Editors' ChoiceNEURAL ENGINEERING

Speaking from the Mind

See allHide authors and affiliations

Science Translational Medicine  27 Jul 2011:
Vol. 3, Issue 93, pp. 93ec118
DOI: 10.1126/scitranslmed.3002934

Brain-computer interface (BCI) systems have been studied by many researchers to extract motor intention from an individual’s brain activity—for example, to control prosthetic limbs—but comparatively few groups have tried to directly decode speech commands from cortical activity with such BCI systems. To this end, Leuthardt et al. report a BCI-based study that allows the potential translation of speech-related cortical (brain) activity to external movement.

Four participants were implanted with standard brain surface electrodes for presurgical clinical brain mapping by using electrocorticography (ECoG). A microscale electrode array was also implanted in one participant so as to study the feasibility of using such miniaturized, high-density brain-surface electrodes for cortical recordings. With very brief training (about 15 min), all of the participants were able to control movement direction (left versus right) of a computer cursor with speech-related cortical activity. Interestingly, for two participants, cortical activities encoding different phonemes (“oo” versus “ee,” for instance) were used to move the computer cursor in opposite directions. Essentially, Leuthardt and colleagues demonstrated that brain activities carrying speech information can be deciphered in a meaningful way by computer algorithms and then used to operate BCI systems. In other words, we can potentially predict the words an individual is trying to generate on the basis of his or her brain activity. It is clear that this system, once optimized, could allow individuals with communication disabilities, such as “locked-in” syndrome, to speak their minds, making a drastic difference in quality of life. In addition to the goal of developing a speech-based BCI system, this line of translational research will also offer new opportunities to further understand the neural basis of speech and language and high-level cognitive functions that are unique to human beings.

E. C. Leuthardt et al., Using the electrocorticographic speech network to control a brain–computer interface in humans. J. Neural Eng. 8, 036004 (2011). [Abstract]

Navigate This Article