Mental imagery is an important cognitive method for problem solving, and the mental rotation of complex objects, as originally described by Shepard and Metzler (1971), is among the best studied mental imagery tasks. Functional MRI was used to observe focal changes in blood flow in the brains of 10 healthy volunteers performing a mental rotation task. On each trial, subjects viewed a pair of perspective drawings of three-dimensional shapes, mentally rotated one into congruence with the other, and then determined whether the two forms were identical or mirror-images. The control task, which we have called the 'comparison' condition, was identical except that both members of each pair appeared at the same orientation, and hence the same encoding, comparison and decision processes were used but mental rotation was not required. These tasks were interleaved with a baseline 'fixation' condition, in which the subjects viewed a crosshair. Technically adequate studies were obtained in eight of the 10 subjects. Areas of increased signal were identified according to sulcal landmarks and are described in terms of the Brodmann's area (BA) definitions that correspond according to the atlas of Talaraich and Tournoux. When the rotation task was contrasted with the comparison condition, all subjects showed consistent foci of activation in BAs 7a and 7b (sometimes spreading to BA 40): 88% had increased signal in middle frontal gyrus (BA 8) and 75% showed extrastriate activation, including particularly BAs 39 and 19, in a position consistent with area V5/human MT as localized by functional and histological assays. In more than half of the subjects, hand somatosensory cortex (3-1-2) was engaged, and in 50% of subjects there was elevated signal in BA 18. In frontal cortex, activation was above threshold in half the subjects in BAs 9 and/or 46 (dorsolateral prefrontal cortex). Some (four out of eight) subjects also showed signal increases in BAs 44 and/or 46. Premotor cortex (BA 6) was active in half of the subjects during the rotation task. There was little evidence for lateralization of the cortical activity or of engagement of motor cortex. These data are consistent with the hypothesis that mental rotation engages cortical areas involved in tracking moving objects and encoding spatial relations, as well as the more general understanding that mental imagery engages the same, or similar, neural imagery as direct perception.