Social Robotics

Social robotics is an interdisciplinary field concerned with the design, implementation and evaluation of robots that engage humans through social behaviours, often encompassing speech, gesture, facial expression and emotional cues. Drawing on robotics, artificial intelligence, cognitive science and human–computer interaction, social robots are intended to operate in daily settings—homes, schools, healthcare facilities and public spaces—where they support companionship, education, therapy and assistance. Key challenges include endowing robots with the capacity to perceive and interpret human intention and affect, to plan appropriate responses in real time and to adapt through learning. Progress in natural-language processing, computer vision and affective computing has enabled robots to recognise emotions, sustain dialogue and tailor interactions to individual preferences. Applications range from cognitive support for older adults and language tutoring for children to collaborative customer service and rehabilitation therapies. The global significance of social robotics lies in addressing demographic shifts, alleviating workforce shortages in care professions and providing novel tools for inclusion and autonomy. By integrating ethical guidelines, user-centred design and robust safety frameworks, the field seeks to deliver socially adept machines that enhance well-being and foster positive human–machine partnerships.

Research from Nature Portfolio

Recent studies have explored the acceptability and adherence of robot-assisted therapeutic protocols for children with autism spectrum disorder. One protocol adapted Pivotal Response Treatment to a humanoid platform, reporting over 80 per cent adherence and high likability ratings from both children and parents. Movement, speech and game scenarios were identified as the most engaging elements, while parental feedback highlighted the need for greater behavioural flexibility in the robotic agent.

Work on ordered interpersonal synchronisation has examined how autistic children from distinct cultures coordinate movements and emotional empathy with both human and robotic partners. Unlike responses to human agents, synchronisation patterns with robots converged across national groups, suggesting that non-human partners can elicit more uniform engagement. These findings underscore the potential of robots to bypass cultural and social barriers in therapeutic synchronisation and empathy training.

Research from all publishers

A meta-analysis of anthropomorphism in service robots synthesised data from over 11 000 interactions, revealing that customer traits, robot design features and task context jointly determine the degree to which users attribute human-like qualities. Physical embodiment, conversational behaviour and perceived usefulness emerged as key mediators of user intention, providing a comprehensive model to guide the social design of robots in retail, hospitality and care environments.

An experimental study of a cyber-physical robotic system assisting speech and language pathologists reported improvements of up to 11 percentage points in articulation therapy among high-school students. By integrating real-time tracking and adaptive feedback, the system enabled clinicians to extend session duration and focus on complex exercises, demonstrating a viable path for robots to co-therapist roles in educational and clinical contexts.

A biologically inspired self-organising map model has been developed to support infant-like learning of social behaviours by robots. Through unsupervised exploration, the map dynamically formed hierarchical feature clusters that grounded symbols and adapted to novel conditions. This approach yields interpretable visualisations of learned patterns and paves the way for robots to acquire social understanding in unstructured environments without extensive manual configuration.

Social Robotics publication trend

The graph below shows the total number of articles in social robotics across all publications each year (not limited to Nature Index journals).

Technical terms

Social robot: An interactive robotic system designed to engage users through social behaviours such as speech, gesture and emotional expression.

Human–robot interaction (HRI): The interdisciplinary study of communication and collaboration between humans and robots, encompassing perception, cognition and behaviour.

Anthropomorphism: The attribution of human-like characteristics to non-human agents, influencing user engagement, trust and acceptance.

Self-organising map: An unsupervised neural network that clusters high-dimensional input data into a lower-dimensional representation, facilitating adaptive feature learning.

References

1. Adherence and acceptability of a robot-assisted Pivotal Response Treatment protocol for children with autism spectrum disorder. Scientific Reports (2020).
2. Ordered interpersonal synchronisation in ASD children via robots. Scientific Reports (2020).
3. Understanding anthropomorphism in service provision: a meta-analysis of physical robots, chatbots, and other AI. Journal of the Academy of Marketing Science (2021).
4. Experimental Analysis of the Effectiveness of a Cyber-physical Robotic System to Assist Speech and Language Pathologists in High School. Journal of New Approaches in Educational Research (2023).