Manus Donahue^1,2, Jeroen Siero^2,3, Hans Hoogduin², Natalia Petridou², Peter Luijten², and Jeroen Hendrikse^2
1Radiology and Radiological Sciences, Vanderbilt University, Nashville, TN, United States, ²Radiology, University Medical Center Utrecht, Utrecht, Netherlands, ³Rudolf Magnus Institute, University Medical Center Utrecht, Utrecht, Netherlands

The overall goal of this study was to investigate negative elements of the blood oxygenation level-dependent (BOLD) hemodynamic response function as a function of cortical depth and stimulus duration by performing high spatial (1.35 m in-plane) and temporal (TR=600 ms) BOLD fMRI in humans at 7 Tesla. Findings indicate that the initial dip is dependent on both cortical depth and stimulus duty cycle, with the largest dip present in peripheral cortical regions and in the first block of event-related paradigms. The post-stimulus undershoot was found to be largest in deeper cortical regions where capillary density is highest.

Natalia Petridou^1,2, Brian Wandell³, Ben M Harvey⁴, Wietske Zuiderbaan⁴, Peter Luijten¹, and Serge O Dumoulin^4
1Radiology, UMC Utrecht, Utrecht, Netherlands, ²Rudolf Magnus Institute, UMC Utrecht, Utrecht, Netherlands, ³Psychology, Stanford University, Stanford, United States, ⁴Experimental Psychology, Helmholtz Institute, Utrecht University, Utrecht, Netherlands

Histological evidence in humans and non-human primates suggests a functional sub-division of extra-striate visual areas V2/V3, with a patchy or ‘stripe’-like pattern of neurons belonging to different information pathways of the visual system. Here, using 3D EPI BOLD at 7T and a dedicated surface coil, we show in vivo evidence of these sub-divisions in humans.

Jozien Goense¹, Hellmut Merkle², and Nikos K Logothetis^1,3
1Physiology of Cognitive Processes, Max-Planck Institute for Biological Cybernetics, Tuebingen, Germany, ²Laboratory of Functional and Molecular Imaging, NINDS/NIH, Bethesda, MD, United States, ³Division of Imaging Science and Biomedical Engineering, University of Manchester, Manchester, United Kingdom

The relative contributions of excitation and inhibition to fMRI responses remain unknown. In principle, inhibition may increase or decrease fMRI signals depending on local circuitry. Negative BOLD signals and CBF decreases were shown for ring stimuli in primary visual cortex (V1). High-resolution fMRI can exploit the functional segregation in V1 to reveal differences between excitatory and inhibitory responses, including layer-specific differences. We measured high-resolution BOLD, CBV and CBF in macaque V1 and found laminar differences in the positive and negative fCBF responses, suggesting different neurovascular coupling mechanisms depending on the location within the cortical sheet.

Robert Friedman¹, Feng Wang¹, Chaohui Tang¹, Anna Roe¹, and Malcolm Avison^1
1Vanderbilt University, Nashville, TN, United States

We used high resolution CBV fMRI to examine the role of inter- and intra-areal processing of spatiotemporally varying tactile inputs in somatosensory motion perception. Simultaneous stimulation of three adjacent digits elicited weaker, more focal activation in area 3b of SI somatosensory cortex than would be predicted from single digit responses. Sequential tapping elicited greater than expected activation in areas 3b and 1. In areas 3b and 1, single digit and simultaneous multiple digit stimulation activated mid-cortical layers, whereas sequential tapping of the digits evoked significant activation in superficial and deep laminae.

Wietske van der Zwaag^1,2, Roberto Martuzzi³, Remy Kusters¹, Olaf Blanke³, Rolf Gruetter^1,2, and José P. Marques^1,2
1Université de Lausanne, Lausanne, VD, Switzerland, ²LIFMET, EPFL, Lausanne, VD, Switzerland, ³LNCO, EPFL, Lausanne, VD, Switzerland

Using high-resolution 7 tesla fMRI, the somatotopic representations of all five digits and the hand and forearm were mapped in the cerebellum of individual subjects. The representations of the digits followed an orderly pattern in lobule 5, with the thumb most posterior and medial and the little finger most lateral and anterior. In lobule 8, an orderly pattern was also found, but this was not consistent across subjects. The hand region fully overlapped with the digits regions in a manner similar to what has been suggested for the primary motor cortex.

Molly Gallogly Bright^1,2, Marta Bianciardi^2,3, Jacco A. de Zwart², and Jeff H. Duyn^2
1CUBRIC, School of Psychology, Cardiff University, Cardiff, United Kingdom, ²Advanced MRI Section, LFMI, NINDS, National Institutes of Health, Bethesda, MD, United States, ³A.A. Martinos Center for Biomedical Imaging, Department of Radiology, Massachusetts General Hospital, Charlestown, MA, United States

Negative fMRI signal changes have been observed in large cerebral vessels, and modeling suggests this phenomenon originates from changes in blood volume rather than deoxyhemoglobin concentration. To better understand the underlying mechanisms, respiratory challenges were used to create transient hypocapnia and global decreases in blood volume in 10 subjects during BOLD-weighted EPI acquisition at 7 Tesla. The global timeseries was correlated with every voxel, and the time-to-peak of the voxelwise response was extracted. Large vessels exhibited anticorrelated signal changes of physiological origin that occurred significantly earlier than the response across gray matter, suggesting early blood volume changes are responsible.

Kevin W.-K. Tsai¹, Thomas Witzel², Tommi Raij², Jonathan Polimeni², Jyrki Ahveninen², Wen-Jui Kuo³, John Belliveau², and Fa-Hsuan Lin^1
1National Taiwan University, Taipei, Taiwan, Taiwan, ²Massachusetts General Hospital, United States, ³National Yang Ming University, Taiwan

Using magnetic resonance inverse imaging (InI) with 100-ms temporal resolution, whole-brain coverage, and uniform spatial resolution achieved by multiple projections, we recorded BOLD-contrast fMRI responses to visual hemifield stimuli. The lateral geniculate nucleus (LGN) was activated approximately 500 ms before the primary visual cortex (V1). The fact that the LGN->V1 delay is longer in BOLD recordings than in evoked electrophysiological responses may reflect interregional vasculature or neurovascular coupling differences, or relative onsets of evoked/induced neuronal activity.

Stephen Smith¹, Karla Miller¹, Steen Moeller², Junqian Xu², Edward J Auerbach², Mark W Woolrich³, Christian F Beckmann^4,5, Mark Jenkinson¹, Jesper Andersson¹, Matthew F Glasser⁶, David Van Essen⁶, David Feinberg^7,8, Essa Yacoub², and Kamil Ugurbil^2
1FMRIB, Oxford University, Oxford, Oxfordshire, United Kingdom, ²Center for Magnetic Resonance Research, University of Minnesota, ³Oxford Centre for Human Brain Activity, Oxford University, ⁴Donders Institute for Brain, Cognition and Behaviour, Radboud University Nijmegen, ⁵MIRA Institute for Biomedical Technology and Technical Medicine, University of Twente, ⁶Anatomy and Neurobiology, Washington University School of Medicine, ⁷Advanced MRI Technologies, ⁸University of California, Berkeley

Current correlation-based approaches for resting-state networks measure average functional connectivity between regions over time, but this is not very meaningful if regions are part of multiple networks. One wants a network model that allows overlap, allowing a region’s activity level to reflect one network’s activity at some points in time and another network’s activity at others. However, even approaches that do allow overlap have often maximised spatial independence, which may be suboptimal if networks have significant overlap. Here we identify functionally distinct networks by virtue of temporal independence, taking advantage of additional temporal richness via improvements in FMRI sampling rate.

Karen J Mullinger¹, Stephen D Mayhew², Andrew P Bagshaw², Richard W Bowtell¹, and Susan T Francis^1
1SPMMRC, School of Physics and Astronomy, University of Nottingham, Nottingham, Nottinghamshire, United Kingdom, ²BUIC, School of Psychology, University of Birmingham, Birmingham, United Kingdom

The exact nature of the vascular and neuronal contributions to the BOLD post-stimulus undershoot remains unclear. Here we use simultaneous EEG-BOLD-ASL recordings during median nerve stimulation to investigate the origin of the post-stimulus undershoot. We demonstrate for the first time that the undershoot amplitudes of both BOLD and CBF measurements in contralateral (positive response) and ipsilateral (negative response) sensory-motor cortex are negatively correlated with natural fluctuations in the power of 8-13Hz, post-stimulus, mu oscillations measured from concurrent EEG recordings. This study provides new evidence for a neural component underlying the BOLD undershoot.

Javier Gonzalez-Castillo¹, Ziad S Saad², Handwerker Daniel¹, and Peter Bandettini^1
1Section on Functional Imaging Methods, NIMH, NIH, Bethesda, MD, United States, ²Scientific and Statistical Computing Core, NIMH, NIH, Bethesda, MD, United States

Over the last 20 years fMRI have emphasized a localizationist view of brain function showing only a handful of regions responding to task/stimulation. Here, using 9 hours of functional data/subject, we challenge that view with evidence that, at high TSNR, fMRI activations extent beyond areas of primary relationship to the task; and that task-correlated signal changes appear in over 90% of the brain for a visual stimulation + attention-control task. Moreover, we show that responses vary greatly across regions; and that whole-brain parcellations based on response differences produce functionally meaningful clusters that are symmetrical across hemispheres and reproducible across subjects.