Scientists have shown that the human brain is engaged by words and numbers flashed so fast that they don't have time to register in a person's consciousness. The findings, which appear in tomorrow's Nature, indicate that more parts of the brain respond to subliminal messages than researchers had thought.
Earlier studies have found that numbers and letters flashed before the eyes often register in the brain, a process called "semantic priming." A team led by Stanislas Dehaene of the Service Hospitalier Frederic Joliot in Orsay, France, wanted to take this a step further and analyze the brain activity patterns triggered by this priming.
Researchers began by flashing, for 4 hundredths of a second, a number on a computer screen in front of 12 subjects. Before and after the number--which they dubbed the "prime"--were nonsensical words to make it harder for the brain to process the number. The subjects were not asked to remember the prime, nor were they even aware it was being flashed on the screen--instead, they were told to recall a different number that appeared for 20 hundredths of a second at the end. Then the subjects were asked to press a buzzer with one hand if the number to be remembered was high, and the other hand if it was low.
In the meantime, the researchers followed the subjects' brain activity measured by electrodes glued to the scalp. While the subjects didn't appear particularly distracted by the prime, their brains certainly did: Almost every time the final number differed from the subliminal one, the brain responded as if it wanted to press the buzzer with the wrong hand--the one a subject would use if reacting to the prime. In other words, the right side of the brain flared up if the prime required a left-handed push on the buzzer. This suggested that the subjects were tempted to respond physically to a number they'd presumably not consciously noticed. "It's the first time we showed that semantic priming goes all the way to the motor system," says Dehaene.
The data are persuasive, says Anthony Greenwald, a psychology professor at the University of Washington in Seattle. He expects this work to fuel interest in probing the boundary separating conscious and unconscious thought.