Words Matter More With Age: Instructions Dictate Self-selected Walking Speed Differently In Young And Older Adults

Variability in gait speed is influenced by both age and assessment methods. Our previous work demonstrated that in young adults, 61% of gait speed variance was attributable to instruction type. However, no study has investigated whether verbal instructions differentially influence gait speed selection in older adults. PURPOSE: This study investigated how walking prompts contribute to gait speed variability in both young and older adults, and whether age-related differences exist in response to various instructions. METHODS: Thirty-six young adults (age 21 ± 2 years) and twenty-eight older adults (age 70 ± 5 years) performed walking trials in response to 24 different instructions. Walking trials were performed in a 12.1-meter lab space using APDM sensors. Multilevel modeling assessed variance components and age-related differences in gait speed. RESULTS: Average walking speed across all instructions was 1.23 ± 0.30 m/s. The first multilevel model showed 28.1% of variance was explained by between-subject differences. Adding instruction type to the model increased total variance explained to 76.5%. A significant interaction between age group and instruction type (χ2 = 86.508, p < 0.001) revealed older adults consistently showed smaller changes in gait speed compared to young adults. Complex instructions (e.g., time pressure scenarios) elicited the largest group differences (β ranging from -0.26 to -0.36 m/s), while simple walking tasks showed minimal differences (β ranging from -0.08 to -0.10 m/s). CONCLUSION: Instructions explain more variance in gait speed (76.5%) when considering both age groups compared to young adults alone (61%). These findings demonstrate that verbal instructions have an even greater influence on gait speed when considering multiple age groups. These results emphasize the importance of standardized instructions in gait assessment and suggest the need for age-specific considerations in clinical and research protocols. Supported by: Research reported in this presentation was supported by the National Center for Advancing Translational Sciences of the National Institutes of Health under award number T32TR004767. The content is solely the responsibility of the authors and does not necessarily represent the official views of the National Institutes of Health.