The study, which is one of the first to look at how fast second-language words are learned and how the brain responds to words with increasing experience with the new language, was published June 13 in the on-line edition of the journal Nature Neuroscience. The research team was headed by Judith McLaughlin, a UW research scientist, and Lee Osterhout, an associate professor of psychology. "Age and reduced brain plasticity are the classic reasons usually given for difficulty in learning a second language. But almost all thinking about this concerns syntax and grammar, while word learning has been ignored," Osterhout said.
"Our results clearly show there are aspects of a second language that can be learned quickly and with amazing ease. A number of our subjects told us they hadn't been studying that hard because the course had just begun or because they were taking French to simply fulfill their language requirement. What's remarkable, considering those factors and that the language wasn't being taught in an immersion environment, was that we saw this rapid change in brain activity."
The researchers recruited 18 students taking first-year French and eight students who were not studying French to serve as control subjects. None of the students had studied French before, although some had studied another language. There were roughly the same number of men and women in the group of French learners, and no gender differences were found in their French-language abilities. The subjects were tested in a laboratory three times, after approximately 14, 60 and 130 hours of French instruction. The control subjects were tested at similar intervals. At each session, the subjects' heads were fitted with a cap containing 13 electrodes that measured brain activity. Seated in front of a computer terminal, the students were shown a series of real French word pairs such as "chien-chat" and "maison-soif" or pseudo-word pairs such as "mot-nasier" one at a time. Their task was to decide whether both items in each pair were real words. Their answers and brain activity were recorded. The test took 15 to 20 minutes to complete, but the students were under no time pressure to answer. However, the researchers were interested in the accuracy of the answers.
The students not studying French performed at chance levels when consciously deciding whether the words were real. They showed no differences in brain activity while taking the test.
The story was entirely different for the students studying French. As their exposure to French increased, the difference in brain response to words and pseudo words also increased. This was true when comparing students at the first testing sessions. At the time of those tests, students had received between 6 and 20 hours of classroom French instruction (an average of 14 hours). The more hours of instruction students had, Osterhout said, the larger the difference in brain responses to words and pseudo words became.
This was also true when each student's performance was measured over the three tests. In every case, the researcher saw larger differences in brain responses at each successive session.
"What is clear is that the learners' brains are doing a much better job at word/non-word discrimination than the learners themselves," said Osterhout.
"At first, I thought this was an impossible result. It seemed incredible that the brain could do this with such facility while the subjects could not do it consciously. When students were asked how they thought they performed on the task, many of them laughed and said that they were just guessing. How could they possibly do well on the task if they had just started French instruction? What this study shows is that students are more successful at this stage of learning a language than they think they are. They learn a lot about French words before they have any obvious proficiency with French. It seems paradoxical that our learners' brains could know more than the learners themselves, but this is generally true when it comes to language. We know a lot more about our first language than we are consciously aware of. Our results suggest that this is true for a second language too," he said.
The researchers next plan to look at syntax and how quickly students learn the syntax of a second language, again French. Co-author of the paper was Albert Kim, a post-doctoral researcher at the UW. The National Institute on Deafness and Other Communication Disorders funded the research.