Electrodes in Brain Give Insight Into How We Use Language


I doubt there’s many people who really like having electrodes stuck in their brains, and for those people who do…eww. Of course if someone is going to be placing wires into your neurons, they might as well do as much as they can while they’re in there. So it was with researchers at Massachusetts General Hospital (MGH) who used probes for epilepsy to simultaneously explore one of the brain’s language processing hubs: the Broca Center. As described in the journal Science, Ned Sahin and his team were able to determine that it takes just 600 milliseconds for the Broca center to receive, translate, and route language information. This discovery gives researchers rare insight into how the brain works – demonstrating that a single cluster of neurons can perform multiple tasks in succession. You can listen to Sahin and others describe the work on an audio clip from NPR’s Morning Edition after the break.

The Broca Center of the brain plays a role in receiving words, adjusting them for grammar, and pronouncing them.
The Broca Center of the brain helps receive words, adjust them for grammar, and pronounce them.

There are many different research teams, such as the Blue Brain Project we discussed earlier, trying to better understand the brain in order to simulate it on a computer. Any discovery about the way in which neurons connect and reconnect to perform tasks advances that research. Direct readings of the brain also help doctors find methods to compensate for or correct damage to the organ. Language, one of the great hallmarks of human intelligence has been broken down and analyzed endlessly to help create the next generation of translation software. It stands to reason that the human brain itself may have processing techniques that are more efficient than those programs and which could inform designers in improving them. In short, as we continue to explore the brain we may see artificial intelligence that helps us heal ourselves and talk to one another effortlessly.

The three volunteers who let Sahin’s team monitor their brain activity were examined through intra-cranial electrophysiology (ICE), which places metal electrodes directly into clusters of neurons in the brain. ICE is typically a medical tool used to help surgeons determine which parts of the brain may be excised to partially correct epilepsy. The research team piggy backed onto the epilepsy analysis and used electrodes in the Broca Center to measure its activity. Each volunteer was asked to repeat words verbatim as well as translate them into different tenses.

During this recitation, Sahin’s team noticed that there were peaks in activity at 200, 320, and 450 milliseconds. These were found to correspond to lexical, grammatical, and phonological processing. In other words, the Broca center was recognizing words, translating them into the correct grammar, and helping guide the brain to pronounce them correctly. Prior to this, scientists believed that another part of the brain, the Wernicke’s Center handled all the receiving of information and the Broca processed it and sent it out. From this work, Sahin and his team believe they have found evidence that parts of the brain may have multiple types of jobs that are accomplished at different times. That’s sort of like finding out a carpenter also works as a lumberjack and a sculptor depending on the day of the week.

With future help from volunteers, Sahin hopes to use ICE to examine different processes in the brain’s various language centers. This current work is fairly narrow, just three patients, and just one center thoroughly examined, but it represents unique data. Unlike fMRI which has good spatial resolution, ICE has excellent temporal resolution, helping us understand the speed and order of neural activity. As we learn the details of how the brain works, we may see researchers try to stimulate as well as measure output. Conceivably a project like Braingate, which measures motor neurons to allow patients to manipulate computer cursors or electric machines, could be adapted to work with the signals in the brain’s language centers. Imagine not only being able to control a computer cursor, but being able to type words simply by thinking about them, or reading electrical signals sent to the brain as easily as you read a book. Many thanks to the volunteers at MGH for letting Sahin poke around their heads. Creepy (or cool) as it may have been for them, it could one day lead to amazing technology for everyone.

[photo credit: Ned Sahin]