Protocol number 93-M-0170, NCT00001360 High resolution fMRI for differentiation of layers and columns: As Dr. Laurentius Huber was finishing up in our lab towards the end of 2018, he continued to spearhead our high resolution functional magnetic resonance imaging (fMRI) development. He has been refining MRI pulse sequences to be more time efficient and to be more selective to small vessel hemodynamic changes which are in closer vicinity to the locations of neuronal activity. His sequences rely primarily on the use of Vascular Space Occupancy Contrast (VASO), which is sensitized to blood volume changes. His latest papers demonstrated increased layer sensitivity and specificity at 9.4 Tesla and a discovery of a novel digit representation in humans. The 9.4 Tesla work was carried out in collaboration with the University of Maastricht and is published in NeuroImage. The paper on digit representation discovered that digit organization in primary motor cortex, M1, is organized separately for grasping and extending movements, with the two different digit representations showing a mirror image organization. Layer-specific fMRI of a cognitive task In the last year, we have optimized and extended layer-specific fMRI methods for higher-order brain regions. Whereas previously, the feasibility of these methods had been demonstrated only in primary modal cortex, we have developed new acquisition and post-processing techniques that allow us to acquire high-quality data in prefrontal cortex. We used these techniques to observe depth-dependent activity in human dorsolateral prefrontal cortex during a working memory task. We were able to distinguish activity in upper layers, which was higher during the delay phase of the task, from that in deeper layers, which was higher during the response phase. A manuscript on these results is in press at Nature Neuroscience. Alternative layer-specific fMRI pulse sequences: We continued to develop new fMRI contrast that can be used to measure brain activity with better specificity and sensitivity, aiming to distinguish activity across cortical layers and columns. We developed a new VASO and Perfusion integrated contrast method and implemented it at 7T. We have applied it to study the layer-dependent activity in primary motor cortex and auditory cortex successfully. We also developed a sequence to acquire a T1-weighted echo-planar image with distortions that exactly match fMRI allowing direct registration. Cognitive correlates of dynamic functional connectivity during rest The dynamical aspects of the human functional connectome have gained attention in recent years, fueling debates regarding its etiology, interpretation and potential clinical value. Some consider this phenomenon neuronal and have started to investigate its developmental, behavioral and clinical correlates. Others believe it is merely a manifestation of fluctuations in arousal, sleep and head motion. Previous work from our lab has demonstrated that dynamic functional connectivity (dFC) is related to externally-driven cognition. Here, we build on these observations to explore the hypothesis that time-varying functional connectivity during rest is a manifestation of an internal flow of covert cognition. We have performed novel experiments this year which suggest that, indeed, rest dFC is influenced by short periods of spontaneous cognitive-task-like processes, and that the cognitive nature of such mental processes can be inferred blindly from the data. Blind deconvolution of activity Inducing events from multi-echo BOLD Signals Over the years, our lab has devised multiple successful techniques to model and remove several noise sources; with a strong emphasis on multi-echo methods in recent years. In single-echo fMRI, one time series is acquired per voxel at a single echo time (TE) known to maximize BOLD contrast. Conversely, in multi-echo (ME) fMRI, N different signals are recorded concurrently per voxel; each of them at a different TE. Because the amplitude of BOLD and non-BOLD fluctuations in fMRI recordings have distinct TE-dependence profiles, it is possible to use this additional information gathered to automatically identify and remove most noise sources. This year, we have worked on the development of a ME-based deconvolution technique which allows the detection of discrete activation events in fMRI. In collaboration with Dr. Caballero-Gaudes, we reformulated an existing single-echo deconvolution technique (SPFM) for its applicability to the ME framework. Evaluation of this novel method in a multi-task event-related paradigm demonstrate that maps of R_2* changes obtained with ME-SPFM at the times of the stimulus trials show high spatial and temporal concordance with the activation maps and BOLD signals obtained with standard model-based analysis. Naturalistic tasks / Individual differences In the last year, we have made advances in collecting and analyzing fMRI data while participants are engaged in naturalistic paradigms (i.e., watching a movie or listening to story). We are exploring the possibility of using such tasks as brain stress tests to draw out individual variation as it relates to intrinsic personality and cognitive traits. Specific accomplishments in the past year include the following: 1) We have developed and used inter-subject representational similarity analysis to model how similarity of brain activity during movie watching is related to behavioral similarity across subjects to demonstrate that individuals with more similar behavioral phenotypes have similar spatial-temporal patterns of brain activity. We are now working on adapting this approach to predict cognitive and personality measures from brain activity during naturalistic stimulation in previously unseen subjects. A manuscript on these results is in preparation for submission to NeuroImage. 2) We have developed a new statistical framework for analyzing naturalistic scanning data. While inter-subject correlation (ISC) analysis is a powerful tool for naturalistic data, drawing appropriate statistical inferences is difficult due to the daunting task of accounting for the intricate relatedness in data structure as well as the intricacy of handling the multiple testing issue. We developed a Bayesian multilevel (BML) framework for ISC data analysis that integrates all the regions of interest into one model. We illustrated the power of this approach to detect effects of age, sex, and a measure of social function on ISC. A manuscript on these results is under review at NeuroImage. 3) We have developed and piloted a new experimental paradigm that combines naturalistic tasks during fMRI with detailed behavioral and phenotypic assessment. We have conducted a large-scale online experiment to determine the optimal stimuli to use to draw out the most individual variability in both neural and behavioral responses. We are currently in the midst of a full-scale data collection effort using these stimuli on both healthy controls and psychiatric patients.
Showing the most recent 10 out of 84 publications
