ConductScience VR

Open field test in virtual reality

People explore a wide virtual field ringed by rocks, and the test tracks whether they keep to the edges, as rats and mice do in the open field.

ConductScience Editorial · Published November 15, 2026 · 5 references

INNER REGION
Drawing · Virtual open field
Task
Explore a virtual grassland ringed by rocks for 15 minutes, from one of eight start points.1
Measures
Time in the outer region, distance to the wall, movement along the wall and entries into the center.1
Virtual version
Gromer and colleagues showed in 2021 that people keep to the outer region of a virtual field.1
Real field
An earlier human open field ran on a real field, with people tracked by GPS.2,1

What the virtual open field tests

Thigmotaxis is the tendency of animals to stay close to walls when they explore an open space.2 The rodent open field test measures it, and Walz and colleagues call the test a well-established indicator of fear in animals.2 The test assumes that rodents avoid open spaces for fear of predators, including those from the air, which makes moving along the walls an anxiety-like behavior.1

Gromer and colleagues built a human open field in virtual reality and scored it with measures taken from animal studies, in three experiments.1 They also asked whether anxiety traits, such as trait anxiety and anxiety sensitivity, change how people move in the field.1 They point to two strengths of virtual reality, a lifelike setting under laboratory control, and repeated testing in the lab.1

How a session runs

The virtual field is a grassland 100 m by 100 m, with stones and flowers spread across it.1 In the second and third experiments, a wall of large rocks about 2.5 to 4.5 m high rings the field.1 The first used a mix of concrete-like walls and rocks about 3 to 6 m high, with remains of old concrete structures across the field.1 The third added more and larger rocks and some tree stumps.1 For scoring, the field is split into an inner region with half its area and an outer region around it.1

  1. Practice moving in a plain scene

    Wearing a headset and holding a controller, the participant first collects blue spheres in a plain virtual scene to get used to moving in virtual reality.1

  2. Start at one of eight points

    The display fades to black, and the participant arrives in the open field.1 Each participant starts at random from one of eight points, in the center facing one of the four walls or at a wall facing the center.1

  3. Explore for 15 minutes

    A recorded instruction asks the participant to explore the field freely.1 They move by teleporting, or by walking within the lab’s floor space of about 2.5 by 3.5 m.1 After 15 minutes, the display fades to black.1

  4. Rate anxiety at three places

    The participant is moved in turn to the center, a wall and a corner, in a mixed order, and rates anxiety and arousal at each.1

  5. Fill in questionnaires

    After taking off the headset, the participant fills in questionnaires, among them measures of trait anxiety, agoraphobic thoughts and anxiety sensitivity.1

One score is the time spent in the outer region, out of a possible 900 seconds.1 Others are the mean distance to the nearest wall, and the walking distance and speed in each region.1 For counts of movement, the field is divided into a grid of 7 by 7 squares, split into inner and outer squares of about equal area.1 Thigmotaxis counts moves between squares of the outer region, and center entries count moves from an outer square into an inner one.1 Each score is compared with a simulation of random movement.1

What the test has shown in people

Keeping to the outer region

In three experiments with 31, 30 and 80 participants, people kept to the outer region of the virtual field.1 In all three, they spent more time in the outer region than random movement predicts, moved more along the wall, and entered the center less often.1 In the second and third experiments they also stayed closer to the wall than chance, and in the first they did not.1 Their paths were longer in the outer region than in the inner one, and their speed was higher in the inner region.1 The authors conclude that the virtual field elicits thigmotaxis, which shows cross-species validity for this part of the test.1

Anxiety ratings and anxiety traits

Anxiety and arousal ratings did not differ between the center, a wall and a corner, in any of the three experiments.1 Links between anxiety traits and behavior were mostly small, not significant and not consistent across the experiments.1 Counting only the direction of each correlation, 46 of 72 pointed toward more center avoidance with higher anxiety, significantly more than pointed the other way.1 The authors suggest adding threats to the scene, or testing people with higher anxiety or with anxiety disorders.1

Objects in the field changed how people moved

The design of the field changed how people moved, which suggests that the number of objects across it, such as stones and flowers, affects behavior.1 The authors suggest that a scene with more objects may draw people away from the outer region, to explore or to relieve boredom.1

An earlier test on a real field, tracked by GPS

Before the virtual version, Walz and colleagues ran a human open field test on a real field, without virtual reality, and tracked people by GPS.2,1 They tested 16 people with agoraphobia, a pathological fear of open spaces, and 18 healthy people with high anxiety sensitivity, a risk factor for agoraphobia.2 Control groups of 16 and 19 people served for comparison.2 Both groups showed more thigmotaxis than their controls: they walked more along the wall and entered the center less, while their speed and the length of their walks were normal.2 Walz and colleagues describe their study as the first, to their knowledge, to verify that the open field test translates to people.2

Wall-following in other virtual tasks

Kallai and colleagues studied wall-following in a virtual water maze task, a navigation test in a round pool, not an open field.3 Among 112 undergraduates, thigmotaxis, keeping continuous contact with the circular wall, was one of three search strategies that predicted spatial performance.3 A later study by Kallai and colleagues followed thigmotaxis in both virtual and physical spaces, in 106 participants.4 In the first phase of learning, more thigmotaxis went with more general phobic avoidance, and in later phases there was no link.4 Participants who did worse on working memory tests and a spatial construction task showed more thigmotaxis and were more prone to fear responses.4

How it compares with the rodent open field

Both versions put the subject in an open space inside a wall it cannot escape, and both score how much the subject keeps to the outer region.5,1 In Seibenhener and Wooten’s mouse protocol, the arena is 50 by 50 cm, scored on a 5 by 5 grid of 10 cm squares, 16 outer and 9 inner.5 More time in the outer region counts as more thigmotaxis and more anxiety-like behavior.5 The virtual field uses the same kind of grid, 7 by 7, on a field 100 m across.1 A mouse session in that protocol lasts 10 minutes, and a virtual session 15.5,1

In mice, thigmotaxis is a validated measure of anxiety-like behavior and rises with anxiety.5 Touch plays a part, since damaged or missing whiskers may lower measured anxiety-like behavior as the mice lose contact with the walls and enter the center more readily.5 In the virtual field its link to anxiety traits was weak, while on the real field people with agoraphobia showed more of it.1,2

Gromer and colleagues report a strong similarity in behavior across species.1 They add that comparing time in the center with rodent studies is not simple, because the open field test is not standardized.1 In a descriptive comparison, people in the virtual field spent a little more time in the center than rodents in animal studies, by about 7.1 to 9.6%.1 They spent less time in the center than people on the real field, by about 15.1%.1 Against the real field, the authors point to differences in how people move and in field of view.1 People had to teleport to cross the virtual field, and the authors cannot rule out that this unnatural way of moving affected their behavior.1

Build a virtual open field with ConductScience VR

We write the task software in house and build each virtual field to the protocol you plan to run.

  • The field’s size and ground, the wall around it, the objects across it and the start points
  • Teleporting or walking, the session length, and rating trials at the center, a wall and a corner
  • Each participant’s path, time in the outer region, distance to the wall, movement along the wall and center entries, saved for analysis
  • A desktop screen with a keyboard or joystick, or a headset
  • The software alone, or a package with the headset and computer

We also build teaching versions, so students in a class can test each other on the task.

Related pages

Questions

Is the virtual open field the same test as the rodent one?
It keeps the rodent design, an open space inside a wall, scored by time in the outer region and entries into the center.1,5 The field is far larger, 100 m across, and people cross it by teleporting.1
Do people keep to the walls as rodents do?
Yes, in three experiments people spent more time in the outer region and entered the center less than random movement predicts.1
Does the virtual open field measure anxiety?
Not clearly, since anxiety traits were only weakly and inconsistently linked to how people moved.1 On a real field, people with agoraphobia and people with high anxiety sensitivity showed more thigmotaxis than controls.2
Did the human open field start in virtual reality?
No, an earlier version by Walz and colleagues ran on a real field, with people tracked by GPS.2,1 Gromer and colleagues then built the virtual version, for its laboratory control and repeated testing.1
Can the virtual open field run on a desktop screen?
Yes, we build it for a desktop screen with a keyboard or joystick, or for a headset. Gromer and colleagues ran the virtual open field in a headset.1

References

  1. Gromer D, Kiser DP, Pauli P. Thigmotaxis in a virtual human open field test. Scientific Reports. 2021. doi:10.1038/s41598-021-85678-5
  2. Walz N, Mühlberger A, Pauli P. A Human Open Field Test Reveals Thigmotaxis Related to Agoraphobic Fear. Biological Psychiatry. 2016. doi:10.1016/j.biopsych.2015.12.016
  3. Kallai J, Makany T, Karadi K, et al. Spatial orientation strategies in Morris-type virtual water task for humans. Behavioural Brain Research. 2005. doi:10.1016/j.bbr.2004.10.015
  4. Kallai J, Makany T, Csatho A, et al. Cognitive and affective aspects of thigmotaxis strategy in humans. Behavioral Neuroscience. 2007. doi:10.1037/0735-7044.121.1.21
  5. Seibenhener ML, Wooten MC. Use of the open field maze to measure locomotor and anxiety-like behavior in mice. Journal of Visualized Experiments. 2015. doi:10.3791/52434

More virtual reality tests