Neuronal oscillations as a tool for attentional stream selection

Columbia University Health Sciences, New York NY

Investigators

Linked publications & trials

Abstract

DESCRIPTION (provided by applicant): The ability to attend to and select a particular stream of input amidst those flooding our senses is a critical cognitive capacity. My broad, LONG TERM OBJECTIVE is to define the neuronal mechanisms by which the brain accomplishes this selection. The striking similarity between the rhythmic structure of many biologically- generated acoustic stimuli, such as natural speech, and of that ongoing rhythmic excitability fluctuations in the brain (i.e., neuronal oscillations) prompts my core hypothesis - that these oscillations themselves are tools that can be used to form a neural representation of the rhythmic structure of an attended stream, and that attention- enforced entrainment of neuronal oscillations is a key mechanism for sensory and perceptual selection. To evaluate these ideas, I will record electroencephalographic (EEG) signals from implanted electrodes in and surrounding auditory cortex in surgical epilepsy patients, while they engage in a series of auditory stream selection tasks. Natural speech is a prototypic example of a rhythmic biologically-generated acoustical stream, whose processing in real life situations often demands selective attention and the ability to ignore irrelevant background noises, as in the well known cocktail party situation. Thus natural speech would be a prime example for testing my core hypothesis of attention-enforced entrainment. However this endeavor is enormously complicated by top-down phonologic, semantic and emotional influences that are also invoked during speech processing. To bypass these problems and yet get directly at the issue of attentional selection based on speech's underlying rhythmic structure, I will the use semantically meaningless sound streams, composed so as to both dissect key issues in stream selection, and mimic key aspects of rhythmic structure in speech. My FIRST SPECIFIC AIM is to define minimal rhythmic conditions for attention-enforced entrainment, and their impact on behavioral performance. I will test a fundamental proposition that attention operates in two distinct modes based on the temporal structure of the input: a rhythmic mode, in which low frequency oscillatory entrainment to the temporal structure of the attended stream is predicted to operate as an instrument of attentional selection, and a random mode, in which the input stream has no detectable rhythm; in this case, I predict that low frequency oscillations will be suppressed, as they would then be a liability to stimulus discrimination. My SECOND SPECIFIC AIM is to examine selective entrainment to streams with hierarchical rhythmic structures mimicking key components of speech. Investigating neuronal oscillations as instruments of selective attention may improve our understanding of attentional dysfunction in a broad spectrum of neuropsychiatric disorders including ADHD and schizophrenia, in which affected individuals reportedly have reduced capacity to entrain neuronal oscillations to rhythmic stimuli. More directly, and perhaps more importantly, findings may also be helpful in understanding attentional deficits that are sometimes observed in epilepsy patients, which is the population participating in the proposed study.