The purpose of this study was to test the sensitivity and reproducibility of dual-mode actigraphy as an objective measure of functional performance in healthy adults under controlled levels of activity intensity. Twenty subjects wore the instrument on the nondominant wrist while performing standardized tasks selected to represent day-to-day activities at three levels of intensity (five tasks in each level): light (1-2 metabolic equivalents), moderate (3-4 metabolic equivalents), and heavy (4-6 metabolic equivalents). Upon completion of each intensity level, subjects were asked to rate their level of exertion using Borg's 15-point rating of perceived exertion (RPE) scale. Eighteen subjects repeated the protocol within 7 days. Zero-crossing and time-above-threshold modes successfully differentiated between light and moderate and between light and heavy activity. Reproducibility correlation coefficients (rs) across activity levels were .80 and .66 for the two modes, respectively. Results suggest dual-mode actigraphy may be useful for the study of performance variation and structure in healthy and chronically ill individuals.