Prof. John T. Hale | Modeling neural time courses with linguistic structure

Guest Lecture

  • Date: Apr 30, 2019
  • Time: 15:00 - 16:00
  • Speaker: Prof. John T. Hale
  • Department of Linguistics, University of Georgia, Athens, USA
  • Location: MPI for Human Cognitive and Brain Sciences
  • Room: Wilhelm Wundt Room (A400)
Human language comprehension offers a grand challenge for the cognitive and brain sciences. What is it that our brains do, such that a stream of words comes together to yield a communicative or literary experience?

Modeling neural signals offers a particular angle from which to approach this grand challenge. With functional neuroimaging, it is possible to extract a timecourse of brain activity from particular regions and ask how well alternative (psycho)linguistic theories account for the measured signal. This approach can be carried out over prolonged periods, for instance during the spoken recitation of a literary text. It offers the opportunity to observed quite detailed facets of linguistic information processing that are repeated many times in a naturalistic stimulus.

This talk considers four specific predictors that are motivated by longstanding theoretical ideas in cognitive science. Formalizing these predictors and scaling them up using techniques from computational linguistics we ask about the functional neuroanatomy of language processing within the brain.

The results cohere well with prior experimental findings and exemplify a quite general method for inferring processing mechanisms during ecologically-natural episodes of cognition.

Poster
Go to Editor View