Utilizing a Socially Assistive Robot to Increase Access to Cognitive Training Resources

Background and Purpose: Accessibility to cognitive health resources can greatly impact older adults ability to age successfully and maintain a high quality of life. Age-related dementias affect more than five million older adults in the United States and 46.8 million globally. Cognitive training interventions, such as learning an instrument, have been identified across the literature as effective strategies toward increasing cognitive reserve, neural plasticity, and reducing risk of dementia. However, due to various barriers to access (e.g., cost, transportation, instructor availability), older adults can have a difficult time in being able to engage in cognitive training. However, new emerging technologies, such as Socially Assistive Robots (SARs), present a unique opportunity towards increasing access and providing older adults with the ability to engage in cognitive training via robot-led customizable piano lessons.

Methods: The present study aimed to understand the feasibility of SAR-led piano lessons, to engage older adults with mild cognitive impairment (N = 11; M= 74.64 ± 6.02 years of age; 72.72% female; 90.1% White/Caucasian) in cognitive training. The data collection occurred across four weekly robot-led piano lessons. Participants were recruited utilizing a combination of emails shared via listservs and flyers posted at local agencies and medical health offices. Participant requirements include the following: age 65+; experiencing forgetfulness; no prior music experience exceeding three years; ability to hear music; desire for more social interactions. During each session, participants engaged in a robot-led piano lesson, completed a battery of scales, and participated in a structured qualitative interview. Cognitive status was assessed using the Mini-Mental State Exam and the CNS Vital Signs to measure cognitive domain-specific functioning. Perceptions and acceptance of the SAR were measured using the Robotic Social Attributes Scale and Technology Acceptance Scale. Qualitative interviews aimed to understand participants’ acceptance, affect, learning experience, perceived connection, and willingness to adopt a SAR.

Results: Participants cognitive function improved in the verbal memory, executive function, reaction time and cognitive flexibility domains, and in the computed neurocognitive index score (t[1,9]=2.20, p=0.03). The SAR was perceived as easy to use, useful, warm and competent. Furthermore, ten out of eleven participants noted willingness to adopt the SAR to engage in piano lessons. Preliminary analysis of qualitative interviews indicated that older adults perceived the robot as an effective and efficient piano instructor, socially engaging, and adaptive to their needs.

Conclusions and Implications: Participants improvement across various domains of cognition, along with perceived usability and willingness to adopt the SAR, indicated feasibility of robot-led cognitive training. Traditionally, accessing cognitive training, such as piano lessons, can be limited due to cost, program availability, and lack of customizability to meet the variability in older adults' needs. SARs can help address this gap in resources and increase access through accessible and cost-effective programs (the cost of the robot is less than year-long lessons) and deliver customizable cognitive training for adult learners in various environments. Increasing access can help alleviate social injustices due to limited accessibility to cognitive health resources and service availability for older adults wanting to age in place.