快猫短视频

THERE鈥橲 no escaping that, compared with many animals, humans seem to have a
pretty poor sense of direction. The navigational skills of migratory birds are
legendary. And bees often forage several kilometres from the hive and, on
returning, signal to their fellow workers exactly where they鈥檝e been feeding.
Not many humans could say they鈥檝e never been lost on their way home, let alone
be able to give the compass bearing and distance to the nearest caf茅.

And if we can find our way around a city or building it鈥檚 surely because our
brains have learnt what roads, buildings, corridors and other 鈥渓andmarks鈥� look
like, rather than because we possess some mysterious inner sense of direction.
Most animal navigators can sense the Earth鈥檚 magnetic field, or, at the very
least, possess specialised nerve circuits that do things like keep track of
changes in orientation. We mammals, on the other hand, appear to have
neither.

Or do we? For years, some neuroscientists have held a very different view.
Armed with a growing body of results from brain imaging experiments and detailed
studies of the brain cells that enable rats and monkeys to find their way
around, these researchers believe it鈥檚 high time for a rethink. The mammalian
brain may not be such a poor navigator after all, they say. Even though we lack
anything like a magnetic sensor, mammals do have something that can be thought
of as a sense of direction鈥攊t鈥檚 just that the evidence is rather subtle
and debate still rages over what this sense actually consists of in the
brain.

Blindfolding people, strapping them into a chair and asking them to swivel
might not sound like cutting-edge science, but sometimes the simplest
experiments can be the most revealing. When James Ranck of the State University
of New York Health Science Center at Brooklyn and his colleagues asked their
spinning subjects to pause at the angle they started at, they found that we鈥檙e
actually pretty good at keeping track of our bearings. People鈥檚 estimates, not
surprisingly, improve if they are allowed to turn round on their feet or remove
the blindfold. But even sitting passively in a chair with no visual clues, the
pauses stayed within 30 or so degrees of the correct angle for several
turns.

And when there鈥檚 food at stake, lab rats can do something similar. If you put
one on a turntable facing some food, spin it and watch, the animal usually heads
straight for the food without hesitation. Even when the lights are turned off
before the animal is spun, the rat still seems to be able to keep track of its
orientation.

For Ranck, this tells us something important鈥攖hat there is a mechanism
in the brain that keeps track of our bearings even when the available sensory
information is limited to the nerve signals that flow from the balance system in
the inner ear. Ranck calls it an inbuilt sense of orientation, but what is it,
and how does it work?

Rats again provided the first clues. One Sunday night in 1984, Ranck was
using an electrode to eavesdrop on neurons in a deep brain structure called the
subiculum. In one rat he missed, and instead found cells that behaved in a way
he鈥檇 never seen before. As the rat roamed about, Ranck noticed that some cells
fired when, and only when, the rat鈥檚 head faced in one specific direction. 鈥淚
had never seen anything like it. I made a TV tape that first night.鈥�

That some neurons are amazingly choosy about what they respond to was nothing
new. In the brain鈥檚 visual system, for example, there are cells that fire only
at the sight of specific colours and others that are excited by vertical rather
than horizontal lines. And, back in 1971, at University College London (UCL),
neuroscientist John O鈥橩eefe discovered cells in the rat that seem to keep track
of its position relative to the walls of a room, cage or maze. But Ranck鈥檚 was
the first evidence that some cells might also be able to sense specific
directions. 鈥淢uch of neuroscience is explaining behaviour we know exists,鈥� says
Ranck. 鈥淭his was a behaviour that few scientists, other than those interested in
bird migration, had even considered.鈥�

In neuroscience, however, merely identifying cells with weird properties is
not enough. Before others will take your finding seriously, you have to flesh it
out with a mass of detail about how the cells work, what connections they make
with other cells and where else they might exist in the brain. In the years that
followed, Ranck and his then colleagues Jeffrey Taube and Robert Muller
uncovered whole networks of these 鈥渉ead-direction cells鈥�. Today, Taube continues
the work in his own lab at Dartmouth College, New Hampshire, where at the last
count the cells had come to light in five different brain regions.

The researchers found that the directional preferences of the cells stay the
same whether the animal is running or stationary, and whether the room is big or
small, square or rectangular, cluttered or empty. Even in the dark, each cell
seems to remember its preferred direction and fires only when the head is
pointing that way. Curiously, though, unlike magnetic compasses, the arrays seem
to make an arbitrary decision as to what is 鈥渇ront鈥� or 鈥渘orth鈥�, and then try to
keep track of the animal鈥檚 orientation relative to that direction. As the animal
turns this way and that, cells with different preferences get excited and calm
down, so that the focus of activity moves from cell to cell in each array as the
angle of the head changes.

Some cells even anticipate head movements by up to a tenth of a second. This
suggests, says Taube, that in addition to receiving information from the eye and
the inner ear鈥檚 balance mechanism, the cells may be monitoring the nerve signals
sent by the brain to the neck muscles to control head movement.

The bad news for the rat as it tries to navigate is that after a few days the
direction cells鈥� sense of arbitrary north, or front, gradually drifts, causing
the animal to lose its bearings. Fortunately, the sight of a familiar landmark
such as a door or window can prompt a rapid recalibration. It can also correct
for the befuddling effect that too many turns have on direction cells. 鈥淚f
you鈥檙e spinning and dizzy, the cells begin to fire all over,鈥� says Taube. 鈥淏ut
when you see a landmark, the firing goes back to normal.鈥�

Taube鈥檚 use of the word 鈥測ou鈥� betrays his belief that the brains of primates,
including humans, also have direction-sensitive cells. For years there was
little evidence for this, but that changed three months ago when Edmund Rolls
and his colleagues at the University of Oxford announced the discovery of
direction-sensitive cells in macaque monkeys. Like their rat counterparts,
each of these monkey cells fires more rapidly when the animal鈥檚 head faces in
their preferred direction. They provide a good model of what happens in the
human brain, Rolls believes.

Of course, there鈥檚 more to finding your way around than knowing which way
your head is facing. To get from A to B, you also need to keep track of your
position in space. This is where the cells that O鈥橩eefe discovered in rats enter
the picture. It鈥檚 impossible to understand the mammalian sense of direction
without knowing about these position-sensitive cells. At UCL, O鈥橩eefe and his
colleague Neil Burgess are still mining them for information.

Maps and compasses

Whereas the neurons discovered by Ranck have preferences for specific
orientations, O鈥橩eefe鈥檚 鈥減lace cells鈥� have amazingly precise preferences for
specific locations relative to walls. Each neuron stays silent until the rat
strays into the closely defined region of space that is its preferred position,
at which point it begins to fire.

If the head-direction cells are the nearest thing the brain has to a compass,
place cells are akin to its map. Located in the brain鈥檚 hippocampus, a structure
that has a role in laying down new memories, the cells don鈥檛 look much like a
conventional map. Nevertheless, each position in a room or street is faithfully
represented by a place cell. And while the direction cells keep the same
preferences even if you pick the rat up and move it elsewhere, the preferences
of place cells are specific to each environment鈥攋ust like maps.

Nor is it only rats that have these maps. In Oxford, Rolls has found
something similar in the hippocampuses of monkeys鈥攂ut with one important
difference. The cells he discovered don鈥檛 keep track of where the animal is, but
where it is looking, encoding positions 鈥渙ut there鈥� in space. And Rolls thinks
there鈥檚 an obvious explanation for this. Rats tend to explore by running around
and sniffing. Primates, by contrast, tend to explore the space around them
without moving, by simply looking around. For them, it makes sense to have a
system that tracks positions 鈥渙ut there鈥� rather than keeping track of where they
are actually standing.

In humans, meanwhile, confirmation that the hippocampus has a vital role in
navigation has come from brain imaging. Working with Burgess and O鈥橩eefe,
Eleanor Maguire at the Institute of Neurology in London has used PET scanning to
identify which brain regions are most active as we try to find our way around.
One experiment snapped people鈥檚 brains as they attempted to navigate virtual
reality environments, including a complex town layout. Another involved taking
pictures of the brains of London taxi drivers as they recalled routes around the city
(This Week, 13 September 1997, p 16).
In both cases, the hippocampus was the most active region of the brain.

The main difference between rats and humans was that in people the activity
was mostly confined to the right side of the brain鈥攑robably because the
left side has become specialised for language skills, says Burgess. However, the
left side of the brain did kick in occasionally, perhaps as subjects recalled
details such as names of landmarks and street names.

For a more detailed understanding of what鈥檚 going on inside the hippocampus,
however, researchers still need to focus on individual neurons. O鈥橩eefe鈥檚 team
has spent a deal of time measuring the actions of a rat鈥檚 place cells as it
moves around a small enclosure. To keep track of the animal鈥檚 position, you
might think that all the cells do is respond to visual cues such as the
positions of walls and corners. But the fact that place cells can keep track of
a rat鈥檚 position even after the lights go out suggests otherwise. Just like
head-direction cells, place cells seem to combine information from different
senses.

What鈥檚 more, even when there are no objects or landmarks, place cells can
distinguish between walls and corners that look identical. O鈥橩eefe thinks it鈥檚
no coincidence that all the head-direction cells studied by Taube are found in
brain regions connected to the hippocampus. Here, they are well positioned to
influence place cells and give them their bearings. O鈥橩eefe argues that the
combination of head-direction cells and place cells gives the brain the closest
thing mammals have to an innate sense of direction, a mechanism that is
鈥渉ard-wired鈥� into the brain from birth and which enables us to learn the
positions of important things such as landmarks, food and home.

All of which is more controversial than it sounds. Many people would
instinctively say that all you need for finding your way around is a good
memory. And some researchers would agree. The cells studied by Taube and O鈥橩eefe
might seem to behave like compass and map, but perhaps they are nothing more
than elaborate memory devices, says Bruce McNaughton of the University of
Arizona in Tucson, another scientist who has spent years studying place cells in
rats.

McNaughton claims that what the cells create in the hippocampus is not much
of a map after all. It works only in open spaces bounded by uniform walls and
readily slips out of register. What O鈥橩eefe is really seeing in the pattern of
firing place cells, says McNaughton, is not the real-time responses of a
hard-wired sensor but the activation of spatial memories. Rolls takes an even
more extreme view, that the hippocampus is a general memory system rather than a
spatial system.

Disoriented rats

Nobody denies that memory has an important part to play in navigation. If
nothing else, it explains how we humans can plan a route before even stepping
outside. Nor is there any question that the navigational skills of rats are
aided by memory鈥攂ecause drugs that block learning disorient rats and alter
the behaviour of place cells. Nevertheless, O鈥橩eefe argues that the evidence
just doesn鈥檛 support McNaughton鈥檚 line.

He points out that at least some aspects of the behaviour of place cells and
direction cells do not change as a result of learning, and so appear to be
hard-wired. The relative angles at which different head-direction cells fire are
so constant that the connections between these cells are probably fixed from
birth. And some of the wiring that connects place cells and direction cells is
also likely to be innate. To O鈥橩eefe, the message from all this is clear: the
mammalian brain doesn鈥檛 rely exclusively on memory for navigation. It has an
extra mechanism for keeping track of position and orientation.

In a similar vein, Taube speculates that head-direction cells and place cells
first evolved to enable the brain to keep track of an animal鈥檚 whereabouts
without the aid of a massive input from learning and memory mechanisms. An
evolutionarily primitive system might have continuously monitored information
flowing through the eyes and other senses, as well as the body鈥檚 movements. If
it had no memories to fall back on to correct drift, the system would have been
error-prone, as well as a drain on resources. So, as the brain evolved,
continuous monitoring gave way to a more economical and accurate strategy that
includes memory.

Taube believes that without this development, navigation would need far more
effort by the brain. We humans take it for granted that we can update our
internal sense of where we are in the world simply by keeping an eye open for
familiar landmarks. 鈥淵ou could look at the Tower of London every now and again,鈥�
says Taube. 鈥淵ou don鈥檛 have to look at it continuously to know your
辫辞蝉颈迟颈辞苍.鈥�

Navigation researchers will surely soon untangle the subtle interplay of
landmarks, memory and the choosy cells of the hippocampus. But while that debate
rumbles on, there鈥檚 a related question that needs to be answered鈥攁bout one
of those tantalising quirks of human nature. How is it that even with all this
complicated neural circuitry, some people get lost as soon as they step outside
the front door? That, as they say, is another story.

Myths abound about the navigational skills and shortfalls of men and women.
But apart from inevitable arguments about who can and can鈥檛 read the map, and
when might be the right time to give up and ask for directions, Taube says that
men seem to be better at comparing the orientations of different environments or
rooms than women. This may reflect subtle differences in the connections between
the head direction cells, or perhaps the way we use them.

Last year, Noah Sandstrom and his colleagues at Duke University in North
Carolina found that men and women seem to use different visual features to help
them navigate in a virtual environment鈥攊n this case a virtual pool of
water in the middle of a trapezoidal room with four different objects in it. The
pool contained a hidden submerged platform, which was always in the same
position, so its position should have been easy to learn. Women became very bad
at finding the platform if the researchers cut the objects out of the scene,
whereas men were able to find the platform using either the room鈥檚 geometry or
the objects.

These differences are not something we鈥檙e born with. Young children seem to
use geometry rather than landmarks to find their way around rooms, according to
a 1994 study by Linda Hermer and Elizabeth Spelke of Cornell University in
Ithaca, New York. They let children aged between 18 months and 2 years watch
them hide objects in a corner of a rectangular room with one brightly coloured
wall.

Although the children were spun round and moved about the room before they
were allowed to search for the objects, the coloured wall should have given them
a very obvious landmark for finding the correct corner鈥攂ut the children
didn鈥檛 use it. Instead, they were just as likely to search the diagonally
opposite corner, presumably because it was geometrically equivalent to the
correct corner. So the ability to use visual landmarks must develop as we grow
up鈥攁nd it seems that women abandon their early grasp of geometry in favour
of landmarks.