Essentials: The Neuroscience of Speech, Language & Music | Dr. Erich Jarvis

The conversation opens with the host asking how the brain organizes speech and language, and whether the two should be thought of as similar or different. Jarvis rejects a sharp split. He describes a common view in which a separate language module holds the algorithms and computations that then drive a speech production pathway and an auditory perception pathway, and says he does not think there is good evidence for that arrangement.

I don't think there is any good evidence for a separate language module.

His alternative is that the complex algorithms for spoken language are built into the production pathway itself, the circuitry controlling the larynx and jaw muscles, while the algorithms for understanding speech are built into the auditory pathway. The two pathways differ in how widely they are shared across species. The production pathway, he says, is specialized to humans, parrots and songbirds, while auditory perception is more ubiquitous across the animal kingdom.

That asymmetry explains a familiar observation: animals can understand far more speech than they can produce. Jarvis says dogs can understand several hundred human words, and great apes can be taught several thousand, yet the apes "can't say a word."

Asked about communication that resembles language without being language in the classic sense, Jarvis points to gesture. The brain regions that control spoken language sit next to the regions for gesturing with the hands, and the hand pathway runs its own complex algorithms. Species differ in how advanced these circuits are. Humans are the most advanced at spoken language, but he says the gap between humans and some other species is not necessarily as large for gestural language.

His everyday evidence is that people gesture while they talk without noticing, even on the telephone when nobody can see them. He agrees with the argument that speech production and gesturing are evolutionarily related, and offers a specific hypothesis to explain why the regions are adjacent.

I think that the brain pathways that control speech evolved out of the brain pathways that control body movement.

He adds that languages such as Italian, French and English each come with a learned set of gestures. His comparative example is Koko, a gorilla raised with humans for 39 years or more, who learned gesture communication or sign language and could understand both signing and spoken speech, but could not produce speech sounds with her voice. His interpretation is that a number of species, though not all, have motor pathways for learned gesturing, "rudimentary language" with the limbs, but lack the extra pathway for sound. In his words, they "can't gesture with their voice in the way that they gesture with their hands."

The host asks whether primitive emotions and sounds might be the early substrate of language. His examples are inhaling and saying "Mm" at a delicious smell, and wincing, turning away and exhaling at a putrid one. He frames these as possible "basic dark and light contrasts" of a pyramid running from primitive sound up to language, and asks whether any evidence supports it.