Stanislas Dehaene

Symbols and languages: a hypothesis about the singularity of the human brain
Stanislas Dehaene, Collège de France and Institut national de la santé et de la recherche médicale (INSERM)

08:30 - 09:30 BST, Monday 24th April 2023

Natural language is often seen as the single factor that explains the cognitive singularity of the human species. Instead, my hypothesis is that human singularity is manifested much more broadly, in the ability to acquire a great variety of symbolic languages, akin to computer languages, to encode and compress structures in various domains (mathematics, music, shape…).

These languages rely on cortical circuits parallel to, but distinct from classical language areas. Each is characterized by (1) the discretization of a domain using a small set of symbols, and (2) their recursive composition into mental programs that encode nested repetitions with variations.

I will present several tasks of elementary shape or sequence perception in which minimum description length in the proposed languages captures human behavior and brain activity. Functional MRI localizes the relevant parallel networks, including activations close to but distinct from the language-related Broca’s area, while magneto-encephalography tracks the postulated mental structures in real time.