The application of functional magnetic resonance imaging (fMRI) to study ac
tivation of auditory cortex suffers from one significant confounding factor
, namely, that of the acoustic noise generated by the gradient system, whic
h is an integral part of the imaging process, Earlier work has shown that i
t is indeed possible to distinguish cortical activation resulting from pres
entation of auditory stimuli despite the presence of background noise from
the gradient system, The influence of acoustic noise from the gradient syst
em of the MRI scanner on the blood oxygen level-dependent (BOLD) response d
uring functional activation of the auditory cortex has been investigated in
six healthy subjects with no hearing difficulties. Experiments were perfor
med using gradient-echo echoplanar imaging (EPI) and a verbal, auditory dis
crimination paradigm, presented in a block-wise manner, in which carefully
aligned consonant-vowel syllables were presented at a rate of 1 Hz, For eac
h volunteer the experiment was repeated three times with all parameters fix
ed, except slice number, which was 4, 16, or 64, The positioning of the cen
tral four slices in each experiment was common, Thus, the fraction of TR du
ring which the stimulus is on but no imaging is being performed, varies fro
m almost zero, in the case of 64 slices, to over 8 seconds, in the case of
four slices. Only the central four slices were of interest; additional slic
es simply generated acoustic noise and were discarded, During the four-slic
e experiment, all subjects showed a robust BOLD response in the superior te
mporal gyrus covering the primary and secondary auditory cortex. The spatia
l extent and the z-scores of the activated regions decreased with longer du
ration of gradient noise from the scanner. For a phonetic discrimination ta
sk, the results indicate that presentation of the stimulus during periods f
ree from scanner noise leads to a more pronounced BOLD response, J, Magn, R
eson, imaging 1999;9:19-25, (C) 1999 Wiley-Liss, Inc.