PSYETA LogoCopies of this journal are no longer available for sale, but our other two journals, Society & Animals and the Journal of Applied Animal Welfare Science, are available and subscriptions are quite affordable. They can be ordered online via our secure order page.

A Home For A Mouse


Author describes the manner in which mice perceive and experience their world and how their perceptions differ from our experience and perceptions. She discusses how to care for mice in order to ensure a humane quality of life based on their needs. Areas discussed include mice posture and movement patterns, their sensory world, social life, and how these considerations should dictate the design of a perfect mouse house.
mice, cages, animal behavior, basic needs, animal welfare, animal housing, environmental enrichment, laboratory animals, husbandry, living conditions, humane
About the Author


Mice have always lived with people, mostly as unwelcome guests, more rarely as pets. In the last century they have taken on a new role as by far the most frequently used laboratory mammal. Their wild cousins make an occasional laboratory appearance but it is mus musculus (the house mouse) who lives and dies, quite literally in millions, in the laboratories of the world.

As laboratory animals, the status of mice is unique. The mouse is the only mammal which has been systematically bred to produce dozens of inbred strains that preserve and "fix" mutations giving a great variety of physical, physiological, and behavioral characteristics. Selective inbreeding has made these strains nearly genetically homogeneous. There are grey mice, brown mice, white mice, piebald mice, hairless mice, diabetic mice, waltzing mice, blind mice . . . the list is long and grows year by year.

Despite the great differences between strains, laboratory mice have remained very much mice; they will easily go feral and breed with wild mice with scarcely a backward glance. It is therefore well worth examining species characteristics which can guide us in devising comfortable conditions for them. Since we owe them a great debt for their involuntary contribution to science and medicine, it would be proper for us to ensure that their quality of life is as good as we can reasonably make it. This is made somewhat easier because laboratory mice are not very interested in people, but are passionately interested (for good or ill) in other mice.

The purpose of this paper is to consider the way in which mice perceive and experience their world, and in particular, to consider how that differs from our own experience and perception. If we try to be "nice" to mice using only our own preferences as a guide, we are likely to be unintentionally cruel. In order to understand how to care for mice it is useful to know how they move, how they perceive things, and the nature of their social life.


An ethogram of a species describes the innate patterns of movement and the invariant activities which characterize the species. The ethogram of the mouse includes the fact that although a mouse is usually quadrupedal; standing, running, and walking on four feet, he/she also frequently adopts a bipedal posture especially when eating, grooming, fighting or just " orienting." The anatomical structure of the mouse facilitates this variation in stance: the back legs are stronger than the forepaws, the tail is large and strong. When a mouse adopts a bipedal stance, he/she is, in fact, supported as much by the tail as by the hind legs, giving the animal a stable tripod as the base. The orienting response, in which the fully alert mouse rears onto his or her hind legs, head up and momentarily still, ears erect [Plate 1] is a behavioral invariant; it is a posture that differs little from one occasion, or one mouse, to another.
Plate 1 - A mouse in the orienting position.
When there is insufficient head room for the mouse to become erect, a more limited version of this posture occurs. Now the mouse stops with his or her ears erect, and usually holds one forepaw off the ground while he/she attends to the change in the environment which has elicited the response. Mice should be caged in a manner which allows the full bipedal stance. A mouse will usually eat in bipedal position, with the food held in the forepaws. This is a much less erect stance than that adopted in the orienting response, and requires much less headroom than the latter.

When a mouse is running freely, the tail acts as a balancing organ fully extended behind the trunk with the last one or two centimeters turned up in a vertical position. A mouse running in this manner is confident. When afraid, the stomach is close to the ground and the tail depressed. Even a small cage will allow a maturing mouse pup to run short distances. In the same cage, the adult mouse will not be able to run but retains the ability when offered sufficient space.

Mice are good jumpers and climbers and will climb anything with a surface rough enough to give them a hold: cloth, brickwork, and wire grids are only a few examples. A cage with a wire grid lid offers the mouse opportunities for climbing exercise which mice appear to welcome. Mice can usually jump at least a meter between two fairly even surfaces, and much greater distances when jumping down from one level to another.

Mice may rest lying on their sides, but more often they go to sleep with their feet tucked under their body with their head down. As they go into deep sleep they roll over onto their sides. If they are cold, mice will curl up and huddle together with other mice. If they are too hot, they will stretch their bodies out when they rest or sleep and move a few centimeters away from their cage mates. Some mice appear more sociable than others, typically sleeping in a heap, at least partially because their coats are thin and they feel the cold more than better provided mice. In this limited sense the apparent sociability of all mice can be controlled by varying the ambient temperature.

The collecting of nest material and the building of a typical nest are apparently unlearned behaviors. Mice born in wire-floored cages without bedding will collect suitable nest material and assemble it when given the opportunity to do so. Although mice do not need nest material in order to live or reproduce, I have observed the frantic efforts they will make to obtain bits of straw, cloth or paper. I find that nesting material adds greatly to their comfort if it is made available to them as a matter of routine.

Patterns of fighting, mating, parturition, and suckling all contain invariant elements which form part of the full species ethogram. But I have found those patterns not quite so critical to good caging as those which have been more fully described here.


Mice in their natural state are crepuscular or nocturnal. That is, they are active at dusk and during the night, and quiet during the day light hours. Their eyes are adapted to this pattern of life. Nearly all mice dislike bright light. Even those with pigmented eyes will usually avoid it, while albino mice with pink eyes will find bright light quite blinding. If albino mice are kept in bright light, their eyes and their visual cortex can be damaged (Clough, 1988). There are two ways of making the light level comfortable for them: one is to keep the level of ambient light in the cage at no more than 60 Lux (Clough, 1988) and the other is to make sure that they are caged in such a way that they have a dark place into which they can retreat. Ideally, they should be offered a combination of the two.

Because the retina of the night-adapted eye of the mouse is largely composed of rod cells, they probably have little or no color vision. They do not distinguish the red end of the spectrum. They may be able to make some color distinctions at the blue end of the spectrum, but that issue remains undecided. If you want to get some idea of what that sort of color vision is like, try fixating a point well to the left or right of a small red light. You will see, in peripheral vision, the red light appears to be a dull white light.

The position of the eyes in a mouse's head [Plate 2] means there is little overlap of the left and right visual fields.

Therefore, effective depth perception depends on head movement. The position of the eyes allows the mouse, when standing on four feet, to see things overhead which is very important for a small animal whose predators include such birds of prey as the owl.

Mice will give a sharp startle response to "visual loom" (a shape descending from above or in front of the mouse). If you want to pick up a mouse he/she will jump away in fright if you drop your hand from above. The better way is to introduce your hand gently at an oblique angle so that the mouse can see you and will not be startled.


Mice have, relative to their bodies, very large mobile ears. Hearing is an important sense for mice and one that differs even more sharply from human sensation than does vision. Very roughly, humans can hear sounds in the range 200 Hz and about 16 kHz (the most important sound range for human speech is below 6 kHz). The sounds in the frequency range over 18-20 kHz are generally referred to as ultrasonic, that is beyond the range of human hearing. Mice do not hear the lower frequencies of the human audible range but they do hear sounds in the range 10-70 kHz and may hear sounds in excess of 100 kHz! Their most important emitted sounds are all in the ultrasonic range. They do make some sounds in the sonic range: high pitched sounds often described as "squeaking." The existence of these noises has led to the mistaken assumption that these are the only sounds they make. In fact, with the exception of warning sounds, all the important mouse calls are in the ultrasonic range. The best documented example is the pup distress call emitted at about 40 kHz when mouse pups are chilled or disturbed (Sales and Pye, 1974). Other important calls are mating calls and submission calls, both of which are in the range 20-30 kHz. The pup distress call has a powerful effect on both male and female adult mice. They will search for and "retrieve" the pup if he/she is out of the nest, or attend to the nest if it has been disturbed. The mouse who responds to the call does not have to be the parent. Any adult mouse will respond to the distress call of any pup.

There is good reason for the distinction between the high and low frequency calls; generally, low frequency sounds carry much further than high frequency sounds. A low frequency pup call might make it easy for a predator to find the nest, whereas an ultrasonic call, while it will attract nearby mice, will not be heard by a predator a few meters away. The other ultrasonic calls are essentially social messages to nearby mice. These need not carry far enough to put the mouse in danger. Audible mouse squeals carry further because of their lower frequency and so act as a general warning to other mice in the wider environment.

The extension of the mouse's hearing into the ultrasonic range has other consequences in a complex, manmade environment. Many pieces of domestic and industrial equipment emit sounds in the ultrasonic range, inaudible to us, but often distressing the mice, other rodents, and occasionally to dogs and cats as well. A few examples must serve to make the point that we should be careful to screen the environment in which we keep mice for intrusive ultrasound. A tap dripping water into a stainless steel sink has a large ultrasonic component and will distress rodents. Animal room sinks should be made of ceramic or plastic so that a dripping tap produces no distressing harmonics. A bunch of keys rattled near a mouse will cause convulsions in a mouse of susceptible strain because of the ultrasonic elements in the metallic noise of which we only hear the harmless sonic jingling. Finally, it is useful to draw attention to a common menace in this call of sound, the TV set or the computer VDU. Do not put your companion mouse near your TV set when it is switched on, and if you have a computer outlet in your animal's room, turn off the screen except when you actually need to look at it. The TV or VDU screen emits a constant signal at 22 kHz, with harmonics in the 16-160 kHz range; inaudible to you, but distressing to mice and rats. By far, the best policy is to use either a "bat detector" or an oscilloscope to screen the environment for mechanical sources of ultrasound (Sales et al., 1992).


People often think mice are dirty because they have an unpleasant musty smell; this smell comes mainly from mature male mice and has nothing to do with cleanliness. Frequent cage cleaning will reduce the intensity of this odor but cannot entirely eliminate it. This smell is one of the many pheromones, or chemical messengers, that pervade the mouse world. It both allows males to avoid one another (which reduces the possibility of fights) and informs female mice where they may find a mate.

All mice have a "class" odor which indicates their age, sexual status, etc. They have a family odor by which they identify their parents and litter mates, and finally, an individual odor. The main sources of the odors are the anogenital area, the palms of the feet, and the mouth. Grooming spreads the saliva and its odor over the whole coat while the pheromones present in the urine serve to mark the territory of a mouse.

Pheromones are a powerful determinant of behavior in most mice. They are as important for recognition among mice as visual appearance is for humans. Mice will recognize their home cage and/or nest, their family members, strangers, females in oestrus, and so forth by their scent (McFarland, 1981). They also recognize other more generalized odors such as the smell of "fear." Bearing in mind their nocturnal way of life, it makes a great deal of sense that smell should be more important than vision as a source of information. It is not always easy for people to grasp the importance of pheromones because so little of the human brain is devoted to olfaction compared with the mouse brain.

The size of the olfactory lobes in the mouse brain, relative to the whole brain, suggests a world dominated by olfactory mes-sages [Plate 3].

Just as people like their homes to "look right," mice like their homes to "smell right." The most common method of cleaning mice cages (throwing out all the bedding and replacing it with clean material) disturbs mice considerably because it removes the acceptable home cage smell. A mouse cage like the Cambridge cage (Wallace, 1993) is advantageous because it allows the caretaker to clean the area of the cage used for excretion without disturbing the nest area, a process which disturbs the mice much less than the total removal of the bedding material.

It is advisable to cage litter-mates together rather than to mix even weanling mice in a random fashion. Fighting is much less likely to occur when all the cage mates have a common family odor. Strange mice, especially strange adult males, are liable to fight. Even a familiar mouse anointed with the odor of a male stranger, will elicit antagonistic behavior. It seems that social relations among mice are odor dependent to an overwhelming extent.


Taste is a sense of which the main component is actually odor. However, sweetness, acidity, sourness, and salinity affect what is eaten. Food is mainly attractive because of its smell and is then eaten if the taste is acceptable. Food eaten by the mother mouse while she is lactating is often the preferred food of the offspring. New foods are added to the diet only after very cautious tasting and exploration.

Although mice are omnivorous, they are largely grain eaters who have no need for either fresh vegetables or a varied diet. Almost any standard laboratory diet based on grain, a little protein, with suitable mineral and vitamin supplements, will keep them happy and healthy. They do have food preferences, and welcome an occasional tidbit, but by and large they are better fed on a good balanced, dry diet with a supply of clean water. The water supply can be more of a problem than the food, especially with companion mice. Open dishes of water are a disaster and many of the water bottles sold for use in cages are unsuitable because the activity of the mice will empty the bottle, making the whole cage wet and uncomfortable. For laboratory mice, water nipple systems can be satisfactory provided they are regularly checked, but they are expensive and impractical with the most commonly used type of cage. The water bottle devised by Wallace (1993) is probably the best available. It is not the purpose of this paper to deal with the mouse's dietary requirements, but this note is included for the sake of completeness in case any reader is in doubt about the needs of a companion mouse.


Mice share with other animals sensitivity to touch. There is one characteristic which enlarges the sense of touch beyond the body surface and that is the whiskers. The importance of these whiskers is an attribute they share with other rodents and possibly with cats. A mouse's whiskers are large in relation to the whole mouse and form a considerable extension of the face. They serve as a sort of "poor man's radar" which allows the mouse to avoid obstacles and run freely in places where the light levels are too low for vision to be of any value.


Mice are gregarious creatures to whom other mice are a source of dominant interest. They can also be savage fighters. The house mouse lives naturally in loose aggregates, spaced out by occasional antagonistic displays and territorial scenting. The closest approximation to this state in captivity is colony breeding. The colony would consist of a mixed-sex, mixed-age group, and works well enough provided there are enough nesting sites and sufficient space. It was shown as long ago as the 1930s that if the population of a colony increased to an unacceptable level, social organization broke down and the whole colony developed into a "behavioral sink" comparable to that described by Calhoun (1962) in rat colonies.

It is more common to cage mice in either like-age, like-sex groups, breeding pairs, or trios. Cage groups vary in number between pairs or as many as 30 mice in a single cage. It is less common to cage mice singly, although that is occasionally necessary, especially with mature males. In this context it must be borne in mind that provided a mouse can hear and smell conspecifics, he/she is not truly isolated. The constraints which govern the establishment of acceptable cage groups are age, sex, and genetic derivation. Mice from some inbred strains are quiet, peaceable animals that show little inclination to fight, while mice from other strains are assertive, pushy animals who readily become aggressive (Scott and Fredrickson, 1951). Mature male mice from these latter strains are better caged singly or with a sterile female when they are not being used for breeding. Litter- mates, caged together from weaning, seldom fight seriously unless they come from a particularly bellicose strain. Stranger mice who meet on neutral territory, though they take an interest in one another, do not usually attack one another. It is when they are caged together that strange mice are apt to fight because the weaker animal cannot run away when threatened and there is territory to be contended. An ample supply of food and water make little difference to these behaviors, which seem to be triggered by pheromones rather than competition for scarce resources (Box, 1973).

I have observed that large groups of mice are socially unstable. Once a group exceeds five or six mice, stable inter-animal relations seem to break down although mutual tolerance may still be displayed, especially in groups of young mice. In smaller groups, mutual grooming serves to spread the individual odors present in saliva throughout the group. This may serve to reduce agonistic encounters.

Young mice need cage mates in order to engage in play activities which are important for development (Box, 1973). Such play mostly takes the form of mock fighting and chasing and is not commonly seen in sexually mature mice.


All laboratory mice and most companion mice will spend their lives in some form of restricted enclosure or cage. If this cage is to be an acceptable home, and not a mere prison, it is important that we design it correctly. A great deal more time and money has been spent on arguing about caging methods than on observation and research. Such caging "standards" that exist for mice owe more to custom than to science. There are three ways in which we can improve on these standards. 1.) Compare the growth, survival, and health of mice caged in different ways.

2.) Examine the ethogram and the known sensibilities of mice and make sure the cage conforms with these.

3.) Conduct preference tests to determine which cage designs are most acceptable to mice. A difficulty with the first method, which has been explored from time to time is that mice seem able to thrive in such a great variety of containers. The other two methods are less commonly applied and are worth a brief consideration here. 2.) If we look at the structure of the mouse and the mouse's species ethogram, it can be strongly argued that a cage should offer sufficient head room for a mouse to adopt the full orienting stance (Lawlor, 1982).

  Equally, it can be argued that a cage should be large enough to permit the mice to run freely, although adults seem to suffer few ill effects from living in cages in which this is impossible. Still, it would seem to be a minimum humane requirement that the mouse is able to run freely in a cage without contortion of his or her body. Such a cage needs to be at least double the length of the mouse from its nose to the tip of the tail and at least as wide as the length of the mouse.

  Cages with solid floors suit the feet of mice better than wire floors which impact the feet in an abnormal manner and may cause foot damage.

3.) There is little in the way of published work involving preference testing for cage features. In these circumstances, it is worth making brief reference to some unpublished results we obtained when we offered mice a choice of nest site conditions. In ordinary commercial mouse cages, mice typically nest and sleep under the food rack. The food rack has three features: it reduces the light level in the cage, it provides a low ceiling area just above the back of a mouse in the feeding posture, and it provides some heat insulation when the rack is full. The last attribute (heat insulation) is likely to be important for mice with neonate pups, who try to maintain a nest temperature of 34 degrees C. Using trios of like-sex adult mice who were not breeding, we found that, when offered a choice, a nest area with a solid low ceiling was preferred to one that was merely dark, and that added overhead insulation did not make the nest site more attractive. Offered a choice, a nest area with a solid front was preferred to one with an open front and finally, they preferred a nest site with a corner, floor-level opening to one with
an opening in the center of the solid front. This is enough
to suggest that further preference testing might prove
rewarding. To summarize, a satisfactory cage should have the following features: 1. Enough space for the mice to adopt a bipedal stance, turn freely, walk, climb, and if possible, run.

2. A nest area which has a lower ceiling and is effectively separated from the rest of the cage.

3. Cage companions, preferably no more than five or six.

4. At least a partly solid floor and bedding material.

5. Light level in the cage no more than 60 Lux if the mice are albino, and no more than 350 Lux if they are pigmented.

6. Not be placed near a source of ultrasound.

7. Acceptable air exchange, correct humidity level, and free from draughts.

8. Clean water available at all times in a container which will not flood the cage.

9. An adequate, balanced diet. Perhaps the most important guiding principle is to devise a form of caging that is compatible with the way the mouse experiences the world and to avoid the simple assumption that because a cage looks good to us (because it is clean, bright and spacious) it would seem good to a mouse. The mouse might well prefer a place which is smelly, dark, and claustrophobically cozy.


Box, H. (1973). Organization in Animal Communities, 7, 160-168. Butterworth, London.
Calhoun J. B. (1962). Population density and social pathology. Sc. Amer. 206(2), 139-48.
Clough, G. in Beynen A.C. & H. A. Solleveld, (Eds.) (1988). Housing and Welfare of Laboratory Rodents. New Developments in Biosciences: Their Implications for Laboratory Animal Science. Martinus Niojhoff, Dordrecht.
Lawlor M.M. (1984). Behavioral Approaches to Rodent Management. Standards in Laboratory Animal Management, University Federation for Animal Welfare, Potters Bar, U.K.
McFarland, E. (1981). The Oxford Companion to Animal Behavior. Oxford University Press, 455-466.
Sales, G.D., Milligan, S.R. & Khirngkh, K. (1992). The acoustic environment of laboratory animals: a report to the RSPCA. University Federation for Animal Welfare, Potters Bar, U.K.
Sales, G.D. & Pye, D. (1974). Ultrasonic Communication by Animals. Chapman and Hall, London.
Scott, J.P. & Fredericson, E. (1951). The causes of fighting in mice and rats. Physiol. Zool. 24, 273-309.
Wallace, M. A Mouse-friendly cage compared experimentally with a person oriented one. (1993). Hum. Innov. & Alt. 7,

Monica Lawlor

National Health Service

  Monica Lawlor is a psychologist who taught psychology at Bedford College, University of London and occasionally at the University of Western Ontario. In 1982, she took early retirement when the College amalgamated with Royal Hollooway College. Since then she has divided her time between research and clinical psychology, working in the National Health Service with people who have severe learning disablities.

  Graduating from the University of London in 1948, and starting work as a University Teacher in 1949, her training and outlook predated the Behaviorist white-out of the 50s and 60s. She has always been as much concerned with experience as behavior. She believes that unless you try to understand what a behavior means to the person or animal who carries it out, you may be able to manipulate it, but you will be a technician rather than a scientist.

  Her main research interests have been in animal behavior, exceptional children, experimental aesthetics, and the psychology of religion with a continuing theoretical interest in psychodynamic psychology.

  Dr. Lawlor has twice chaired the British Psychological Society Psychotherapy Section. In the last decade, most of her work has been on humane caging conditions for rodents and the treatment of self injurious behavior in people with learning disabilities.

  She has an interest in the Philosophy of Science and Scriptural Theology and writes in both subjects.

  In her spare time, Dr. Lawlor reports, "I enjoy the company of my friends, reading, gardening, and making things, and I am an active member of the Roman Catholic Church."

PSYETA LogoCopies of this journal are no longer available for sale, but our other two journals, Society & Animals and the Journal of Applied Animal Welfare Science, are available and subscriptions are quite affordable. They can be ordered online via our secure order page.