Scientists have long conceptualized the part of the brain known as the primary somatosensory cortex (S1) as where it first registers touch sensations. Prick your finger and S1 springs into action, sending raw information about the injury's location to higher brain areas for further interpretation, according to most neuroscience textbooks.

Those textbooks may need new editions. S1 doesn't simply catalogue physical sensations: It also registers sensory illusions that are generated elsewhere in the brain, according to a recent study in PLOS Biology (Vol. 4, No. 3, pages 459-466). In fact, as far as S1 is concerned, there's no difference between a real or imaginary touch, says lead author Felix Blankenburg, PhD, a neuroscience researcher at University College London (UCL). Other researchers, including David Ress, PhD, a neuroscience professor at Brown University, are finding similar results in S1's cousin, the primary visual cortex.

Together, the research paints a picture of a deeply integrated brain, one that begins making sense of information at the earliest stages of perception, says Ress.

"You use a lot of your brain to make a visual decision," he says. "The whole system is probably used as an integrated whole in order to create visual consciousness."

Tactile illusions

Tap people's arms rapidly at the wrist and then at the elbow, and they will feel a phantom tap right in the middle, as if a rabbit were hopping the arm's length. Blankenburg and his colleagues, including Jon Driver, PhD, director of the Cognitive Neuroscience Institute at UCL, harnessed this phenomenon, known as the cutaneous rabbit illusion, to see how tactile illusions play out in the brain.

The researchers strapped electrodes to the arm of 10 adult participants, placing the electrodes at three points between each participant's elbow and wrist. While the participants lay in a functional magnetic resonance imaging (fMRI) machine, the researchers delivered pulses to the electrodes. In one condition, participants experienced real sensations hopping up their arms, as experimenters activated the three electrodes in succession. In another condition, participants only thought they felt the sensation hopping up their arms, as researchers delivered pulses first to the electrode near their wrist and then to one by their elbow.

Participants reported feeling the illusory touch and the real one equally strongly, and their brains agreed-the S1 area registered both sensations at the same location in the brain and with a similar amount of neural activity.

"This is quite remarkable because traditionally we thought S1 formed a map of the body that faithfully represents the actual touch on the skin, but our results suggest this is not always the case," says study author Christian Ruff, PhD, a psychology and neuroscience researcher at UCL. Instead, S1 seems to be representing what we feel-not what is actually there, he adds.

Where is S1 getting its false information? One possibility is that higher areas of the somatosensory cortex, the ones that would integrate information about the time and location of a tap on the skin, also register raw sensory information and then force their interpretation on S1, says Ress, who also studies perception.

In fact, while S1 showed no differences in activation during real and imagined touch, the right premotor cortex showed increased activation during the illusory touch, and that area may be at least part of the illusion's source, Ruff observes.

"It could be that signals from higher-level brain structures can influence the primary sensory cortex via neural feedback connections," Ruff notes.

Visual mistakes

Researchers who study an entirely different sense-vision-are coming to similar conclusions. Scientists traditionally claim that the primary visual cortex, or V1, registers sensory information and then kicks it to higher-level areas for processing. A study published in a 2003 issue of Nature Neuroscience (Vol. 6, No. 4, pages 414-420) suggests otherwise.

The study's four adult participants lay in an fMRI machine while watching a screen that showed a faint vertical grating on a similarly patterned background, or just the background alone. Participants had one second to view the screen and then one second to press a button indicating whether they had seen the vertical grating -a process repeated several hundred times for each participant.

Analysis of their brain activity showed high levels of activity in V1 both when the participants saw the grating and when they just thought they saw it. What's more, the V1 area was similarly quiet when participants did not see the grating as when they just missed it.

These results, like those of the Blankenburg study, help explain why false perceptions sometimes feel quite real, says Ress.

"If you think you perceive a sensation, then the lower-level primary sensory area that is associated with that false perception actually becomes involved," he notes.

However, he cautions that fMRI data doesn't always match up with the electrical activity of the brain.

"It's a very indirect measure of neural activity, and we are still not exactly sure what it means," he notes.

That said, this line of research could eventually help amputees who suffer from phantom limb pain, Blankenburg says. If phantom pain comes from the lowest level of the sensory system, effective drugs or therapy could target that area.

In the distant future, research on the translation of sensation to perception may lead to machines that transmit visual signals directly into the brains of blind people, allowing them to see. But if higher level areas of the brain feed information to the lower areas, as is suggested by this line of research, such applications wouldn't just be able to transmit raw data straight into people's primary cortices, Ress posits.

"The design of something that emulates cortical processing becomes more complicated when the brain is a recursive network," he says.