Hemodynamic-based brain-mapping techniques, including functional magnetic resonance imaging (fMRI), are widely used for clinical and basic neuroscience research. However, the exact relationship between neural activity and the hemodynamic response with regard to spatial extent and amplitude is not yet clear. A few studies have examined this issue with the most widely used fMRI technique, which is based on blood oxygenation level dependent (BOLD) image contrast. However, the BOLD effect has a complex mechanism that depends on alterations in cerebral metabolic rate of oxygen (CMR02), cerebral blood flow (CBF), and cerebral blood volume (CBV) in response to increased neuronal activity. More importantly, the spatial specificity of the conventional BOLD signal is hampered by contributions from large draining vessels, which can be a few centimeters from neuronally active sites. Further, BOLD contrast depends on vascular dimensions and geometry, as well as imaging techniques (e.g., gradient-echo, spin-echo) and experimental parameters (e.g., static magnetic field strength, echo time). Thus, a correlation between neural activity and the BOLD effect found in one experimental condition cannot be easily generalized to other conditions. Therefore, we propose to determine (i) the spatial correspondence between neural activity and tissue-specific CBF-based fMRI devoid of large vascular contributions and (ii) the quantitative relationship between neural activity and CBF, which is independent of magnetic field strength and imaging parameters. We will investigate these issues using the well-established cat orientation column model, which has been extensively investigated with single-unit recording, 2-deoxyglucose (2-DG) autoradiography, and optical imaging, and which has already been implemented in the PI's laboratory. Since individual orientation columns are separated by a) 1- 1.4 mm, it is an ideal model for our proposed studies.
Showing the most recent 10 out of 50 publications
