NEUCOD

Information-theoretic approach to infer encoding patterns in a decision making processes


Abstract

The question of how stimuli and actions are encoded in the neuronal signals of the brain is at the central core of neuroscience research. In particular, very little is known about the mechanism for which a given decision posterior to a stimulus is manifested in the brain activity. Nowadays, simultaneous recordings of neurons in different brain regions bring an exceptional dataset to investigate the complex encoding mechanisms that underlie a decision making process. In the present project, the candidate will have exclusive access to one of these datasets which consists of brain recordings in primates during a decision making task. The candidate will employ his expertise in some areas of information theory and coding to formulate a novel hypothesis to be tested on these available data. This hypothesis states that causal interactions between neurons in a population encode information about decisions. This interdisciplinary line of investigation represents a changing point in the candidate’s career. On the one hand, the candidate will be trained in the neurophysiological aspects of the problem. On the other, he will develop emerging methods to capture causal interactions between neuronal responses and then represent the hypothesized “encoding patterns”. From a scientific point of view, this project will provide one of the key tools to understand information processing in a decision-making process and study sources of brain disorder based on the dynamics of encoding patterns. More generally, the present project will launch the candidate’s career in an interdisciplinary field that has potential application in academia and industry research, as well as will foster the collaboration between the information theory and the computational neuroscience communities, thus making this research area in Europe more competitive and attractive for future students and researchers.

Marie Curie fellow: Adrià Tauste-Campo

PI and Supervisor: Gustavo Deco

Funded by: European Union’s Seventh Framework Programme for research, technological development and demonstration. EU Marie Curie Intra-European Fellowship.

Project reference: 329837