Functional Magnetic Resonance Imaging (fMRI) has rapidly become a desirable neuroimaging tool for cognitive scientists and clinicians, because of the tremendous advantage of non-invasively assessing human brain function. Hundreds of successful initial fMRI studies on clinical MRI machines have used robust paradigms which elicit measurable signal changes at modest image resolution. However, there exists considerable controversy over the ultimate spatial resolution and functional specificity of the BOLD technique, because neither the underlying mechanisms of the cerebral vascular response nor their effect on the BOLD signal are understood. Therefore, the investigators' objective is to establish the specificity of the cortical vascular response to well defined units of neuronal activity, as well as determine the sensitivity of fMRI to BOLD changes in such basic units of cortical organization. This should allow the extension of fMRI techniques to the level of coarse electrophysiology in humans. In this grant, the investigators propose to develop techniques necessary to image ocular dominance columns in the human visual system. These columns form an unique and ideal in-vivo resolution system for fMRI since they constitute point sources of neuronal activation that can be driven by a simple paradigm. The investigators' hypothesis, for which they will present preliminary supporting evidence, is that fMRI can be performed at the resolution of coarse electrophysiological measurements. This implies that the vascular response is appropriately localized to functional subunits. In order to test this hypothesis, the investigators will develop methods to (1) increase image SNR using ultra-high field fMRI and phased array coils to the point that thermal (Johnson) noise is not the dominant source of image-to-image fluctuations in high resolution images, (2) assess in a quantitative statistical manner the degree to which the brain and body motions and instrumental instabilities contribute to the remaining inter-image fluctuations by studying signal magnitude and phase in consecutively acquired fMRI images using several different imaging pulse sequences and (3) develop new hardware and software strategies for the compensation of the quantities underlined in (2) in order to reduce the remaining image-to-image fluctuations and consequently reduce the detection threshold for high resolution cortical mapping using fMRI.
