This proposal aims to advance our understanding of the quantitative link between hemodynamic responses observed in BOLD functional MR imaging and the underlying neuronal activities. We hypothesize that the BOLD response reflects the sum of synaptic activity in a local region of neural tissue. Combined fMRI and in vivo electrophysiological experiments will be used in awake, behaving rabbits to test this hypothesis. We will focus equally on understanding both the spatial and temporal domains of the BOLD signal, and its relationship to electrophysiological activity at the level of single- and multi- neuron activity, and of the local field potential. Investigation of the relationship between BOLD and electrophysiological responses will be conducted with two paradigms: simple stimulus-response using whisker stimulation alone, and a complex learning related paradigm using trace eyeballing conditioning with whisker stimulation as the conditioning stimulus. Stimulation of the well-characterized vibrissae barrel system will be used in the first paradigm to examine the dependence of response properties on stimulation parameters including changes in amplitude, rate and duration. The spatial origin and temporal characteristics of the controversial initial """"""""dip"""""""" in BOLD fMRI will be investigated using high spatial and temporal resolution fMRI. The relationship between neuronal input and output measured as neuronal spiking and local field potentials will be compared with hemodynamic responses measured as BOLD signal evolution in individual cortical layers of identified barrel columns. The second paradigm will study learning acquisition and consolidation while focusing on the hippocampus, somato sensory cortex and cerebellum. These regions function as the site of temporary memory storage, permanent memory storage, and the basic essential associative site, respectively during the course of learning. The inverse activity relationship of the cerebella cortex and deep nuclei during learning will be used to examine activation and deactivation simultaneously in two different regions. Our comparison of simple and cognitively-demanding paradigms will verify if the neural-hemodynamic dependence is valid across diverse levels of neural activity, and if BOLD fMRI can potentially be used as a noninvasive assay of cognitive functioning.
| Li, Limin; Wyrwicz, Alice M (2016) Modularized architecture of address generation units suitable for real-time processing MR data on an FPGA. Rev Sci Instrum 87:063705 |
| Li, Limin; Wyrwicz, Alice M (2015) Design of an MR image processing module on an FPGA chip. J Magn Reson 255:51-8 |
| Aksenov, Daniil P; Li, Limin; Miller, Michael J et al. (2015) Effects of anesthesia on BOLD signal and neuronal activity in the somatosensory cortex. J Cereb Blood Flow Metab 35:1819-26 |
| Li, Limin; Weiss, Craig; Talk, Andrew C et al. (2012) A MRI-compatible system for whisker stimulation. J Neurosci Methods 205:305-11 |
| Song, Xiaomu; Li, Limin; Aksenov, Daniil et al. (2010) Mapping rabbit whisker barrels using discriminant analysis of high field fMRI data. Neuroimage 51:775-82 |
| Song, Xiaomu; Wyrwicz, Alice M (2009) Unsupervised spatiotemporal fMRI data analysis using support vector machines. Neuroimage 47:204-12 |
