Reducing Loneliness for Long Term Care Older Adults through Collaborative Augmented Reality
Lead PI:
Nilanjan Sarkar
Co-Pi:
Abstract

This project seeks to reduce loneliness in older adults who reside in long term care (LTC) communities through new augmented reality (AR) technology. Loneliness is a serious condition that is related to increases in heart disease, depression, suicide, mental and physical decline, and reduced quality of life and death. Two out of five older adults in the U.S. report being lonely. Even more alarming, three out of four LTC older adults experience loneliness. The COVID-19 pandemic, with its accompanying safety protocols, has intensified loneliness across the LTCs. The project will discover how augmented reality can reduce loneliness in LTC older adults by linking them with family members who reside elsewhere. This project will allow older adults and family members to see each other’s 3-dimensional realistic images, eat meals together, and interact with one another in various activities, such as playing cards. Investigators of this project are experts in engineering, computer science, gerontology, nursing, medicine and social health science. Working with older adults and family members in the design and testing of the AR technology, the team will compare AR to 2D interactive communication technologies, such as Zoom or Facetime. Initial understanding of the feasibility and acceptability of this enhanced AR technology among older adults, families and LTC staff will guide future studies targeting loneliness, ultimately improving quality of life for older adults. The community focus for this project will be older adults residing in LTC communities in Middle Tennessee with the potential to scaling the solution across the nation.

The project will fundamentally advance the scientific and the technological methodologies of collaborative Augmented Reality to enhance social presence and thus social connectedness, to create realistic and socially appropriate interactions. It will make several fundamental contributions in both technology and social science during the course of this research: 1) create a novel multi-objective optimization based framework that minimizes positional errors of the hand of the avatar while preserving its nonverbal behavior with respect to the human it represents; such an ability will allow shared activities (e.g., drinking tea together) with appropriate social nonverbal behavior (e.g., gaze and postures), a critical component of communication; 2) create a new methodology of a user’s motions onto its avatar to generate naturalistic, socially appropriate motion that respects dissimilarities between the user’s and its avatar’s environments (e.g., differences in room geometries) through novel motion mapping and optimization that ensures natural walking patterns; 3) develop a greater understanding of the feasibility, acceptability and social presence in the use of varying collaborative AR activities and environments for older adults with different levels of cognitive impairment and their family members; 4) develop a greater understanding of the impact of collaborative AR on loneliness based on level of cognitive impairment; 5) gain a greater understanding of the logistics and deployment of this technology in LTCs and family homes to inform scalability; and 6) create activity design guidelines for reduction of loneliness in older adults. The research will be conducted through participatory design using key stakeholders (e.g., older adults, activity directors, LTC management) and evaluated using a two-arm experimental design comparing collaborative AR to current state-of-the-art 2D interactive communication technologies.

Nilanjan Sarkar
I am interested in the analysis, design, and development of intelligent and autonomous systems that can work with people in a versatile and natural way. The applications of this research range from helping individuals with autism and other developmental disabilities in learning skills, aiding stroke patients to regain some of their movement abilities through robot-assisted rehabilitation, and providing more autonomy in robots for a variety of tasks. We are developing new generations of robots and computer-based intelligent systems such as virtual reality systems that can sense human emotion from various implicit signals and cues such as one’s physiology, gestures, facial expressions and so on, to be able to interact with people in a smooth and natural way. My current research involves both theoretical analysis and experimental investigation of electromechanical systems, sensor fusion and machine learning, modeling of human-robot and human-computer interaction, kinematics, dynamics and control theory leading to the development of these smart systems.
Performance Period: 10/01/2022 - 09/30/2026
Institution: Vanderbilt University
Award Number: 2225890