Spatiotemporal structure of perception

<strong>JAWAHARLAL NEHRU UNIVERSITY SCHOOL OF COMPUTATIONAL AND INTEGRATIVE SCIENCES</strong> S E M I N A R <strong>Dr. Arpan Banerjee</strong> Cognitive Brain Lab &amp; Centre for Excellence in Epilepsy and MEG, National Brain Research Centre, India. <strong>Spatiotemporal structure of perception</strong> Date and Time : <strong>23rd September, 2015 </strong> <strong>Abstract: </strong>Several neuroimaging supports the view that perceptual processing of environmental stimuli requires harmonious interaction of large-scale networks composed from distributed and interconnected neuronal populations. Typically these networks are omnipresent in neurophysiological data such as electroencephalogram (EEG) / magnetoencephalogram (MEG) and metabolic data such as functional magnetic resonance imaging (fMRI) and have been defined as "neurocognitive" networks. Our lab is engaged in developing techniques for visualization and quantitative characterization of neurocognitive networks as well as development of experimental paradigms for studying cognitive behavior. I will present an example of a fMRI experimental paradigm in speech perception where a neurocognitive network in the brain involving auditory cortex (for sound processing), visual cortex (processing vision), parietal cortex (attention), inferior frontal areas (processing linguistic representations) and posterior superior temporal sulcus (processing audio-visual stimuli) interacts for different perceptual categories. Interestingly, the presence of multisensory (such as audio-visual) stimuli doesn't guarantee the activation of multisensory brain areas but the context such as synchronous or asynchronous nature of the stimuli plays a huge role. Subject response behavior close to synchronous presentation of stimuli was markedly different from asynchronous case, featuring creation of illusory perceptual objects (speech sounds). When audio visual signal onsets are separated in time, lower activations in multisensory areas were observed along with a decrease in coupling strengths among cortico-cortical and sub-cortical to cortical connections. Some preliminary EEG data will be presented from this paradigm to show how neuro-cognitive network dynamics transcends scales of space and time. Finally, a computational model will be introduced for mechanistic explanation of behavior and neural dynamics.