Neurophysiologic theory and some empirical evidence suggests that fatigue caused by physical work maybe more effectively recovered during “diverting” periods of cognitive activity than during passive rest; a phenomenon of great interest in working life. We investigated the extent to which development and recovery of fatigue during repeated bouts of an occupationally relevant reaching task was influenced by the complexity of a cognitive activity between these bouts. Eighteen male volunteers performed three sessions, consisting of six 7-min bouts of reaching alternating with 3minutes of a memory test differing in complexity between sessions. Throughout the session, recordings were made of upper trapezius muscle activity using electromyography (EMG), heart rate and heart rate variability (HRV) using electrocardiography,arterial blood pressure, and perceived fatigue (Borg CR10 scale and SOFI). A test battery before, immediately after and 1 hour after the work period included measurements of shoulder elevation strength (MVC), pressure pain threshold (PPT) over the trapezius muscles, and a submaximal isometric contraction. As intended, perceived fatigue, HRV, and EMG amplitude increased during the physical work bouts. Recovery did occur between the bouts, but fatigue accumulated throughout the work period. Neither EMG changes nor recovery of perceived fatigue during breaks were influenced by cognitive task complexity, while heart rate and HRV recovered the most during breaks with the most difficult task. Recovery of perceived fatigue after the 1 hour work period was also most pronounced for the most difficult cognitive condition, while MVC and PPT showed ambiguous patterns, and EMG recovered similarly after all three cognitive protocols. Thus, we could confirm that cognitive tasks between bouts of fatiguing physical work can,indeed, accelerate recovery of central components of fatigue, even if benefits may be moderate. Our results encourage further research into combinations of physical and mental tasks in an occupational context.