Editorial: Digital health applications of social robots

1 Introduction

The intersection of Artificial Intelligence (AI), robotics and healthcare is gaining prominence, particularly the use of socially assistive robots (SARs) (Stanojevic et al., 2023; Bennett et al., 2022a; Bennett et al., 2024; Bennett et al., 2018; Collins et al., 2024; Stanojevic et al., 2025; Bennett et al., 2017; Hsu and Šabanović, 2013; Perumandla et al., 2025; Šabanović et al., 2013; Toombs et al., 2024). As one of the emerging forms of digital therapeutics (DTx) tools, SARs can improve patient care and engagement alongside technology (i.e., mobile phones, wearables, smart home devices) (Stanojevic et al., 2023). This transformative shift in healthcare introduces new tools, while also shaping how we think about care delivery, patient engagement and the everyday ecologies in which health and wellbeing are supported.

SARs, as a DTx tool, can benefit patients directly through interactive capabilities as well as part of the broader “Digital Health” (DH) systems. Data from sensors onboard the robot, for example, can help identify in-home behaviors, activity patterns and health status of patients remotely (Stanojevic et al., 2023). Furthermore, linking the robotic sensor data to other DH system components (e.g., mobile phones, wearables, smart home devices, virtual/augmented reality) enables the SAR to function as part of an Internet of Things (IoT) ecosystem, creating a broader long-term picture of patient health outcomes in daily life beyond the clinic (Bennett et al., 2022b). In this way, SARs are not simply “devices” placed in a home or clinical setting; they become active, sensing, adaptively-responding components of a larger networked care infrastructure. Such a paradigm shift is transformative for robots in healthcare.

In 2021, $57.2 billion was invested in DH systems around the world (Rock Health, 2022), recognizing the promise this concept holds for aiding in delivery and care management. DH systems traditionally include a blend of various technologies, AI and physiological biomarkers, which combined have the potential to provide support for individuals with a broad array of health conditions. DTx is a more specific set of technology-enabled interventions within the broader DH sphere intended to produce a measurable therapeutic effect (Goldsack et al., 2019), empowering both patients and healthcare providers. This dual therapeutic and informational role positions DTx, and SARs within it, as a crucial bridge between moment-to-moment patient experience and longer-term clinical decision-making.

The main challenge with social robots within DH systems is that the sheer volume and limited oversight of different robotic platforms is hindering validation efforts from technical, clinical, system and privacy standpoints, thus consequently slowing widespread adoption of SARs as treatment tools. Regulatory frameworks, clinical trial designs and implementation strategies are struggling to keep pace with rapid technological development. Therefore, the goal of this Research Topic was to solicit contributions that demonstrate the use of social robots in DH applications, either via empirical research studies or via proposed theoretical solutions, as well as to position SARs more clearly as part of rigorously validated, ethically grounded, and clinically meaningful DTx ecosystems. The Research Topic features several papers that move us closer to that goal.

2 Contents of the special issue

Authors were encouraged to submit papers for this Research Topic that contribute to advancing the integration of social robots with other DH components, so as to enhance the overall understanding of patient health outcomes. This Research Topic aimed to gather insights focused on overcoming the obstacles related to validating and adopting technologies for DTx in general (robots or otherwise), particularly concerning technical, clinical, system and privacy aspects. We were especially interested in work that does not treat SARs as stand-alone tools, but rather situates them within multi-device ecosystems, complex care networks, clinical organizational structures, and everyday life contexts.

The authors who submitted work for this special topic have aptly responded to the call. It is evident from the work provided that SARs are increasingly being considered as DTx tools with the potential to be integrated into the broader DH systems. The papers published in this Research Topic cover usage of SARs across the lifespan, from children in acute care settings to older adults living at home or in long-term care. Papers framed as perspectives, systematic literature reviews and original research give both breadth and depth to the subject matter that is being explored. One of the key, commendable themes that connects all these works is the participatory nature of research being done in this field. The construction, refinement and deployment of robots is conducted through approaches that actively involve end users, caregivers, clinicians and other stakeholders in the design of robot systems for healthcare. Furthermore, in the SARs field, particularly, there is a strong and growing recognition of the need for theory-driven, context-aware and trust-sensitive design if these systems are to move from promising prototypes to reliable DTx components of digital health. In other words, “safety” is of paramount importance if social robots or other AI systems are going to be used for healthcare purposes in the real world. That is a common thread throughout all the papers included in this Research Topic, whether it stems from careful alignment of robot behavior during design a priori or incorporating appropriate guardrails for the AI after deployment a posteriori.

2.1 Highlights of the collected papers in this issue

Boudouraki et al. use the NASSS (Non-adoption, Abandonment, Scale-up, Spread, Sustainability) implementation framework to show how robotics and sensor platforms in care homes are shaped by complex organizational, regulatory and material constraints, as well as competing stakeholder values. Building on their previous work and engaging healthcare stakeholders, the authors introduce the concept of an adaptive system where sensor technologies are integrated into robotic platforms through a series of interactive demonstrations. This research provides a better understanding of SARs as a DTx tool that would support and fit within broader DH systems, particularly in settings such as care homes where organizational realities and resource constraints are ever-present.

Klier and Lugrin proposed a wellbeing-centered conceptual model for SAR applications with cognitively healthy older adults, grounded in Self-Determination Theory (SDT) and Positive Emotions, Engagement, Relationships, Meaning, and Accomplishment (PERMA) model, to systematically align robot behaviors with basic psychological needs and flourishing rather than only deficit reduction. The contribution of this work to the field of social robotics is very significant, as the use of SARs in treatment should be aligned with strength-based, person-centered approaches that are moving away from traditional deficit-based frameworks. If SARs are to become effective DTx tools and part of validated DH systems, the use and design of these robots should be informed by models such as SDT addressing autonomy, competence and relatedness of patients and PERMA, with special focus on positive emotions, engagement, relationships, meaning and accomplishment. Presenting a step forward in the right direction bridging clinical practice and robotic design this article explicitly connects psychological theory, wellbeing outcomes and SAR behavior design.

Extending these insights into an acute care environment, Foster et al. describe the participatory designed, AI-enhanced planning and real-world usability testing of a SAR deployed in pediatric emergency departments. The concrete design guidelines for clinical SAR applications focus on managing children’s pain and distress during intravenous procedures. Achieving successful autonomous adaptiveness of the robotic agent is commendable in itself, and even more commendable when we consider what this DTx tool is used for. Beyond technical challenges, which were addressed through iterative refinement, this team successfully integrated the SAR into clinical workflows, with positive responses from children, caregivers and healthcare providers. Their work exemplifies how SARs can be woven into busy clinical environments in ways that respect existing routines while opening new possibilities for emotionally attuned, AI-supported care.

Complementing these perspectives, Olatunji et al. conducted a synthesis of literature on design considerations for SARs that support older adults’ healthy activities at home, emphasizing user-centered design, integration within existing care networks and a roadmap for embedding SARs into DH systems. Systematic reviews often offer much-needed context and build arguments and directions for certain points of view. A somewhat obvious conclusion, that SARs need to be designed for specific needs and be accepted by and useful to the users, is a much-needed reiteration in this field that paves the way to the third conclusion these authors make - integration of SARs into the healthcare ecosystem and care network. Digital health systems need this type of thinking if adoption and validation of these systems is to be done. The emphasis on design according to specific needs of the DTx tool, with adaptive abilities, is another aspect that would provide more usefulness of the tool itself and increase the likelihood of sustained, meaningful engagement.

Along these lines, Gul et al. systematically review how trust in SARs is conceptualized and measured with older adults, highlighting methodological gaps and introducing the Subjective Objective Trust Assessment Human Robot Interaction (SOTA-HRI) framework combining subjective and objective trust indicators as a prerequisite for sustained real-world adoption. A critical aspect that hinders ubiquitous SAR use as DTx is that most previous studies were conducted in lab settings with subjective self-report measures like questionnaires, interviews and surveys to measure the trust of older adults in SARs. Also, convenience sampling was identified as a concern, where many studies reviewed focus on healthy older adults without age-related disabilities. Furthermore, findings reported by these authors align with previous claims that the lack of standardization is one of the key problems in adoption of DH systems and is emblematic of deeper issues in the way studies are conducted (Mathews et al., 2019). Gul et al. analyzed factors influencing trust and found no consensus on any single factor that exerted greater influence. This paper should be observed as a call to action in the field of SARs and, perhaps, a pointer toward the next Research Topic topic focused specifically on trust, ethics and real-world validation.

Collectively, these papers motivate and inform work to showcase SARs as DTx tools, and tackle the intertwined challenges of ecosystem integration, real-time data use, AI-driven adaptation, and the validation, acceptability and trustworthiness required for SARs to become reliable components of next-generation DH systems. They frame SARs not as isolated gadgets, but as socio-technical interventions whose therapeutic impact depends on careful attention to context, wellbeing outcomes, ecosystem integration, robust autonomy and trustworthy interaction. As editors of this Research Topic, we see these contributions as both a reflection of how far the field has come and an invitation to continue pushing forward, toward SARs that are clinically rigorous, ethically grounded, person-centered and deeply embedded within the digital health systems that increasingly shape contemporary care.

References

• 1

  Bennett C. C. Sabanovic S. Piatt J. A. Nagata S. Eldridge L. Randall N. (2017). “A robot a day keeps the blues away,” in 2017 IEEE international conference on healthcare informatics (ICHI).

  • Google Scholar

• 2

  Bennett C. Stanojević Č. Šabanović S. Piatt J. Kim S. (2018). “When no one is watching: ecological momentary assessment to understand situated social robot use in healthcare,” in Proceedings of the 9th international conference on human-agent Interaction2021.

  • Google Scholar

• 3

  Bennett C. C. Stanojevic C. Kim S. Lee J. Yu J. Oh J. et al (2022a). “Comparison of in-home robotic companion pet use in South Korea and the United States: a case study,” in 2022 9th IEEE RAS/EMBS International Conference for Biomedical Robotics and Biomechatronics (BioRob). (IEEE), 1–7.

  • Google Scholar

• 4

  Bennett C. C. Ross M. K. Baek E. Kim D. Leow A. D. (2022b). Smartphone accelerometer data as a proxy for clinical data in modeling of bipolar disorder symptom trajectory. NPJ Digit. Med.5 (1), 181. 10.1038/s41746-022-00741-3

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 5

  Bennett C. C. Stanojevic C. Oh J. Ahn J. Jeon Y. Son K. et al (2024). “Digital health sensor data in autism: developing few shot learning approaches for traditional machine learning classifiers,” in 2024 IEEE 20th International Conference on Body Sensor Networks(BSN) (IEEE), 1–4.

  • Google Scholar

• 6

  Collins S. Henkel Z. Henkel K. B. Bennett C. C. Stanojević Č. Piatt J. A. et al (2024). “The ins and outs of socially assistive robots: sensors and behaviors of a therapeutic robot for depression management,” in 2024 33rd IEEE International Conference on Robot and Human Interactive Communication (ROMAN). (IEEE), 1624–1629.

  • Google Scholar

• 7

  Goldsack J. Coder M. Fitzgerald C. Navar-Mattingly N. Corvos A. Atreja A. (2019). Digital Health, digital medicine, digital therapeutics (DTx): what’s the difference. DiMe Source18 (02), 2022.

  • Google Scholar

• 8

  Hsu L.-J. Šabanović S. (2013). “Designing AI for partnership in care,” in Proceedings of the AAAI symposium Series2024.

  • Google Scholar

• 9

  Mathews S. C. McShea M. J. Hanley C. L. Ravitz A. Labrique A. B. Cohen A. B. (2019). Digital health: a path to validation. NPJ Digital Medicine2 (1), 38. 10.1038/s41746-019-0111-3

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 10

  Perumandla R. Bae Y. H. Izaguirre D. Hwang E. Murphy A. Hsu L. J. et al (2025). Developing conversational speech systems for robots to detect speech biomarkers of cognition in people living with dementia. arXiv preprint arXiv:2502.10896.

  • Google Scholar

• 11

  Rock Health (2022). 2021 year-end digital health funding: seismic shifts beneath the surface. Available online at: https://rockhealth.com/insights/2021-year-end-digital-health-funding-seismic-shifts-beneath-the-surface/.

  • Google Scholar

• 12

  Šabanović S. Bennett C. C. Chang W.-L. Huber L. (2013). “PARO robot affects diverse interaction modalities in group sensory therapy for older adults with dementia,” in 2013 IEEE 13th international conference on rehabilitation robotics (ICORR).

  • Google Scholar

• 13

  Stanojevic C. Bennett C. C. Sabanovic S. Collins S. Baugus Henkel K. Henkel Z. et al (2023). Conceptualizing socially-assistive robots as a digital therapeutic tool in healthcare. Front. Digital Health5, 1208350. 10.3389/fdgth.2023.1208350

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 14

  Stanojevic C. Piatt J. Sabanovic S. (2025). Attitudes of special educators towards use of socially assistive robots for individuals with autism Spectrum disorder in Serbia and the United States. J. Special Educ. Technol.40 (3), 283–297. 10.1177/01626434241270592

  • CrossRef
  • Google Scholar

• 15

  Toombs A. L. Montague K. Wong R. Y. Brewer R. N. Naik S. Parsons P. C. et al (2024). “Future dialogues: personal AI assistants and their interactions with us and each other,” in 2024 Companion Publication of the 2024 Conference on Computer-Supported Cooperative Work and Social Computing, 675–678.

  • Google Scholar