Closed-loop Intervention to Promote a Supportive and Interactive Environment around Children
Lead PI:
Ou Bai
Co-Pi:
Abstract

For both parents and educators, monitoring and adjusting their behaviors to ensure that children develop appropriate prosocial and learning behaviors is a complex balance between nurturance and limit setting. When these interactions are strained, negative or coercive cycles may emerge that delay appropriate development and exacerbate existing impairment. To disrupt the development of coercive cycles, adults must have the ability to accurately assess the quality of their interactions with children and integrate this information into personal change. Approaches to measuring these types of interactions will inform what we know about the mechanisms of child social, emotional, and learning development in STEM learning settings, and enable the creation of adaptive interventions for those moments when support is most needed. This project envisions a closed-loop intervention framework to promote a supportive and interactive environment around children. Smart wearables will sense interaction and responses between the children and their parents or educators, using embedded machine learning technology to recognize supportive behaviors. The perceived behaviors will be sent to a cloud server where adaptive interaction strategies will be identified from either online psychological consultation or artificial intelligence. These interaction strategies will then be provided to the parents and educators in the form of guidance cues to promote a supportive STEM learning environment around the children.

This planning project aims to understand the barriers and critical problems in the implementation of smart technology and psychological strategies to support adult-child interactions in STEM learning settings. The work will proceed by convening key stakeholders (parent organizations, formal educational institutions, and informal educational institutions) in a series of iterative discussions to produce a set of adult-child behavioral targets that are essential to children’s development of social, emotional, and learning skills. Further discussions will then identify mechanisms to enhance these behaviors, and reduce competing, less effective approaches. Qualitative thematic analysis of the discussions will be used to capture these behaviors and mechanisms. Then technologies will be developed to measure, provide feedback on, and improve these behaviors. These devices will be piloted with adult-child dyads. Audiovisual data collected by the devices will be human coded as well as processed by algorithms to vet the technological capacity of the devices to detect and respond to targeted behaviors. A series of debriefing interviews and surveys with adult-child dyads will be used to determine the feasibility, acceptability, and utility of the devices. The collected preliminary data will support the forming of critical technological and social science research questions that co-inform one another: questions about the social engagement between adults and children will drive the technical research, and what can be discovered via the technological research will open up new questions that can be posed about social engagement between children and adults. Adult-child interactions are key social factors that integrate to produce student social, emotional, and academic outcomes. Within our informal educational communities, our formal educational communities, and our familial communities it is essential to find the best mechanisms for measuring, providing feedback, and improving these interactions. This work thus seeks to advance a new approach to, and evidence-based understanding of, the development of STEM learning. This Smart and Connected Communities project is also supported by the Advancing Informal STEM Learning program, which seeks to (a) advance new approaches to and evidence-based understanding of the design and development of STEM learning in informal environments; (b) provide multiple pathways for broadening access to and engagement in STEM learning experiences; (c) advance innovative research on and assessment of STEM learning in informal environments; and (d) engage the public of all ages in learning STEM in informal environments.

This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.

Ou Bai
Dr. Bai serves as the Director of Human Cyber-Physical Systems (HCPS) Laboratory in FIU. He leads the HCPS Lab to develop new sensor, network and computing technologies to accelerate both the creation and our understanding of the complex and increasingly coupled relationships between humans and computing with the broad goal of advancing human capabilities. Dr. Bai has successfully developed a number of Intent-of-Things (IoT) applications that may promote better human interactions with physical world, such as brain-computer interfaces (BCIs), neuro prosthetics, and human-robot interactions. Dr. Bai has published more than 100 Journal papers and conference proceedings. He is member of IEEE EMBS society.
Performance Period: 10/01/2021 - 03/31/2024
Institution: Florida International University
Award Number: 2125549