This study will use a socially assistive robot to deliver cognitive training in the form of a music (piano) learning intervention to socially isolated older adults with mild cognitive impairment.

Cognitive training has been shown in past research to improve cognitive function in older adults with and without mild cognitive impairment. Music learning has been previously shown to be a particularly effective form of cognitive training: in addition to the cognitive demands of learning notation, hand positioning, coordination and timing, those who engage in music learning also benefit from emotional connection to the music and social engagement with a music teacher.

Not all older adults are able to access music lessons. Older adults who are homebound, live in rural areas and do not have access to reliable transportation (or a monthly income with sufficient funds for recurring payments for lessons) are unlikely to be able to access these benefits. Socially assistive robots that can deliver both education and companionship on an ongoing basis within the home environment may be a viable solution to these limitations.

The purpose of this study is to deliver cognitive training in the form of a music (piano) learning intervention to older adults who are socially isolated, and who may be able to benefit from the cognitive training aspects of the intervention. A socially assistive robot will deliver the intervention.

Inclusion Criteria:

• Be able to speak, read and write in English with a high degree of fluency
• Cumulative music learning experience < 3 years
• Able to sit upright for two hours at a time
• Able to see images on a screen at a distance of approximately 20 inches
• Must have independent transportation to and from the Institute of Gerontology, weekly, for 27 weeks
• No hearing impairment beyond what is correctable via hearing aid
• Must have a safe location in the home to store and use an 88-key keyboard + stand and piano bench
• No more than mildly cognitively impaired: global deterioration scale no higher than 2-3; Clinical Dementia Rating Scale no higher than 0.5 (sum of boxes)
• Must have a close companion, family member or loved one who can attend the first and last visit
• Must be capable of providing informed consent (based on responses to questions on study purpose, tasks, benefits, incentives and risks).
• Must have a computer or tablet that is connected to broadband internet, to be used three times over the course of the study
• Must have access to an e-mail account that can be checked three times during the study

Exclusion criteria:

• No previous diagnosis of Alzheimer's disease or related dementias
• No condition that limits movement of the hands or fingers

IPD will not be shared with researchers outside of the study site.