Social Robotics Technology and Engagement in Young Children: A Systematic Review

• In Event: Poster Session 04
  In Poster Session: PS 04 Topic - Technology, Media & Child Development

Abstract

Introduction: The use of innovative robotics technology is becoming more prevalent in developmental research. Robots are rapidly evolving as engagement tools in educational, therapeutic, community, and home settings; however, best practices for supporting young children’s social engagement within human-robot interaction (HRI) remain unclear (Crompton et al., 2018; Di Lieto et al., 2020; Scassellati et al., 2018). While a vast number of HRI studies with children focus on whether the child prefers a robot partner over a human partner, or which robot characteristics promote likeability (Fernandez-Llamas et al., 2018; Sandoval et al., 2020), we instead draw attention to the methodological design elements and HRI qualities that specifically promote children’s social engagement, motivation, and communication. For example, a recent HRI study has implemented a novel memory and emotion cognition model in which a robot progressively stored and retrieved information from prior child-robot exchanges (Ahmad et al., 2019). Additionally, researchers have combined novel robotic models with technologies such as Kinect sound recorders (Taheri et al., 2019), eye-tracking (Manzi et al., 2020, Cao et al., 2019), Photovoice (Carlson et al., 2016), and even machine learning (Javed et al., 2020, Rudovic et al., 2018). Social robotics research paradigms are promoting long-term gains in children’s social engagement skills in a variety of HRI contexts by specifically targeting social interest. And, given the pivotal role of social engagement and social interest in shaping children’s early developmental trajectories (Adamson et al., 2019; Heymann et al., 2018; Parlade & Iverson, 2015), we aim to review empirical evidence in the field of social robotics technology and explicate how researchers have utilized HRI to investigate and bolster child gains in communication as well as related developmental skills.

Methods: An initial search with PsychINFO, MEDLINE, and Web of Science databases initially yielded 6,600 articles (see Figure 1). Following record screening and eligibility assessment, the remaining 194 articles were coded and assessed for research focus, sample characteristics, robot characteristics, methodology, child-robot interaction qualities, and targeted child behaviors (see Table 1 for an expanded description of article content synthesized). Intra and inter-rater reliability were calculated for agreement on coding criteria listed above and  ranged from .92 to .96.

Results: This review aimed to examine current and developing social robotics technologies in the context of HRI and to identify emerging best practices for promoting young children’s social communication development with HRI methodologies. Heterogeneity was observed in reported robot features (i.e., brand, motor and speech capabilities), child populations targeted (i.e., neurotypical children versus children with developmental disorders like autism spectrum disorder), and targeted child behaviors (i.e., joint attention, vocabulary learning, storytelling, imitation, play skills). In contrast, studies were more consistent in reported use of analysis strategies and child-robot interaction setup.

Conclusion: While current evidence investigating the use of HRI with young children to promote early developmental skills varies greatly in precise aims, methodologies, and analytical strategies, the themes and trends identified in our review highlight the benefits of incorporating HRI to support social communication and related skills in a variety of contexts for young children.

Authors

• Jessica Elaine Blume, Texas Tech University
  Non-Presenting Author

• Miranda Cox, Texas Tech University
  Presenting Author

• Katheryn Webb, Texas Tech University
  Non-Presenting Author

• Ann M Mastergeorge, Texas Tech University
  Non-Presenting Author