Wild mice actually enjoy running on exercise wheels | @GrrlScientist

Research findings dispel the idea that wheel running is a product of captivity, indicative of a neurotic or repetitive stereotyped behaviour that may be associated with poor welfare

by GrrlScientist for The Guardian | @GrrlScientist

According to a newly published study, wild mice frequently and voluntarily run on an exercise wheel if provided access to them in nature, even in the absence of a food reward. Further, the length of running bouts by wild mice matched those of captive mice. These findings dispel the idea that wheel running is a product of captivity, indicative either of neurosis or a mindless repetitive behaviour that may be associated with poor welfare or close confinement.

Activity wheels come in a variety of sizes and styles, but their basic purpose is to provide confined animals with the opportunity to exercise, as you see these adorable dwarf hamsters doing here:

Reading on a mobile device? Here’s the video link.

That’s an amusing video, and the hamsters appear to be enjoying themselves, but are they really? And how might you figure this out?

I ran across an interesting little research paper that addresses a fundamental — and controversial — question in exercise physiology: is running in an exercise wheel as exhibited by small pets, like hamsters, mice and rats, an artefact of captivity? Why do they do it? Is it intended to alleviate stress or neurosis caused by close confinement, is it a repetitive and invariant behaviour that is devoid of any obvious goal or function, such as cage-pacing seen in some zoo animals, or might it signify something else?

“When it comes to stereotyped behaviour, there are competing theories. One issue, as an example, is whether stereotyped behaviour is a symptom of bad welfare that should be prevented, or whether it is a coping strategy that actually increases welfare”, explained the study’s co-author, Yuri Robbers, in email.

Mr Robbers, who works as a grammar school biology teacher, has a master’s degree in animal behaviour and is currently researching animal behaviour, theoretical biology and ecology as he works towards his PhD under the mentorship of neurophysiologist Johanna H. Meijer, a professor in the Department of Physiology at the Leiden University Medical Centre in The Netherlands.

The controversy swirling around running on an activity wheel worried the Dutch researchers so much that they designed several experiments to clarify the basis of this behaviour. Specifically, they wanted to test whether wheel running fulfils the criteria for a stereotyped behaviour: (1) it occurs only in captive animals, (2) it is repetitive, invariant and devoid of obvious goal or function, (3) if it consists of natural behavioural elements, these elements occur at higher rates and for longer durations than found in nature, and (4) it is partially or not at all dependent on external stimuli.

Will wild mice use a running wheel if one is provided in nature?

Famous animal behaviourist Konrad Lorenz once remarked that rodents that had either escaped or been released will enter and run on exercise wheels if one is accessible to them (as cited here: doi:10.1126/science.155.3770.1623). Intrigued, Professor Meijer and Mr Robbers decided to follow up on this observation by going one step further: they asked whether free-living animals that had never before seen an activity wheel would voluntarily use one if it was accessible to them.

They designed a cage that specifically excluded large animals (so they wouldn’t knock over the exercise wheel set up). Inside this cage, which could be freely accessed by small animals, they placed a running wheel along with some food intended to attract mice. These exclosures were set up in nature at two different field sites; a green urban area (Professor Meijer’s back garden) and a dune area that was inaccessible to the public (see below):

Every visit to the experimental set-up was recorded by a night-vision camera, using passive infrared motion detection. At night, the camera relied upon infrared light (infrared light is invisible to mice), which did not interfere with motion detection.

Data were collected for more than three years using this experimental set-up (green urban area data were collected from October 2009 to February 2013, and dune area data were collected from June 2011 to January 2013; figure 2):

The running wheels proved popular with a variety of free-living animal species. During a time period of more than three years, Professor Meijer and Mr Robbers made more than 200,000 recordings of animal visitors and analysed more than 12,000 video fragments in which wheel movement was detected.

In the first 24 months, the urban green area resulted in 1011 animal visits (of which 734 were mice; data video). The first 20 months in the dunes resulted in 254 animal visits (of which 232 were mice). Although the cage was baited with food specifically intended to attract mice, other animals — shrews, rats, snails, slugs (data video), and frogs (data video) — also stopped in to do some running. (Or sliming?) Interestingly (to this ornithologist), even small birds popped in occasionally, although they were never recorded running on the wheel. (I suspect they were laughing at the mice.)

“Our data indicates that wheel running occurs in nature, being performed by free-living wild animals”, said Mr Robbers. Thus, “captivity, extended or otherwise, cannot possibly be the cause of wheel running”. Since all the experts agree that stereotypical behaviours only occur in captivity, this finding indicates that wheel running does not fit well into the established criteria.

In view of the wide variety of diets preferred by these other species of animal visitors, it would appear they were not coming solely for the food. So of course, Professor Meijer and Mr Robbers then asked what would happen if they stopped providing any food at all — might these animals still pop in for a run on the exercise wheel?

Will free-living animals use a running wheel without a food reward?

Although Professor Meijer and Mr Robbers stopped providing food in the urban area enclosures for more than a year (October 2011–February 2013), animal visits continued. Professor Meijer and Mr Robbers observed 78 wheel running visits (62 mouse visits — 36 of which were very small mice, indicating they were too young to know the cages had previously been baited with food).

The data revealed that the number of visits to the activity wheel dropped significantly as soon as the food was removed, but visits that included wheel running actually increased by 42 percent, suggesting that the animals (wild mice, mostly) were visiting the cage specifically to run on the activity wheel, that exercise is rewarding in itself. The experimental device acted as a sort of neighbourhood mouse gym, it would appear.

Were activity bouts comparable between wild and lab mice?

Another test of whether wheel running fulfils a behavioural stereotypy was to compare wheel running activity between wild and lab mice. To get a clearer picture of the animals’ activity patterns, Professor Meijer and Mr Robbers compared wheel-running data from their wild mice to wheel-running data from laboratory mice that had been collected previously by another researcher (doi:10.1126/science.155.3770.1623, see figure 3):

The data reveal that most of the wheel running wild mice were juveniles, and 20 percent of them had running bouts that lasted longer than one minute, with the maximum bout lasting 18 minutes. This is similar to activity bouts reported for 200-day old lab mice.

The wild mice were actually working hard, too: although the average speed of running was slightly less than for lab mice (1.3 versus 2.3 km/h), the maximum running speed recorded for wild mice was higher than the maximum for lab mice (5.7 versus 5.1 km/h).

Why do wild mice voluntarily run on exercise wheels?

This study shows that wheel running is voluntary in wild mice, is not dependent upon a food reward, and occurs in bouts that are comparable to those recorded for captive mice, so it does not satisfy the established criteria for a stereotypical behaviour, as some scientists have argued. Of course, this raises the question: why do they do it?

“We’re considering play behaviour as a viable explanation, and have started follow-up experiments to test that hypothesis”, said Mr Robbers in email.

Wheel running looks rather boring (suggestive of my least favourite piece of gym equipment, the treadmill), but perhaps these animals perceive an activity wheel similarly to humans when given the option to climb musical stairs instead of riding an escalator, as we see in this video?

Reading on a mobile device? Here’s the video link.

Why do we care about mice and their exercise wheels?

As anyone who reads the newspapers or listens to the radio knows, we are daily told to get more exercise. Basically, daily physical activity slows ageing, reduces the incidence of many cancers, helps maintain a reasonable body weight, reduces the incidence of diabetes, heart attack and stroke, and improves brain function — and that’s just to name a few benefits that I’ve read about this week. Research into the human health benefits of physical activity depend upon the use of running wheels used by lab animals. If this wheel running is a stereotypy, this could be problematic.

“Some scientists … have had a tendency to dismiss research findings based on the use of running wheels outright”, explained Mr Robbers in email. “This is rather more difficult, now that our data is there to suggest that wheel running can occur in free living animals.”

Reading on a mobile device? Here’s the video link.


Meijer J.H. & Robbers Y. (2014). Wheel running in the wild, Proceedings of the Royal Society B: Biological Sciences, 281 (1786) 20140210–20140210 | doi:10.1098/rspb.2014.0210

Also cited:

Sherwin C.M. (1998). Voluntary wheel running: a review and novel interpretation, Animal Behaviour, 56 (1) 11–27 | doi:10.1006/anbe.1998.0836

Kavanau J.L. (1967). Behavior of Captive White-Footed Mice, Science, 155 (3770) 1623–1639 | doi:10.1126/science.155.3770.1623 [Free PDF]


Love science? Me too! Follow me on Medium for more like this.

.. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. ..

GrrlScientist is very active on twitter @GrrlScientist and you can follow all her writing by subscribing to her TinyLetter

Originally published at The Guardian on 21 May 2014.

Like what you read? Give GrrlScientist a round of applause.

From a quick cheer to a standing ovation, clap to show how much you enjoyed this story.