Hemodynamic response timing of human lateral geniculate nucleus and visual cortex

Abstract

INTRODUCTION Accurate timing information is important for understanding the flow of neuronal activity in the brain. While BOLD-contrast fMRI is widely used to localize brain activations, hemodynamic responses secondary to neuronal events have been considered to have insufficient temporal resolution to detect interregional spread of neuronal activity. Magneto-and electroencephalography (MEG/EEG) have excellent temporal resolution and are directly measuring post-synaptic neuronal activity [1]. However, the sensitivity of MEG/EEG is mainly limited to cortex and accurate localization of activated areas is confounded by the ill-posed electromagnetic inverse problem. This biophysical dilemma has impeded the understanding of relative timing between subcortical and cortical activations in humans. Recently developed magnetic resonance inverse imaging (InI) that uses highly parallel RF detection has allowed sampling of the BOLD signal at a 100 ms temporal sampling [2, 3]. InI has a whole brain coverage and almost uniform spatial sensitivity throughout the brain. Combining multiple projection acquisitions, InI can even maintain the whole-brain sensitivity and 100 ms temporal sampling without compromising any spatial resolution [4]. Thus it is feasible to use InI to probe cortical/subcortical timing differences at neuronally relevant scales. Here we report timing differences between lateral geniculate nucleus (LGN) and visual cortex (V1) BOLD responses to visual hemifield checkerboard stimuli. These preliminary results indicate that the activity of LGN significantly precedes that of V1 by approximately 500 ms. METHOD Twelve subjects were recruited with informed consent. Visual stimuli consisted of a radial checkerboard (contrast reversal at 8 Hz) pattern presented in the left or right visual hemifield; stimulus duration was 500 ms, and the latency between the two hemifields was varied parametrically. Subjects were instructed to push a button as quickly as possible for every visual hemifield stimulus using the ipsilateral hand. InI was used to measure the BOLD signal using a 3T MRI scanner (Tim Trio, Siemens Medical Solutions, Erlangen, Germany) and a 32-channel head coil array. The imaging parameters were: TR=100 ms; TE=30 ms, flip angle=30