Cortical Networks for Flexible Categorisations

The fundamental problem for cognitive neuroscience is to describe how perceptual and cognitive states arise from states of the brain. To understand the relationships, functional imaging techniques such as PET and fMRI can address the where question of functional anatomy: Which areas of the brain show increased activity when a particular task is being resolved and psychological functions solicited? But brain processes are not static, they are dynamic, and the questions of when and how perceptions and cognitive activities arise are as essential as where they arise. In principle, EEG and MEG provide the millisecond temporal resolution that is necessary to study the dynamics of perceptual and cognitive states. Furthermore, new developments in spatial filtering methods have recently enabled a reliable cortical localization of MEG signals, which are less distorted by the intervening tissue than EEG signals. Lastly, and central to this research proposal, high temporal resolution allows for an optimal decomposition of the measured MEG signals into frequency components, in order to analyse oscillatory activity and coupling in relation to cognitive tasks. Here, we will render the a flow-chart of information processing in the brain for the main nodes of the cortical processing networks associated with biologically important face categorisation tasks (i.e. gender and expression).

A proposal with a long history suggests that the dynamics of psychological states is related to the oscillatory activity of the neural substrate (see VanRullen & Koch, 2003; Ward, 2003, for recent reviews). Simply stated, synchronized discharges of cortical cell assemblies with a prominent rhythmicity would be the neural correlates supporting discrete cognitive and perceptual states. This is based on evidence of a consistent relationship between behavioural variables and the frequency of both the theta (4-8 Hz) and alpha rhythms (8-12 Hz) in memory and perceptual tasks. An important task, then, it is ascribe information content to the measurable parameters of oscillatory brain activity. Here, we will for the first time merge methods to identify the form of cognitive states and ascribe them information content. To isolate form, we will use the proven MEG DICS methods for localising oscillatory cortical networks (e.g. Gross et al., 2001; Gross et al., 2004). For ascribing content, we will use Bubbles (Schyns et al., 2003; Smith et al., 2004; Smith et al., 2005). To tap into biologically important cortical information processing networks we will use biologically-relevant face categorization tasks (e.g. gender and categorization of expressions). Our goal is to provide a flow-chart of information processing in the brain for the main nodes of the cortical processing networks associated with each one of the face categorization tasks tested.