Pictures of the Future    Spring 2007

Watching Thoughts

Until recently, the human brain, with its complex matrix of billions of nerve cells, was largely inaccessible for scientific inquiry, and researchers had no alternative but to use indirect means to determine how human beings are able to perceive their surroundings, think and learn.

A new method introduced a few years ago, however, is making it possible to show how the brain works while it is processing our sensory impressions. Known as functional magnetic resonance imaging, fMRI "enables us to literally watch the brain while it’s thinking," reports Dr. Robert Rauschenberger, a cognitive psychologist at Siemens Corporate Research in Princeton, New Jersey. Rauschenberger is part of a team that also includes Prof. Frans Verstraten of the Helmholtz Institute at the University of Utrecht, which funds the project, and Tom Carlson of Harvard University in Cambridge, Massachusetts.

Functional magnetic resonance imaging technology examines brain metabolism and superimposes the results on MR images to map areas of increased brain activity. Firing neurons need more oxygen, which is why more blood flows into areas that are active. In contrast to blood with low oxygen content, the increased concentration of oxygen-rich blood triggers a change in the signal of the tomograph’s magnetic field. The value actually being measured is the amount of iron in the hemoglobin of the red blood cells.

"With a resolution of about 1 mm³, we can precisely delineate areas of activity," Carlson explains. "Not only were we able to show for the first time which areas of the brain are active during which cognitive processes, but also how they function."

"The results are quite amazing," adds Rauschenberger. "We have demonstrated, for instance, that there is an unconscious flow of information between brain cells and the eyes." The findings of this basic research could lead to the development of new technologies and systems that would help make people aware of things they might otherwise fail to notice.

Several American companies are planning to market new kinds of lie detectors that use functional magnetic resonance imaging to determine the veracity of statements by measuring brain activity. However, experts doubt that this method will be ready for practical use in the next few years. It is unclear, for example, if functional MRI can accurately identify suspicious persons who are not aware of having committed any wrongdoing or who are pathological liars. Furthermore, lie detector tests are influenced by a large number of factors. "Although our experiments are similarly complex, they are conducted under controlled conditions and are therefore reliable," explains Rauschenberger.

Visual Feedback. Using a Siemens tomograph with a magnetic field strength of three Teslas, Rauschenberger and Carlson viewed test subjects’ brain reactions to a variety of images. Previous tests with words that flashed for 30 ms showed that these not consciously perceived stimuli are in fact registered by the brain. This is proved by a measurable signal that the brain returns to the eye.

These results led scientists to believe that it might be possible to use subliminal messages to manipulate cinema audiences to buy certain products. "But it doesn’t work that way," says Rauschenberger. Only under very specific conditions can such messages trigger desires, and it’s almost impossible to steer those desires toward a specific brand. "Instead of words, we showed our test subjects squares with a corner missing," says Rauschenberger. Such shapes generate simpler patterns in the brain than is the case with words, which also create associations and bring back memories that can set off a veritable shower of sparks in a functional MRI.

As with the experiment using words, test subjects viewing a square were shown an image for about 30 ms. After a pause, this was followed by a stimulus that lasted for 30 to 100 ms and masked the first impression. Because the subjects were lying inside an MR tomograph for the entire duration of the test, "we were able to measure their brain activity at the same time," says Carlson. Although it takes several seconds for the blood to flow through the brain, the process can be precisely tracked if the chronological sequence of the images is known.

Researchers asked the test subjects to say which of the square’s corners was missing. They couldn’t provide the answer, though, because the image was masked too quickly. "That proves they weren’t consciously aware of the square," Rauschenberger explains. "As we can see from the signal, the sensory impression is first transmitted to the brain. The brain then asks for confirmation—which it doesn’t receive, because the image has already been masked by a new sensory impression."

In the fMRI the signals for the square and the masking stimulus are usually distinct. But why does the masking impression cause the square to vanish from people’s memories? "That was the question we wanted to find answers to," recalls Rauschenberger. Is it because the masking stimulus deletes the impression of the square, or because this information is no longer accessible to the conscious mind—even though it’s still stored in the brain?

To find the answer to the puzzle, Rauschenberger and Carlson decided to change the experiment. Following the masking stimulus, they once again showed the square with the same corner missing. If the first sensory impression was still stored in the brain, the second signal would be weaker because the nerve stimulus would be repeated. "But the signal remains undiminished, as if the first square had never existed in the first place," Rauschenberger says. According to currently accepted theory, the brain processes the second signal completely independently from the first, which means the masking stimulus entirely deletes the first sensory impression from the brain.

Psychological Science will soon publish new findings from the experiments Rauschenberger conducted together with Carlson and Dutch psychologist Verstraten. The results demonstrate that the brain can’t be fooled. Although the conscious mind can be deceived, neurological processes are objectively displayed by the fMRI. "I myself was one of the test subjects," reveals Rauschenberger. "And although I knew exactly how the test would proceed, my fMRI values are exactly the same as those of all the other test subjects. Even I couldn’t manipulate the result."
                                                                                                  Norbert Aschenbrenner