PURPOSE: Adequate image post-processing software, essential for functional magnetic resonance imaging (fMRI)of cerebral cortex activation, is not readily available; this study aimed to evaluate the usefulness of a homemade post-processing system for fMRI and to investigate the physiologic and anatomic sources of activation signals infMRI. MATERIALS AND METHODS: Twenty nine fMRIs of motor and visual cortices activated by hand movement and photic stimulation were performed using the GRE technique (TR/TE/flip angle : 80/60/40 °, 64x128 matrix) in 10 normal volunteers. By using post-processing software, final functional maps were subsequently obtained. In order to investigate physiologic and anatomic sources of activation signals, fMRIs of motor cortices were repeated with different echo times, flip angles and presaturation of adjacent sections. Activation signals were compared tocortical veins on MR venograms. RESULTS: All fMRIs were successfully performed and those activation signals were detected in regions well known as motor and visual cortices. Mean percentage changes of signal intensities (PCSIs)of activation signals at echo times of 15, 30, and 60 msec were 2.9%, 3.5%, and 12.5%, respectively (p < 0.01).Mean PCSIs of activation signals with 40 ° and 10 ° lip angles were 11.1% and 6.6% (p < 0.01), and those of activation signals without and with presaturation of the adjacent sections were 8.9% and 5.2% (p < 0.01). The shape and location of activation signals corresponded well with those of cortical veins. CONCLUSION: The imagepost-processing system developed in this study is a useful method for fMRI of cerebral cortex activation. Both BOLD (blood oxygen level-dependent) and inflow effects contribute to the physiologic source of activation signals, and the cortical veins draining activated cortex are the main anatomic source of activation signals.