The open field: what an empty box reveals

Opening

The box is almost insultingly plain. Four walls, a flat floor, bright light from above, and nothing else. No platform, no food, no hidden goal. A camera looks straight down from the ceiling.

A mouse is set down in one corner. For a moment it freezes. Then it starts to move, but only along the edges, whiskers brushing the wall, tracing the outline of the box. It stops, rises up on its hind legs, sniffs the air, and drops back down. Minute by minute, its loops get a little bolder. Eventually it makes a quick dash across the bright, empty middle, and scurries back to the wall on the far side.

Nothing is being asked of it. And yet, in those few minutes, it's telling us a great deal.

An empty box

This is the open field test, one of the oldest and most widely used tests in behavioural neuroscience. It's also one of the most argued about. How did an empty box become a standard measure of anxiety? And is that really what it measures?

The apparatus

The apparatus is simple. An arena, usually square or round, with walls high enough to stop the animal climbing out. Traditionally the floor was marked with a grid of squares, so an observer could count how many lines the animal crossed. The centre of the field is treated as its own zone, and so is the strip along the walls.

A trial is just as simple. The animal is placed in the arena, often in a corner or the centre, and left to explore for a fixed time, often five or ten minutes, sometimes longer. Then it's returned to its home cage, and the arena is cleaned before the next one.

Today, a camera and tracking software usually do the counting. They record where the animal is many times a second, and build up a picture of its whole session.

The behaviour falls into a few families. There's how much it moves, measured as total distance travelled or line crossings. There's where it moves, especially the time spent in the centre and the number of entries into it. And there are particular actions. Rearing is standing up on the hind legs to look around. Grooming is self-cleaning. There's a cautious posture called stretch-attend, where the animal stretches its head and shoulders forward and then pulls back, as if weighing up a risk. And then there's the one that started it all, defecation.

The same arena often doubles for other tests. Put two objects in it, and you have the basis of novel object recognition. Swap one object for a new one, and an animal that remembers the old one will spend more time exploring the newcomer. So the bare box is also a stage.

Where it began, the 1930s

The test goes back to the nineteen thirties and to Calvin Hall, then a young psychologist who had trained at the University of California, Berkeley. Hall was interested in emotionality, the idea that some animals are naturally more fearful or reactive than others, and that this might be inherited.

He needed a way to measure it. His answer, published in 1934, is right there in the title of his paper. Emotional behaviour in the rat, part one. Defecation and urination as measures of individual differences in emotionality.

The reasoning was straightforward. Put a rat in a brightly lit, unfamiliar open space. A frightened animal, like a frightened person, may have a nervous gut. Count the droppings, and you have a number for fear. Hall also looked at how much the rats moved, and the idea that fearful animals explore less became part of the test.

It's a slightly undignified start for such a famous method. It's also a good reminder that early behavioural science made do with what could be counted by eye. Hall, incidentally, later became far better known for an entirely different subject, the scientific study of dreams.

Hall went further than measuring. He was interested in whether emotionality runs in families, and his work on the inheritance of emotionality in rats was an early contribution to what became behaviour genetics. The open field was the yardstick that made that possible. If you can give every animal the same simple test, you can compare parents and offspring, and strains with strains.

Over the following decades, the open field grew into a general-purpose test. Researchers stopped relying on droppings alone, and paid more attention to where the animal went. One pattern stood out. Rodents placed in an open space tend to stick to the edges and avoid the middle.

Thigmotaxis

Hugging the wall has a name. Researchers call it thigmotaxis, from Greek words meaning touch and arrangement. In 1994, a team in Rouen led by Simon, with Dupuis and Costentin, studied it directly in mice. They noted that mice introduced into an open field mainly explore the outer zone, and that this tendency fades over the first few minutes.

They proposed using the degree of wall-hugging during that early period as an index of anxiety. Then they tested the idea with drugs. Several drugs known to increase anxiety, such as yohimbine and pentylenetetrazole, increased wall-hugging. Drugs that reduce anxiety, such as buspirone and phenobarbital, reduced it. That's the kind of check researchers call pharmacological validation. If a test is measuring anxiety, anxiety drugs should move it in the expected direction.

A rich picture emerges from a single session. Picture two mice, each tracked for ten minutes. Both travel about forty metres. The first spends a quarter of its time in the central zone. The second spends less than a twentieth there, keeping almost entirely to the walls. Same activity, very different use of space. Most researchers would read the second mouse as more cautious.

That's also why the open field is such a useful baseline. Before you interpret any other test, you want to know whether an animal is simply more or less active. A mouse that barely moves will look poor in many memory tasks. The open field tells you whether that's the explanation.

Modern tracking adds even more. Time can be broken into short bins, to see how behaviour changes as the animal settles. That matters, because the most revealing moments are often the first few minutes, when the space is new. Remember that Simon and colleagues saw wall-hugging fade over exactly that period. Average over a long session and you blur it. Watch minute by minute and you can see an animal getting used to its surroundings, which is a simple form of learning in itself. Speed, turning and resting can all be measured. Some systems use two cameras to measure rearing in three dimensions, rather than relying on someone watching and pressing a key.

Is it an anxiety test?

So, is the open field an anxiety test? Here's where it gets complicated. In 2003, Laetitia Prut and Catherine Belzung published a large review of how drugs behave in the test. Classic anti-anxiety treatments, such as benzodiazepines and certain drugs acting on serotonin receptors, usually produced the expected effect, in roughly two thirds of cases.

But other drugs that help people with anxiety disorders did poorly. Some close relatives of the benzodiazepines, including alprazolam, and the selective serotonin reuptake inhibitors widely prescribed for anxiety, were poorly effective in the open field. The authors concluded that the test may not model features of anxiety disorders such as panic, generalised anxiety, or obsessive-compulsive disorder. It does pick up drugs that increase anxiety, though.

There are other cautions. The droppings measure that started it all is now contested. Some scientists see it as a sign of emotional arousal, but not necessarily of anxiety.

The details of the set-up matter a lot. Bright light makes the centre more aversive than dim light. The size and colour of the arena, the time of day, how the animals were handled, and the strain of mouse can all change the results. Different labs using different boxes can get different answers.

And the test is built on a conflict between curiosity and caution. Less centre time might mean more fear. But it might also mean less interest in exploring, or simple fatigue. That's why many researchers pair the open field with other tests, such as an elevated plus maze or a light and dark box, and look for a consistent picture. If an animal avoids the open arms of a plus maze, the bright side of a light and dark box, and the centre of an open field, the case for caution is much stronger than from any one test alone.

Back to the box

Back to that plain, bright box, and the mouse tracing its edges before daring the middle. Nothing is asked of it, and that's the point. Left to its own devices, an animal reveals how it balances the urge to explore against the urge to stay safe. The open field can't tell us everything about anxiety. But as a first look at how an animal behaves, it's hard to beat.

hvsimage.com has more on tracking and zone analysis in the open field. You'll find the papers we mentioned, and a transcript, at water-maze.com.