fMRI of cortical layers

Abstract

Functional magnetic resonance imaging (fMRI) is a tool to look at activity in the human brain in vivo and has developed into one of the most prominent techniques used in neuroscience today. A typical investigation involves creating MR images of the brain in rapid succession, the whole brain being depicted every 2-3 seconds. This high frame-rate comes at the cost of a relatively coarse spatial resolution, the smallest discernable image element being limited to approximately 3 mm. As mentioned, fMRI is used to depict brain activity, a process that takes place in the cortex, the thin outer layer of our brain. The cortex itself consists of six histologically defined layers, each having a different function. Imaging these layers individually could prove very useful in various fields of neuroscience. Brain network analyses in particular may benefit as input, output and feedback signals are known to be located in different layers. Unfortunately, standard fMRI cannot depict these layers: the cortex in its entirety is 3 mm thick, i.e. equally thick as the smallest details that standard fMRI methods can discern. In our research we used recent MRI developments to improve the spatial resolution of the fMRI experiment aiming to measure layer-specific activity signals. We started with proof-of-principle studies with a very low frame-rate where we showed that there is no fundamental fMRI limitation hindering laminar research. Once the analysis framework was in place, we restored the frame-rate to 2.5 seconds while maintaining laminar resolution, showing that laminar fMRI is possible after all.