• BOLD
• connectivity
• fMRI
• motion
• visual pathway

Visual motion processing is one of the main subsystems of the visual system of primates, as the ability to perceive and detect motion is essential for survival in a complex environment. Speed and direction of motion in a complex visual scene are processed by the human middle temporal (MT) complex (hMT+), a region of the extrastriate cortex that plays a central role in visual motion perception. It is well known that the main pathway that conveys visual motion information from the retina to MT involves two main processing stages, one at the level of the thalamus, in the Lateral Geniculate Nucleus (LGN), and the other one in the primary visual cortex (V1) . However, anatomical studies in primates have suggested the existence of at least two other pathways that convey visual motion information from the thalamus straight to area MT, that is, without passing through V1. While all these findings may suggest a direct link from the thalamus to hMT+, a direct functional influence exerted by thalamic nuclei on hMT+ remained to be proven.
We conducted functional fMRI experiments to record brain activity in response to moving visual stimuli. Performing CGC analysis on the BOLD fMRI time-series of LGN, hMT+ and V1, a significant direct influence of the BOLD signal recorded in LGN over that in hMT+, not mediated by V1 activity, was demonstrated, suggesting the co-existence of an alternative route that directly links LGN to hMT+. Moreover, it has been observed that hMT+ response latencies and amplitude vary as a function of speed, raising the question whether these characteristics reflect a property of the functional connection with LGN. Thus, we performed a new experiment to measure the causal influence between LGN and hMT+, in the presence of fast and slow visual stimuli in order to better understand the influence of speed on this direct connection. Computing differences between the strength of the causal connection from LGN to hMT+ in the two speed conditions, we detected two clusters of voxels in hMT+ whose connections with LGN are stronger in either the slow or the fast condition.
In order to move our evidence of a direct connection between LGN and hMT+ based on the dynamics of BOLD signals towards a neuronal interpretation, more direct measurements of neural activity are needed. The neurovascular and neurometabolic coupling between blood flow supply and neural activity is still debated since the BOLD is an indirect effect of a combination of changes in Cerebral Blood Volume (CBV), Cerebral Blood Flow (CBF), and oxygen consumption (CMRO2) . Using dynamic responses of BOLD signal using Gradient Echo acquisition techniques, CBV (using superparamagnetic iron oxide), CBF (by arterial spin labeling and laser-Dopper flowmetry) and Temperature (using copper-constantan thermocouple wire), we characterized the coupling between these parameters across three transcortical segments along the vertical axis of the rat's primary somatosensory cortex during forepaw stimulation. Since BOLD, temperature and CMR02 modeling dynamics vary their relationship across each segment, layer-specific changes of these neurophysiologic, hemodynamic, and metabolic measurements are needed to better interpret high-resolution functional magnetic resonance imaging (fMRI) data. Combination of multi modal neuroimaging techniques and advanced analysis will substantially contribute to quantitatively relate the BOLD signal to changes in neurotransmission and electrical activity and will help to elucidate the neurophysiological processes of the human brain.