Voxel based morphometry showing acute alteration in a visual paradigm

Abstract

Aims: Previous studies used voxel based morphometry (VBM) for imaging long term alteration within the gray matter of the brain. Increased and decreased densities were reported after special conditions such as learning to juggle or pain application and were interpreted as signs of cerebral neuroplasticity over time. Acute changes of cerebral activation have been only studied by functional imaging such as functional magnetic resonance imaging (fMRI) and positron emission tomography (PET). Objectives: To investigate whether VBM can also detect acute changes induced by a simple checkerboard paradigm. Methods: Twenty healthy subjects were investigated using VBM. Three MRIs were applied to each single subject: one before looking at a flickering checkerboard, one after looking at a flickering checkerboard, and one 1 hour later. For each subject a map of grey matter volumetric differences between the scans of different time points was calculated. As control experiment we performed regular fMRI with checkerboard stimulation in all participants. Results: Changes within the occipital lobe were detected, most pronounced during right after looking at the flickering checkerboard. Changes almost entirely disappeared within one hour. The same region showed activation in a regular fMRI block design using the checkerboard. Discussion: The observed changes appear to be generated by looking at the checkerboard as they occur in the occipital lobe which is known to be activated in visual paradigms. Up to now, VBM changes were interpreted as either neuronal growing/degeneration, or dendrite spine and synapse turnover, or changes of the extracellular space/ microvascular volume. Acute VBM changes rather seem to reflect actions that must occur much faster than the hithero suggested, for instance water displacement. This phenomenon might explain the wide diversity of results by different VBM studies not controlling for acute changes or using cross sectional design.