Open Field Test in Mice: Arena Setup, Thigmotaxis, and IACUC Requirements

The open field test is one of the most widely used behavioral assays in rodent research, applied to measure locomotor activity and anxiety-like behavior in a single, relatively short session. Its popularity rests on simplicity: a mouse is placed in a novel open arena and allowed to explore while its movement is tracked. What the test actually captures, though, and how well it captures it, depends heavily on arena design, habituation practice, lighting, and testing order — details that are easy to standardize badly and hard to fix after the fact. This guide covers the methodology as published in the behavioral-neuroscience literature, the confounds most likely to distort results, and what an IACUC protocol needs to specify before this test can be approved.

What the open field test measures

The open field test yields two broad categories of readout from the same session. Locomotor activity is measured as total distance traveled, average velocity, and rearing frequency — a general index of motor function and exploratory drive, useful on its own and as a control measure for interpreting other behavioral assays (a drug or genotype that suppresses locomotion can produce false “anxiety-like” or “depression-like” results in downstream tests for the wrong reason). Anxiety-like behavior is inferred from the animal’s spatial distribution within the arena, specifically the balance between time spent near the walls versus in the open center.

The logic linking spatial distribution to anxiety rests on an approach-avoidance conflict: rodents are innately drawn to explore a novel space but are also averse to open, brightly lit, exposed areas where a predator could see them. An anxious or stressed animal resolves that conflict by minimizing time in the exposed center and maximizing time along the perimeter. This same open-versus-protected spatial logic underlies other rodent behavioral assays reviewed by an IACUC, including the Morris water maze, and it is why open field data is typically reported as both a locomotor measure and an anxiety measure rather than a single combined score.

Arena setup and standard dimensions

Published protocols vary in exact size, but the design principles are consistent. The arena is an enclosed, open-topped square or circular field with opaque, non-reflective walls high enough to prevent escape or jumping (commonly in the 30–40 cm range for mice) and a floor that is easy to clean and does not offer hiding cover. Reported floor dimensions for mice commonly fall in the 40 × 40 cm to 50 × 50 cm range for square arenas, or an equivalent-area circular arena; rat protocols scale up accordingly. The walls must be opaque — a clear-walled arena undermines the “open, exposed” conditions the test depends on and confounds thigmotaxis with simple visual barrier-seeking.

The arena is conventionally divided, for scoring purposes, into a center zone and a peripheral (wall) zone, most often by treating the innermost portion of the floor area as “center” and a fixed-width band adjacent to the walls as “periphery.” The exact boundary (a fixed distance from the wall, or a fraction of total floor area) differs across labs and tracking software, which is why a protocol document should state the specific center-zone definition used rather than assume it is self-evident — this matters for reproducibility and for comparing results against the literature, and it is one of the ARRIVE 2.0 reporting items relevant to behavioral assays (see the full ARRIVE 2.0 checklist).

Testing is usually conducted under automated video tracking (a camera above the arena feeding software that logs coordinates, distance, and zone time), which removes a substantial source of observer-dependent variability that affected earlier hand-scored protocols.

The habituation period before testing, and why it matters

Because the open field test relies on the animal’s response to a genuinely novel environment, habituation has to be handled at two separate levels, and conflating them is a common design error:

• Habituation to handling and the testing room is standard practice and does not compromise the assay: animals are typically transported to the testing room and left undisturbed for a period (commonly on the order of 30–60 minutes) before testing begins, so that transport stress and room novelty are not confounded with the arena itself. Handling habituation over the days preceding testing (brief, repeated gentle handling by the person who will run the test) similarly reduces acute handling-stress artifacts.
• Habituation to the arena itself is generally avoided for a first open field test, because the arena’s novelty is the manipulation the test depends on — a pre-habituated arena is no longer “open field” novelty exposure, it becomes a familiar-environment locomotor test with a different behavioral profile. Repeated testing in the same arena (e.g., across a longitudinal study) is a distinct design choice with its own interpretation, since exploratory drive and thigmotaxis both habituate with repeated exposure.

Protocol documentation should distinguish these explicitly: “habituation” in this context usually means acclimating the animal to the room and handling, not to the apparatus.

Thigmotaxis: the primary anxiety readout

Thigmotaxis — literally, orientation toward touch/contact, in practice the tendency to remain close to walls rather than crossing into open space — is the field’s standard anxiety-like readout in the open field test. It is typically quantified as the proportion of total session time (or distance traveled) spent in the peripheral zone relative to the center zone, sometimes supplemented with latency to first center entry and number of center entries. Higher thigmotaxis (more time at the wall, less in the center) is interpreted as more anxiety-like; a shift toward more center time and entries, without a corresponding change in total distance traveled, is interpreted as an anxiolytic-like effect rather than a general locomotor change — which is precisely why locomotor activity and center-zone measures are always reported together rather than either alone.

This center-versus-periphery logic and its pharmacological validation (anxiolytic compounds increasing center time/entries without simply increasing overall locomotion) is well established in the behavioral pharmacology literature reviewing the open field as an anxiety paradigm, alongside the classic methodological literature on the open field test’s origins as a broader emotionality assay. Because thigmotaxis is sensitive to arena size, lighting, and center-zone definition, absolute center-time values are not comparable across studies that used different apparatus parameters — only within-study, same-apparatus comparisons (e.g., genotype or treatment groups tested under identical conditions) are directly interpretable.

Common confounds

Several variables reliably shift open field results independent of the experimental manipulation of interest, and a well-designed protocol controls or at least documents each:

• Time of day. Mice are nocturnal and their baseline activity level varies substantially across the light/dark cycle. Testing all animals within a consistent time window, and reporting that window, is standard practice; comparing groups tested at different times of day confounds treatment effects with circadian activity differences.
• Handling stress. How an animal is picked up, how recently it was handled, and who is doing the handling all affect baseline anxiety-like behavior at test onset. Consistent handling technique and habituation to the handler (see above) reduce this as a source of unwanted variance.
• Lighting. Illumination level is one of the strongest environmental determinants of open field behavior — brighter light increases the aversiveness of the open center and increases thigmotaxis, while dim lighting reduces it. Lighting must be held constant across all animals in a study and reported in the methods, not left to whatever ambient light happened to be present.
• Prior test-order effects in a battery. The open field test is frequently run as one component of a behavioral battery alongside assays like the elevated plus maze or rotarod. Prior test exposure (or even prior handling stress from an earlier procedure that day) can alter subsequent open field performance, and order effects are not always symmetric — running the open field first versus last in a battery can produce measurably different results. A fixed, documented test order (and, where feasible, adequate washout between tests) is necessary for the data to be interpretable, and the order itself belongs in the methods section, not just the list of tests performed.

ARRIVE 2.0’s Essential 10 items on randomization and blinding apply directly here: allocating animals to treatment/control groups without confounding group assignment with cage, litter, or testing-order position, and scoring/tracking software analysis blinded to group assignment where feasible, both reduce the risk that one of the confounds above masquerades as a treatment effect.

What an IACUC protocol should specify

The open field test is non-invasive and generally low on the pain/distress spectrum relative to many other procedures reviewed by an IACUC, but “non-invasive” is not the same as “requires no protocol detail.” Consistent with the review criteria in 9 CFR 2.31(d)/(e) for animal-use protocols generally, a complete open field test protocol should specify:

• Scientific justification and species/number. Why the open field test is the appropriate assay for the study’s hypothesis, and how group sizes were determined (a priori power justification, per ARRIVE 2.0) rather than an unstated convention.
• Session duration limits. The exact length of each test session and total handling/restraint time per animal per day, especially where the open field is one component of a same-day battery — cumulative time out of the home cage across a battery is a welfare consideration, not just a methods detail.
• Cleaning between animals. The arena must be cleaned (commonly with a dilute ethanol solution, allowed to dry before the next animal) between every subject to remove olfactory and excretory cues left by the previous animal; residual odor cues are a documented confound that can bias exploration and anxiety-like measures in the next animal tested, independent of any treatment effect.
• Personnel qualification and monitoring. Who is authorized to handle animals and run the test, what training they have received (species-specific handling technique, apparatus operation), and how animals are monitored during and immediately after testing.
• Humane endpoints. Criteria for removing an animal from testing (excessive freezing, evidence of injury from a fall or jump attempt, signs of acute distress inconsistent with normal exploratory behavior) and what happens to that animal’s data and continued participation in the study.

Because the open field test is frequently embedded in a multi-assay battery, protocols should also describe the full test sequence and inter-test intervals, not just the open field component in isolation — reviewers need to evaluate cumulative animal use and welfare across the whole battery, and this is where the confounds discussed above (test order, handling frequency) become protocol-review questions, not just statistical ones.

The 3Rs in practice

The open field test supports Reduction and Refinement directly: as a same-session source of both locomotor and anxiety-like data, it avoids running separate assays (and separate animal cohorts) to obtain each measure independently, and its short, non-invasive design is comparatively low-burden relative to many alternative anxiety paradigms. Refinement considerations specific to this assay include arena size and wall opacity appropriate to the species and strain being tested, humane handling technique to minimize acute stress that would confound the very anxiety measure being collected, and defined humane endpoints as above. See the 3Rs framework for how these considerations fit into protocol review more broadly, and the Guide for the Care and Use of Laboratory Animals for the housing and husbandry standards that apply to animals used in behavioral testing.

Frequently asked questions

Is the open field test painful or invasive?

No. It involves handling and a period of exploration in a novel but non-aversive enclosed arena, with no injection, surgery, or restraint device involved. It is generally classified at the lower end of the pain/distress spectrum, though acute stress from novelty and handling is a real, if brief, welfare consideration that a protocol should still address.

How is thigmotaxis different from simple locomotor activity?

Locomotor activity (total distance, velocity) measures how much an animal moves. Thigmotaxis measures where it moves — the proportion of that movement confined to the peripheral wall zone versus the open center. A treatment can reduce thigmotaxis (more center time, an anxiolytic-like signature) without changing total distance at all, which is why the two measures are reported separately rather than combined into one score.

Can the open field test alone establish an anxiety phenotype?

A single behavioral assay is rarely treated as sufficient on its own in the literature; open field results are typically corroborated with at least one other anxiety-relevant paradigm (such as the elevated plus maze) before an anxiety-like phenotype claim is made, partly because single-assay results are sensitive to the specific confounds described above.

Does the open field test need to be run at the same time of day for every cohort?

Yes, or at minimum within a consistent, documented time window. Circadian variation in baseline activity is large enough in mice that testing different groups at systematically different times can produce group differences that have nothing to do with the experimental manipulation.

What should the IACUC protocol say about combining the open field test with other assays in a battery?

It should specify the full test sequence, the interval between tests, and the cumulative time each animal spends in testing/handling across the battery, not just describe the open field test as an isolated procedure — test order and cumulative handling both affect data interpretation and welfare, and both are proper protocol-review questions.