A Brain for Language and the Language of the Brain

Author

Guillermo A. Cecchi

Date of Award

1999

Document Type

Thesis

Degree Name

Doctor of Philosophy (PhD)

RU Laboratory

Magnasco Laboratory

Abstract

This thesis consists of three sections. In the first one, we overview the basic concepts and results of computation theory, the relationship with formal language theory, and present an introduction to the possible applications to different biological systems. The second section deals with the role of noise in the transmission of information in neurons. We show that a key component of the dynamics of action potential generation in neurons, the rectification of ionic flow and negative incremental resistance to ionic gradient, can occur in Brownian transport in the environment of the cell by a purely entropic effect, i.e. driven by thermal noise. We also focus on the problem of temporal coding of information by neurons. vVe show that a wide class of models of spiking elements display a high sensitivity of the output message noise to the particulars of the input message. We show that predictions based on these models are corroborated by measurements of spiking activity in the visual geniculate of cats. The last section is centered in the problem of coding of complex spaces by the brain. We investigate a case of animal communication, that of songbirds. We show that there exists in the auditory brain of canaries a topographic representation of song, whereby song components are mapped into unique patterns of activity, preserving stimulus identity as well as inter-stimuli relationships. We postulate that this representation reveals a song code embedded in the topography of neural space. We discuss different theories of neural code, their relevance to the findings of syllabic representation, and preliminary approaches to models of brain function compatible with these results. These approaches are based on the hypothesis of the existence in auditory areas of birdsongs of a self-organizing neural network that, driven by statistics of the input, constructs a consistent representation of the auditory input ensemble in neural space.

Comments

A thesis presented to the Faculty of Rockefeller University in partial fulfillment of the requirements for the degree of Doctor of Philosophy

Recommended Citation

Cecchi, Guillermo A., "A Brain for Language and the Language of the Brain" (1999). Student Theses and Dissertations. 443.
https://digitalcommons.rockefeller.edu/student_theses_and_dissertations/443