Last, year, 2004, was the 5?th year of existence for the Unit of Functional Imaging Methods, directed by Peter A. Bandettini, in the Laboratory of Brain and Cognition. The overall theme of the unit is to develop more robust, informative, and quantitative methods for mapping human brain function. The specific research goals of the unit include the following: A) Development of a better understanding of the neuronal correlates of fMRI signal changes. B) Development of methods for extraction of quantitative physiologic information from the fMRI signal. C) Demonstration and characterization of neuronal current contrast in MRI. D) Development of integrated paradigm design, behavioral, and processing methods allowing improvement of the quality and usability of information obtain from fMRI time series data. E) Development of methods to assess and use ?resting state? fMRI data (both the baseline flluctuations in the time series and the baseline anatomic images). Six of the most interesting findings over the past year include: 1. In vitro detection, using MRI at 3T and MR proton spectroscopy at 7T, of changes in signal that appear to be correlated with spontaneous and synchronous spiking activity of rat fetal brain tissue. This is ongoing work in collaboration with Dr. Dietmar Plenz and Dr. Afonzo Silva. We are working to confirm our findings by modulating neuronal activity and performing other similar experiments with direct electrical recordings. If we confirm that the source of the MRI signal changes are small magnetic fields induced by the propagation of currents within neurons. This improvement potentially represents the introduction of a new way to use MRI for understanding brain activation: that of a new method for detecting neuronal activity directly. 2. We have created of a general framework for the characterization of the effects of overt movement on fMRI signal changes as a function of paradigm timing. This work has resulted in an in press manuscript, authored by Dr. Rasmus Birn. 3. We have shown that fMRI can resolve cascaded activity with the use of a novel brand of paradigm known as a """"""""deadline paradigm"""""""" in which the duration of the stimuli is systematically increased in such a way that the subjects? performance of a task involving the stimuli improves as a function of stimulus duration. Using this, we are able to titrate the neuronal correlates of this performance improvement by observation of the BOLD activation maps that correspond to each stimulus duration. Anthony Boemio, Ph.D. has developed this method in the context of separating early vs late processing with house vs. face stimuli. These results appear similar to those achieved with MEG. 4. We have shown that an individual can be conditioned without being consciously aware of the stimulus. This behavioral finding was the result of a study originally intended to simply be a pilot study for an fMRI study. The fMRI study is currently being conducted to characterize the neuronal circuit involved with conditioning and awareness in this context. This study was carried out primarily by David Knight and so far has resulted in a paper published in PNAS. 5. We have shown the brain region presumably associated with comparing stimuli during perceptual decision-making tasks. The task was to decide whether the stimulus (having modulated degrees of scrambledness) was a face or a house. In the easy tasks, the respective ?face? or ?house? regions were observed to activate. On comparing the difference between the activation of the two extreme conditions (easy house or face vs ambiguous house face) we found that a specific region (left dorsal lateral prefrontal cortex) correlated with the magnitude of this difference in activation between the face and house regions. This is suggestive that this region is associated with integrating information from each region associated with the decision. This work, carried out by Hauke Heekeren, resulted in a manuscript recently accepted in Nature. 6. We have derived a method for determining the ?optimal resolution? based on the signal to noise and minimum desired activation change to be detected. These studies have led us to develop methods for separation of physiologic from scanner and/or thermal noise. This work is primarily being carried out by Jerzy Bodurka and Natalia Petridou.
Showing the most recent 10 out of 36 publications
