Instructions and skill level influence reliability of dual-task performance in young adults

The purpose of this study was to assess the trial-to-trial repeatability of dual-task performance and establish the minimal detectable change (MDC95) of gait-related dual-task interference. Thirty-one healthy young adults (22.5, SD 2.1 years) performed texting and walking tasks in isolation (single-task) and in combination (dual-task). The dual-task was repeated with three different instructional sets regarding how attention should be prioritized (no-priority, gait-priority, texting-priority) in two different environments (low-distraction, high-distraction). Participants performed two trials for each condition. Trial-to-trial repeatability of gait speed, texting speed, texting accuracy, and the relative dual-task effects (DTE) on each was examined using intraclass correlation coefficients and standard error of measurement. MDC95 scores were also computed for each performance measure. Among young adults, reliability of gait speed in a challenging dual-task situation is excellent, even in a high-distraction environment. In the absence of specific task prioritization instructions, changes in dual-task gait speed greater than 0.15m/s or 11.9% DTE represent real change. Reliability of the more novel, non-gait task has poor to good reliability. Dual-task effects are more reliable when participants are given specific instructions about how to prioritize their attention. The findings also suggest that reliability of dual-task performance in a novel or challenging task is greater when individuals are more skilled at the task. Implications for clinical assessment of dual-task performance are discussed.