An information analysis of the interaction between sensory signals and ongoing cortical activity using a novel mechanistic cortical model, behavioural and MEG studies.

BF Psychology;QA75 Electronic computers. Computer science;QP Physiology

Dempster, Michael Andrew ; Gross, Joachim

University:University of Glasgow

Department:School of Psychology

关键词: MEG, computer Model, neural model, leaky integrate and fire, mechanistic model, brain rhythms, auditory steady state, perception;   

Others  :  http://theses.gla.ac.uk/4223/1/2013dempsterphd.pdf

来源: University of Glasgow

【 摘 要 】

In this work I present a novel mechanistic cortical model derived from the most current cortical anatomical data. The model is built at the cellular level representing the mean laminar distribution and connectivity of the neo-cortex. From this model I derive an extracellular field potential signal and simulate ongoing cortical activity using top down, local and bottom up sensory input. An information theoretic analysis is applied to the simulation data in the context of a bottom up input. This identifies a relationship between cortico-cortical oscillatory activity across a number of frequencies and the information contained in spiking neurons that have long range afferent connections. From these model predictions three auditory perception experimental paradigms are developed, implemented and analysed. I show that the behavioural data is explained by the model predictions and offer a mechanistic explanation of the effect derived from model behaviour. I perform an information analysis on magnetoencephalography dataacquired from a simple 50 % auditory perception task and demonstrate prestimulus ongoing activity frequency power and phase features facilitate perception. In addition there is evidence of attention related interaction between the auditory and visual early cortices.