PURPOSE: To evaluate the feasibility of functional MR imaging (fMRI) with a 1.0 T scanner, fMRI of normal cerebral motor cortex at 1.0 T was compared with that at 1.5 T. MATERIALS AND METHODS: FMRI of bilateral cerebral motor cortices (left, seven; right, six) was performed in seven healthy male volunteers aged 26-34 (mean 29) years, with BOLD contrast at both 1.0 T and 1.5 T units (Siemens MR scannners). Using both these systems, two-dimensional (2D) FLASH images were obtained with TR/TE of 90/56, flip angle of 40°, matrix size 128*128, slice thickness of 5 mm, and FOV 23 cm. A sequence consisting of five-image-off phase (rest phase) followed by five-image-on phase (activation with finger movement) was repeated four times without pause at a single plane. The same study was performed for the contralateral motor cortex in each volunteer. Using the z-test, activation images were obtained for the signal difference between on- and off-phases (p < 0.05) and were then superimposed on 2D FLASH anatomic images at the same plane. Percentage changes of signal intensities (PCSIs) and numbers of activated pixels were compared, using the non-parametric t-test, and periodicity of signal changes was compared, using the Mantel-Haenszel Chi-square test. RESULTS: Mean PCSIs at 1.5 T and 1.0 T in the left motor cortex were 3.13 ±1.20% and 1.43 ± 0.56%, respectively (p = 0.009), and in the right, 1.78 ± 0.95% and 1.34 ± 0.28%, respectively (p = 0.32). The mean number of activated pixels at 1.5 T and 1.0 T in the left cortex was 21.14 ± 10.67 and 19.86 ± 11.36, respectively (p = 0.83), and in the right, 22.5 ± 6.47 and 16.8 ± 8.47, respectively (p = 0.22). At 1.5 T, periodicity of signal changes was seen in the left cortex in six of seven volunteers, and in the right cortex, in four of six. At 1.0T, all showed periodicity (left : p = 0.32 ; right : p = 0.14). CONCLUSION: PCSIs in the dominant hemispheres were significantly higher at 1.5 T, but no other indicators showed significant differences between 1.0 T and 1.5 T. Acceptable fMRI can therefore be carried out with a 1.0 T field strength scanner.