improving Well-Being among Older Adults with Animatronic Pets. Aging Ment. Health 2021, 25, 1239–1245. [Google Scholar] [CrossRef]
  44. Kalil, M. Year of the Humanoid Robot: Top AI Robots to Watch in 2025. Available online: https://mikekalil.com/blog/2024-year-of-the-humanoid-robot/ (accessed on 19 January 2025).
  45. Emotion Detection for Social Robots Based on NLP Transformers and an Emotion Ontology—PMC. Available online: https://pmc.ncbi.nlm.nih.gov/articles/PMC7917797/ (accessed on 19 January 2025).
  46. Humanoid Era: List of All the Most Prominent Companies Building Humanoid Robots. Available online: https://roboticsandautomationnews.com/2024/12/30/humanoid-era-list-of-all-the-most-prominent-companies-building-humanoid-robots/87989/ (accessed on 17 January 2025).
  47. Sidner, C.L.; Bickmore, T.; Nooraie, B.; Rich, C.; Ring, L.; Shayganfar, M.; Vardoulakis, L. Creating New Technologies for Companionable Agents to Support Isolated Older Adults. ACM Trans. Interact. Intell. Syst. 2018, 8, 17. [Google Scholar] [CrossRef]
  48. Al Mazroui, K.; Alzyoudi, M. The Role of ChatGPT in Mitigating Loneliness among Older Adults: An Exploratory Study. Online J. Commun. Media Technol. 2024, 14, e202444. [Google Scholar] [CrossRef]
  49. Follmann, A.; Schollemann, F.; Arnolds, A.; Weismann, P.; Laurentius, T.; Rossaint, R.; Czaplik, M. Reducing Loneliness in Stationary Geriatric Care with Robots and Virtual Encounters—A Contribution to the COVID-19 Pandemic. Int. J. Environ. Res. Public Health 2021, 18, 4846. [Google Scholar] [CrossRef] [PubMed]
  50. Wang, N.; Goel, S.; Ibrahim, S.; Badal, V.D.; Depp, C.; Bilal, E.; Subbalakshmi, K.; Lee, E. Decoding Loneliness: Can Explainable AI Help in Understanding Language Differences in Lonely Older Adults? Psychiatry Res. 2024, 339, 116078. [Google Scholar] [CrossRef] [PubMed]
  51. Chen, N.; Song, J.; Li, B. Providing Aging Adults Social Robots’ Companionship in Home-Based Elder Care. J. Healthc. Eng. 2019, 2019, 2726837. [Google Scholar] [CrossRef] [PubMed]
  52. Hudson, J.; Ungar, R.; Albright, L.; Tkatch, R.; Schaeffer, J.; Wicker, E.R. Robotic Pet Use Among Community-Dwelling Older Adults. J. Gerontol. Ser. B 2020, 75, 2018–2028. [Google Scholar] [CrossRef]
  53. Yen, H.-Y.; Huang, C.W.; Chiu, H.-L.; Jin, G. Social Companion Robots for Alleviating Depression and Loneliness in Institutional Older Adults. Psychiatry Res. 2023, 328, 115425. [Google Scholar] [CrossRef]
  54. Igarashi, T.; Sugawara, I.; Inoue, T.; Nihei, M. Research Participant Selection Bias in the Workshop Using Socially Assistive Robots for Older Adults and Its Effect on Population Representativeness. Int. J. Environ. Res. Public Health 2023, 20, 5915. [Google Scholar] [CrossRef]
Disclaimer/Publisher’s Note: The statements, opinions and data contained in all publications are solely those of the individual author(s) and contributor(s) and not of MDPI and/or the editor(s). MDPI and/or the editor(s) disclaim responsibility for any injury to people or property resulting from any ideas, methods, instructions or products referred to in the content.

Article Metrics

Citations

Article Access Statistics

Journal Statistics
Multiple requests from the same IP address are counted as one view.
Download PDF
A Cost-Minimization Analysis of Teleconsultation Versus In-Person Care for Chronic Diseases and Rehabilitation in Medically Underserved Areas of South Korea
Previous
Examination of Non-Specific Low Back Pain, Pain Perceptions and Disability Between Brazilian Jiu Jitsu, Muay Thai and Boxing Athletes
Next