Understanding How We Speak

A new study revealed the patterns of brain activity that produce human speech. The findings may one day lead to new approaches for treating speech disorders.

Most of us barely give a conscious thought to the process of speaking, but it’s one of the most complex actions we perform. In order to speak, your brain needs to quickly and precisely coordinate your lips, jaw, tongue and larynx (voice box). Speech disorders, such as stuttering, affect roughly 5% of children by the first grade. The underlying causes of most speech disorders, however, aren’t well understood.

Past studies have found that a part of the human brain called the ventral sensorimotor cortex, or vSMC, controls speech. Using electrical stimulation, researchers were able to discover which general areas of the vSMC controlled which parts of the face and mouth. However, this kind of stimulation couldn’t evoke meaningful utterances. The finding implies that rather than being stored in discrete brain areas, speech sounds might arise from coordinated motor patterns involving multiple areas.

A team led by Dr. Edward Chang of the University of California, San Francisco, set out to better understand speech processing in the brain. To do this, they recorded neural activity from the brain surfaces of 3 people who were implanted with electrode arrays as part of their preparation for brain surgery. The electrical recordings were matched with microphone recordings as the subjects read syllables aloud. The study was funded in part by NIH’s National Institute of Neurological Disorders and Stroke (NINDS) and National Institute on Deafness and Other Communication Disorders (NIDCD). Results appeared online on February 20, 2013, in Nature.

The researchers identified about 30 active electrode sites per subject. Analysis of the recordings revealed that different sounds were coordinated in dynamic, complex patterns of activity involving different brain regions. Electrical patterns in the brain transitioned within tens of milliseconds between distinct representations for different consonants and vowels.

Regions of brain activity during speech have a hierarchical, overlapping structure organized by phonetic features, the scientists found. For example, consonants that require similar tongue locations have overlapping areas of activity. The patterns of brain activity differ most between consonants and vowels. The researchers say that this may explain why slips of the tongue usually involve substituting consonants for consonants, or vowels for vowels, but very rarely mixing them up with each other.

“Even though we used English, we found the key patterns observed were ones that linguists have observed in languages around the world — perhaps suggesting universal principles for speaking across all cultures,” Chang says.

This work sheds light on a unique human ability: the power of complex speech. The findings may help guide potential treatments for speech disorders and the development of brain-computer interfaces for artificial speech.

— by Harrison Wein, Ph.D.