Barbering in Mice: Welfare Significance and What an IACUC Protocol Should Require

Barbering is the stereotypic behavior in which one mouse (the “barber”) plucks the fur and whiskers of its cagemates, and sometimes its own, leaving characteristic patches of symmetric hair loss around the muzzle, eyes, and body. It is one of the more common abnormal repetitive behaviors (ARBs) reported in laboratory mouse colonies, and it matters on two separate axes: it is a recognized indicator of impaired welfare and abnormal social dynamics, and, independently of that, it can confound any study that depends on an intact vibrissal (whisker) system or on an unbiased, evenly distributed experimental population. An IACUC-approved husbandry and monitoring protocol needs to address both.

What barbering is

Barbering is a derivative of normal allogrooming that has become excessive and one-directional. Sarna, Dyck, and Whishaw’s foundational study of the behavior in C57BL/6 mice — “The Dalila effect: C57BL6 mice barber whiskers by plucking” — described the barber grasping individual whiskers with the incisors and plucking them out during what otherwise looks like mutual grooming, with the recipient passively tolerating it. Barbers in that study tended to be heavier than the mice they barbered, consistent with a social-dominance mechanism rather than random aggression. A larger epidemiological survey by Garner, Weisker, Dufour, and Mench (Comparative Medicine, 2004) of roughly 2,950 mice found the behavior was female-biased, typically had its onset around puberty, and followed a consistent topography — hair removed preferentially from the scalp, around the eyes, and the genital region — closely paralleling human trichotillomania. That parallel is well established enough that barbering is used deliberately as a spontaneous model of trichotillomania and obsessive-compulsive spectrum disorders in some behavioral-neuroscience protocols; this guide addresses it as the far more common incidental colony-management and welfare issue that shows up in cages never intended to model anything.

Why it matters for welfare

Barbering is treated as a welfare indicator, not a cosmetic curiosity, for two reasons. First, its association with dominance hierarchies means a barbered cage is a cage with unresolved or unstable social dynamics — the recipient is tolerating repeated, unwanted handling from a cagemate it cannot avoid in a standard shoebox cage. Second, barbering can escalate. What begins as hair and whisker removal can progress to skin trauma, and persistent plucking at the same site can produce dermatitis requiring veterinary intervention. A monitoring protocol has to be able to tell early, low-grade barbering (a husbandry observation) from barbering that has crossed into self-trauma or interanimal injury (a clinical and potentially a humane-endpoint issue) — see CASRAI’s guide to the Mouse Grimace Scale for one validated tool used alongside body-condition and coat-condition scoring to make that call. Under the Guide for the Care and Use of Laboratory Animals, animal environment and management is one of the core sections institutions are assessed against, and abnormal behavior is squarely inside that scope, independent of whether it also threatens a particular study’s data.

Why it matters for study validity

Barbering is not only a husbandry finding — it is a real experimental confound in any protocol that uses a whisker-dependent assay. Mice rely heavily on their vibrissae for tactile exploration, texture discrimination, gap-crossing, and object investigation, and several commonly used behavioral paradigms (open-field exploration, novel-object recognition, and other assays where thigmotaxis or active whisking drives the measured behavior) assume an intact vibrissal system across the cohort. Sarna and colleagues found that despite grossly normal cortical barrel fields, barbered mice showed altered dendritic morphology in the corresponding barrel cortex — evidence that whisker loss from barbering is not merely cosmetic but has a measurable central-nervous-system correlate. Practically, that means: if barbering is unevenly distributed across treatment groups — more common in one cage, one room location, or one genetic background than another — it becomes an uncontrolled between-subjects variable that can bias results in exactly the direction ARRIVE 2.0 randomization and blinding requirements are meant to prevent. A protocol using a whisker-dependent readout should specify how barbering status is checked, recorded, and accounted for before animals are allocated to groups, not discovered after the fact when a result doesn’t replicate. See CASRAI’s ARRIVE 2.0 guide for the reporting side of this problem.

Cage-level contributing factors

The largest dataset on this to date is Ratuski, Theil, Ahloy-Dallaire, and colleagues’ 2025 Scientific Reports study, which tracked 2,544 cages (7,007 mice) over a full year specifically to identify barbering risk factors. Their analysis identified sex, breeding status, genetic background, and housing system as significant predictors of barbering prevalence, and it also found a spatial “hotspot” clustering effect — barbering was not evenly distributed across a facility but concentrated in particular locations, along with measurable seasonal variation. Three practical factors follow from the existing literature and are worth building into a husbandry SOP:

• Strain predisposition. C57BL/6 and related substrains are the most consistently implicated genetic background in the barbering literature (both the Sarna and Garner studies used C57BL/6 mice), which argues for strain-aware baseline expectations rather than treating any hair loss on this background as automatically abnormal or automatically benign.
• Social hierarchy and group composition. Because barbering tracks dominance status rather than random aggression, group stability matters more than group size alone — introducing or removing cagemates, or housing unfamiliar adults together, is a plausible trigger for the dominance-testing behavior that barbering reflects.
• Housing system and environmental enrichment. Ratuski et al.’s finding that housing system itself predicts prevalence supports auditing cage type, group size, and enrichment provision as a colony-level variable rather than assuming enrichment alone will resolve the behavior — the evidence for enrichment as a reliable fix is mixed, and a protocol should treat it as one lever among several, not a guaranteed remedy.

See CASRAI’s Mouse Husbandry guide for the broader housing, group-size, and enrichment standards this sits inside.

Detecting and monitoring barbering

Because barbering clusters unevenly — by cage, by room, by season — per-cage observation catches what a room-level or colony-level summary can miss. Routine husbandry checks (daily observation is the baseline expectation under PHS Policy and AAALAC-accredited programs) should include a coat- and whisker-condition check alongside standard body-condition scoring, with findings logged in the cage or colony health record rather than only escalated verbally. That written record is what lets an attending veterinarian or IACUC distinguish an isolated, stable case from an emerging colony-wide pattern, and it is the data an investigator needs to demonstrate, after the fact, that barbering status was tracked rather than assumed.

What an IACUC-approved protocol should specify

A protocol that houses mice in group cages — which is most protocols — should specify barbering-related detection and response explicitly, not leave it to informal husbandry judgment:

• Monitoring frequency and documentation. How often cages are checked for hair/whisker loss, who checks (species-specific IACUC training for husbandry and research staff should cover recognizing barbering versus dermatitis, mite infestation, or fight wounds), and where findings are recorded.
• Escalation criteria and veterinary consult triggers. Defined thresholds — extent of hair loss, presence of skin trauma, or recurrence after a management change — that require notifying the attending veterinarian rather than continuing to monitor informally. Differentiating barbering from dermatitis, dermatophytosis, or interanimal aggression is a clinical call, not a technician judgment call, once the picture is ambiguous.
• Humane endpoints. Criteria for when barbering that has progressed to self-trauma or open skin lesions triggers a defined endpoint, consistent with the AVMA Guidelines for Euthanasia where euthanasia is the endpoint in question.
• Management response and 3Rs refinement. What happens to an affected cage — regrouping, environmental modification, or, only as a documented last resort, single housing (itself a welfare exception requiring separate justification, since social isolation is its own recognized stressor for a social species). This is squarely a 3Rs Refinement question, and IACUC review should treat it as one.
• Data-validity handling. For any protocol using a whisker-dependent behavioral assay, how barbering status is checked before group allocation, and how a barbered animal identified mid-study is handled — excluded, re-randomized, or flagged and analyzed as a covariate — documented prospectively rather than decided post hoc.

When it triggers a protocol amendment: a colony-wide barbering rate high enough to threaten the validity of a whisker-dependent assay, or a pattern that emerges only after animals are already allocated to groups, is the kind of finding an IACUC would expect reported as a protocol deviation or addressed through an amendment describing the revised monitoring/exclusion criteria — not something a PI quietly works around at the analysis stage. See CASRAI’s guides to the IACUC protocol and to IACUC and animal research oversight more broadly for how that review process works.

Frequently asked questions

Is barbering painful for the mouse being barbered?

Plucking itself does not appear to be overtly painful in the way an open wound is, and Sarna et al. described recipients as passively tolerating it during grooming interactions. That does not make it welfare-neutral: it reflects an unresolved dominance dynamic the recipient cannot escape in standard housing, and it can progress to genuinely painful skin trauma if unaddressed.

Can established barbering be reversed once a cage develops the behavior?

Fur and whiskers regrow once plucking stops, but the underlying social dynamic driving it does not necessarily resolve on its own. Management responses (regrouping, environmental changes) can reduce recurrence, but the literature does not support treating any single intervention, including enrichment, as a reliable fix in every case.

Does every barbered cage represent a welfare failure?

No — low-level, stable barbering in an otherwise healthy, socially settled cage is common enough in some strains that it is treated as a monitored finding rather than an automatic adverse event. The distinction an IACUC protocol needs to make is between stable, low-grade barbering and a pattern that is escalating, spreading, or accompanied by skin trauma.

How is barbering distinguished from mite infestation or fight wounds?

By pattern and clinical exam: barbering produces a relatively clean-edged, symmetric pattern of hair loss without the pruritus, skin lesions, or scabbing typical of ectoparasites, and without the puncture wounds typical of fighting. Where the picture is ambiguous, that differential is an attending-veterinarian determination, not a husbandry-staff judgment call.

Should barbered animals be excluded from a behavioral study?

Only where the protocol’s own pre-specified criteria say so. For a study with no whisker-dependent or sensory-exploration component, barbering status may not be a relevant covariate at all. For a study that depends on an intact vibrissal system, exclusion or covariate-adjustment criteria should be written into the protocol and applied consistently, in line with ARRIVE 2.0 randomization and reporting expectations, rather than decided case by case once the data are in hand.