Selective attention is the process by which the brain enhances its representation of task-relevant input at the expense of irrelevant input, and it is essential to adaptive behavior. A critical, yet poorly understood component of attention, is the ability to deploy processing resources in advance of anticipated stimuli. Such preparatory attention goes to the very heart of the brain's capacity to predict, and to organize its data gathering and processing accordingly. Our broad goal is to provide incisive information on neural mechanisms of preparatory attention by capitalizing on the resolution provided by direct electrocorticographic (ECoG) and Utah (multiple microelectrode) Array recording within the human brain in surgical epilepsy patients. This novel combination of macro- and micro- electrophysiological methods provides an unprecedented level of resolution and a unique opportunity to advance the understanding of the brain mechanisms of preparatory attention. Using a novel combination of high-resolution ECoG methods, we will address two propositions central to the current debate on how the brain anticipates and predicts. The first proposition is that neuronal ensemble excitability fluctuations (i.e., oscillations) in mid and high frequencies play complementary mechanistic roles in spatially- directed preparatory attention. Prior studies suggest that: 1) attention-related increase in gamma (30-50 Hz) synchrony reflects spatially-selective enhancement of a sensory input representation, while associated high gamma (>90 HZ) power indexes related neuronal firing patterns, and 2) alpha (8-14 Hz) oscillations reflect an attentional suppression of task-irrelevant input. We hypothesize that in preparatory attention, combinations of these mid and high frequency neuronal oscillations are deployed as complementary instruments to amplify the representation of anticipated task-relevant inputs and to suppress that of irrelevant inputs. The second proposition is that lower frequency (delta/theta: 1-7 Hz) oscillations play a role in temporally-directed preparatory attention. Prior findings show that when a task-relevant event stream is rhythmic and predictable, these oscillations, entrained to the temporal structure of the stream, can both amplify neuronal responses to the events in that stream and suppress responses to out of phase (irrelevant) events. We hypothesize that in adapting to ever-changing task demands, ranging from rhythmic (e.g., listening to music) to random (e.g., waiting for a traffic light), the brain shifts dynamicaly between rhythmic and random sampling strategies; in rhythmic mode, low frequency dynamics, keyed to the pace of the task, dynamically regulate higher frequency preparatory activity, while in random mode, higher frequencies operate more continuously, albeit perhaps less effectively. While both modes likely entail top-down control via prefrontal modulation of sensory areas, there is indication that the rhythmic mode also strongly engages motor cortex. If successful, this project will bridge the gap between noninvasive ERP and MEG studies of preparatory attention in humans and more incisive studies at cellular and cell ensemble levels in monkeys.
Selective attention is essential to adaptive behavior, and its disruption contributes significantly to a range of psychiatric disorders including autism and schizophrenia. This proposal targets a critical, yet poorly understood component of attention, preparatory attention. It will yield information critical to improving the understanding and treatment of these disorders.
