Motor sequencing in older adulthood: relationships with executive functioning and effects of complexity

Abstract

Objective: Older adults’ motor sequencing performance is more reliant on executive functioning (EF) and more susceptible to complexity than that of younger adults. This study examined for which aspects of motor sequencing performance these relationships hold. Methods: Fifty-seven younger and 90 non-demented, community-dwelling, older adults completed selected subtests from the Delis–Kaplan Executive Function System as indices of EF and component processes (CP; graphomotor speed; visual scanning; etc.), as well as a computerized motor sequencing task (Push Turn Taptap task; PTT). The PTT requires participants to perform motor sequences that become progressively more complex across the task’s four blocks, and is designed to assess action planning, action learning, and motor control speed and accuracy. Results: Hierarchical regressions using each discrete aspect of performance as the dependent variable revealed that action planning is the only aspect of motor sequencing that is uniquely related to EF (beyond the CP composite) for both age groups. Action learning and motor control accuracy are uniquely associated with EF for older adults only, and only if the sequences are complex. Component processes do not fully account for the unique relationships between motor sequencing and EF in older adults. Conclusions: These results clarify prior findings by showing (a) more aspects of motor sequencing relate to EF for older compared to younger adults and (b) for these unique relationships, EF is only related to action during the generation of sequences that are complex. These findings further our understanding of how aging shapes the links between EF and motor actions, and can be used in evidence-based and theoretically driven intervention programs that promote healthy aging.

Keywords:

• Executive functioning
• motor sequence learning
• action planning
• aging
• motor control

Acknowledgments

Yana Suchy, the second author on this article, is the Editor-in-Chief of The Clinical Neuropsychologist (TCN). Per policy of TCN and Taylor & Francis, Dr. Suchy had no influence on the peer review process of this article, was not involved in reviewer selection, was completely blind to the identity of the reviewers, and did not participate in any editorial decisions. Additionally, the ScholarOne article submission portal indefinitely blocks Dr. Suchy from accessing any information about the peer review process as it pertains to this article, including the identity of the reviewers and any correspondence between the reviewers and the Action Editor. For this particular article, Dr. Robin Hilsabeck served as the Action Editor.

Notes

1. Demographic data were available for all participants except for measures of depression (n = 143) and race/ethnicity (n = 142); all missing data were from the young adult group. It should also be noted that given violations of Levene’s test for homogeneity of variances, t-tests that do not assume equal variances were used. Findings remained unchanged if equal variances were assumed.

2. All analyses were also conducted with left-handed young adults excluded (n = 7). The main findings remain virtually unchanged, and any changes that emerged are noted in the results section.

3. Simple slopes were computed using the online utility (Preacher, Curran, & Bauer, 2006).

4. Without the left-handed younger adults, the interaction becomes a trend-level association (p = .065). Additionally, the R² – change value for EF in the hierarchical regression – becomes statistically significant [R² Δ = .03, p = .033]. However, the results of the simple slopes testing do not change (i.e. relationship between action learning and EF remains statistically significant for the older adults only [B = 5.04, p = .005]).

5. We computed a health score composite that includes a history of high blood pressure, heart disease, diabetes, sleep apnea, and chronic obstructive pulmonary disease where a higher score is indicative of more self-reported conditions. Though not explicitly asked, 71.4% of participants with any of these medical conditions voluntarily self-reported that all of their conditions were medically controlled.