Musical instrument engagement across the life course and episodic memory in late life: An analysis of 60 years of longitudinal data from the Wisconsin Longitudinal Study

BACKGROUND: As the global burden of dementia increases, the absence of treatment underscores the need for identification of factors that may improve cognitive reserve–the ability to stave off cognitive decline in old age. The beneficial association between musical instrument engagement and episodic memory has been identified in children, young adults, and older adults. Yet, previous studies in musical instrument engagement have rarely examined the potential for adolescence and adulthood exposures to independently improve cognition, nor have they been linked with the rate of memory decline over time in older adults. We investigated whether adolescent musical instrument engagement and continued musical instrument engagement over the adult life course were separately associated with higher episodic memory, as well as rate of decline in a large longitudinal cohort. METHODS: Data were from a prospective cohort of high school graduates from 1957. High school music engagement (HSME) was ascertained through graduate yearbooks and assessed as membership in musical performance groups. A questionnaire was used to assess musical engagement through adulthood (MEA) at ages 35, 55, and 65. The episodic memory score was composed of immediate and delayed recall task scores, and was assessed when participants were aged approximately 65 and 72 years old among 5,718 individuals. Linear mixed models were used to assess the association between music, and memory performance and decline over time. RESULTS: Of high school graduates who participated in the study, 38.1% played music in high school, and 21.1% played music in adulthood. While musical engagement was more common in those who played in childhood, 40% of those who played continuously as an adult did not play in high school. High HSME (B = 0.348, p = 0.049) and continuous MEA (B = 0.424, p = 0.012) were associated with higher memory scores at age 65 after covariate adjustment. When examining memory decline, the benefits of high HSME decreased over time (B = -0.435, p = 0.048), while the rate of decline did not differ between MEA groups. Exploratory models revealed differential benefits for HSME and immediate recall, and MEA and delayed recall. CONCLUSION: This study provides further evidence that musical engagement in childhood or adulthood is associated with non-musical cognitive reserve. These two exposures may act differentially in different domains of episodic memory. Further work is needed to determine the relationship between musicianship and the rate of cognitive decline.