Temporal Dynamics of Endogenous and Stimulus-Driven Attention: Combined ECoG/fMRI Studies

Selective attention allows us to filter out irrelevant sensory information in the environment and focus neural resources on information relevant to our current goals, while being able to flexibly shift our focus to potentially rewarding or harmful stimuli. Functional brain imaging studies have identified networks of broadly distributed brain regions that are recruited during goal-driven attention (i.e. based on internal expectations or goals) and/or stimulus-driven attention (i.e. driven by salient or unexpected stimuli); however, the dynamics by which these networks enable selection of attended sensory information are not well understood due to the low temporal resolution of functional neuroimaging. Here, we first used functional MRI to localize attention-related and other task-relevant and -irrelevant brain networks in human epileptic subjects, prior to localization of their seizure foci using electrocorticography (ECoG), electrodes placed directly on the cortical surface. We subsequently recorded cortical physiology from the ECoG electrodes during a spatial attention task, involving both goal-driven and stimulus-driven attention, and co-registered electrode positions with the fMRI-defined networks to study network-specific dynamics during these two processes. We found that low frequency local field potential (LFP) oscillations, which are thought to reflect fluctuations in local neuronal excitability, became selectively phase modulated over task-relevant brain regions/networks during the same task epochs in which they are recruited in fMRI. This mechanism may alter the excitability of task-relevant regions or the effective connectivity between them to enable selective neural processing of attended stimuli. Furthermore, different attention processes (holding vs. shifting attention) were associated with phase modulations at different frequencies, possibly to multiplex different cognitive processes and minimize unnecessary cross talk between unrelated neuronal populations