The benefits of exercise for health are becoming increasingly clear. Wheel running is often used in the laboratory to study the effects of enhanced activity levels in rodents. The biological significance of wheel running remains elusive, however, making the behavior difficult to interpret. Some have argued that wheel running is not a natural behavior but an artifact of captivity and that it may be a sign of neurosis or stereotypy.

Johanna Meijer and Yuri Robbers (Leiden University Medical Centre, the Netherlands) sought to address these concerns by studying whether wheel-running behavior is exhibited voluntarily by animals in the wild. The team placed a running wheel with automatic movement detection, a passive infrared motion sensor, a camera equipped with night vision and a food tray to attract animals inside a cage-like structure that could be entered by any animal up to the size of a rat. Cages were placed in two locations where feral mice live: a green urban area and a dune area not accessible to the public.

Meijer and Robbers collected video at these sites for more than 3 years and analyzed more than 12,000 video fragments in which wheel movement was detected by the motion sensor. The wheel was moved in both locations not only by mice but also by shrews, rats, snails, slugs and frogs (Proc. Royal Soc. B 281, 20140210; 2014). Mice and some shrews, rats and frogs that manipulated the wheel were seen to leave the wheel and then enter it again within minutes, indicating that wheel running was intentional for these animals.

Wild mice ran in the wheels for lengths of time similar to those reported in captive mice. The median running speed of wild mice was lower than that of their captive counterparts, but wild mice reached higher maximum speeds. When food was no longer provided at the urban site, the number of visits to the recording site decreased, but the fraction of visits that included wheel running increased, suggesting that the animals found the activity inherently rewarding.

Research into the health effects of exercise depends on the use of running wheels, and for such research, it would be potentially problematic if wheel-running behavior was stereotypic rather than elective. But this study suggests that wheel running is in fact a voluntary behavior. The authors wrote, “Our findings may help alleviate the main concern regarding the use of running wheels in research on exercise.”