Item

Abstract

The brain is characterized by ultra-high structural complexity and massive connectivity, both of which change and evolve in response to experience. Information is processed in both a parallel and a hierarchical fashion, and connectivity is bidirectional and continuously modulated. One major problem in studying such systems is adequately defining elementary operational units, because such modules can be complex systems in their own right. In addition, the synergistic organization of complex systems means that their behavior cannot be reduced to, or predicted from, their components. Traditionally, topographic connectivity between different brain areas has been studied using degeneration methods and anterograde and retrograde tracer techniques. Although such studies have yielded valuable information, they require fixed, processed tissue for data analysis and say nothing about dynamic or effective connectivity. To localize and comprehend the neural mechanisms underlying cognition, we must combine multimodal methodologies to concurrently study both large-scale networks and their components. One possibility is to combine global imaging technologies such as functional magnetic resonance imaging (fMRI) with invasive measurements of the brain’s electrical activity at the microcircuit level. Now, novel MR-visible tracers can be infused into a specific brain region and are transported anterogradely transsynaptically, allowing us to study anatomical connectivity in vivo. Simultaneous direct electrical stimulation (DES) and fMRI (DES-fMRI) let us visualize the networks underlying electrostimulation-induced behaviors, map neuromodulatory systems, and study the effects of regional synaptic plasticity on cortical connectivity. Our own recently developed and optimized “neural event triggered functional MRI” (NET-fMRI) uses multiple-contact electrodes and fMRI to map activations induced by neural events. This lecture will address some central questions in this research. Can we identify networks and study “relationships” between their nodes? Can we make activity maps of such networks that are robustly related to behavior? Ultimately, can we really study emerging properties, such as perception and memory, by tracking the behavior of such small- and large-scale assemblies?