Authors: Greg S Harrington Sarah Tomaszewski Farias Michael H Buonocore Andrew P Yonelinas
Publish Date: 2006/05/09
Volume: 48, Issue: 7, Pages: 495-505
Abstract
The goal of the present study was to evaluate the inter and intrasubject reproducibility of FMRI activation for three memory encoding tasks previously used in the context of presurgical functional mapping The primary region of interest ROI was the medial temporal lobe MTL Comparative ROIs included the inferior frontal and fusiform gyri which are less affected by susceptibilityinduced signal losses than the MTL regions Eighteen subjects were scanned using three memory encoding paradigms wordpair pattern and scene encoding Nine subjects underwent repeat scanning Intersubject reproducibility of FMRI activation was evaluated by examining the percent of subjects who showed activation within a given ROI and the range to which individual laterality indices LIs varied from the mean Intrasubject testretest reproducibility was evaluated by examining the LI testretest correlation the average difference between LIs from two separate imaging sessions and concordance ratios of activation volumes Rvolume and Roverlap For scene encoding the reproducibility of activation volume and LIs within the MTL were as good as or better than the reproducibility within the fusiform and inferior frontal ROIs For pattern encoding and wordpair encoding the reproducibility of activation volume and LIs within the MTL tended to be worse compared to the fusiform and inferior frontal ROIs The differences in FMRI reproducibility appeared more dependent on the task than the susceptibility effects The results of this study suggest that FMRIbased assessment of the neural substrates of memory using a scene encoding task may be a useful clinical tool
Keywords: