Blood oxygenation level dependent (BOLD) functional magnetic resonance imaging (fMRI) has become an important tool in localizing brain functions in vivo at millimeter to centimeter scale. At this coarse scale, however, pivotal questions about the brain's functional architecture must remain obscured. The applicability of BOLD fMRI for addressing questions down to the submillimeter, columnar scale, is however, severely questioned, as it remains unknown whether and how the foci of the activity derived from BOLD signals are spatially correlated with the electrical activity of the underlying neurons. The overall objective of this proposal is two-fold. First, we aim to define and validate the neural correlate of BOLD by using conjunctive single unit recordings. To this end, we hypothesize that only the early negative BOLD changes following sensory stimulation are indicative of neuronal activity at a columnar scale. The delayed positive BOLD changes on the other hand, are hypothesized to be indicative only for the pattern of overall activation per se, but incapable of discriminating between electrically active and inactive columns. This hypothesis will be tested by detailed single unit recordings in conjunction with ultra-high field (9.4T) fMRI studies in cat orientation columns. In the mammalian cortex, neurons preferring similar receptive field properties are spatially clustered into iso-functional domains in vertical and tangential directions to the cortical surface. To date, the tangential layout of the individual cortical maps has been studied only as averaged images across all cortical laminae due to the limitations of current brain mapping techniques. Therefore, as the second aim of this study, we propose to label the laminar-specific architecture of the multiple cortical maps by utilizing the unique advantages of the ultra-high field fMRI (9.4T) at this columnar scale. We hypothesize that different organization of iso-functional domains takes places within different laminae. Consequently, the spatial relationships between multiple cortical maps are hypothesized to differ across cortical layers. We will test this hypothesis by utilizing the unique advantages of ultra-high field fMRI (9.4T) at columnar resolution in order to yield the laminar-specific architecture of the orientation, ocular dominance, direction, and spatial frequency maps in cat visual cortex. The main contribution of this study will be: first, in establishing a direct correlation between neuronal activity and the spatio-temporal pattern of the BOLD fMRI signals down to the submillimeter scale. Such a direct validation of BOLD in an animal system is imperative, as we move towards columnar-resolution BOLD fMRI in humans, where the means of a direct validation are evidently very limited. Secondly, by determining the rules, which govern the representation of the multiple receptive field properties within a common 3-dimensional neural circuitry, the results of our study will greatly facilitate our understanding of the principle of cortical information representation per se.
| Kim, Dae-Shik (2005) The cutting edge of fMRI and high-field fMRI. Int Rev Neurobiol 66:147-66 |
| Kayser, Christoph; Kim, Mina; Ugurbil, Kamil et al. (2004) A comparison of hemodynamic and neural responses in cat visual cortex using complex stimuli. Cereb Cortex 14:881-91 |
| Olman, Cheryl; Ronen, Itamar; Ugurbil, Kamil et al. (2003) Retinotopic mapping in cat visual cortex using high-field functional magnetic resonance imaging. J Neurosci Methods 131:161-70 |
| Ugurbil, Kamil; Toth, Louis; Kim, Dae Shik (2003) How accurate is magnetic resonance imaging of brain function? Trends Neurosci 26:108-14 |
