Stimulus-specific influence of gender on mental-rotation-related brain activity

Abstract

Mental rotation (MR) is a cognitive process during which individuals mentally simulate the rotation in space of an object's image (stimulus). The traditional assertion that men outperform women in MR tasks may be influenced by methodological biases, such as treating gender as a secondary or post hoc variable, and relying solely on binary comparisons between two classes of MR stimuli. Furthermore, a comprehensive understanding of how nuanced the effects of the interaction between gender and stimulus type are on MR-related brain activity remains lacking. To fill these gaps, we recorded functional magnetic resonance imaging (fMRI) data while 57 participants (29 women, aged 18-35 years) performed MR of abstract objects, human bodies, and real objects. Whole-brain fMRI data analyses indicated that, with respect to women, men had larger activations in inferior frontal regions during MR of abstract objects, and in superior and medial frontal regions during MR of human bodies. Compared with men, in women we found larger activity in the superior parietal lobe during MR of human bodies with respect to abstract objects, and in the inferior occipital cortex in the MR of real objects versus human bodies. Finally, while in men we found a positive correlation between MR accuracy and brain activity in the precuneus, in women the correlation between MR accuracy and activity in motor and premotor areas was negative. These results indicate that brain activity during MR is modulated by the type of stimulus, differently for women and men.

Keywords: Bottom-up; FMRI; Gender differences; Mental rotation; Somatosensory processes; Stimuli.

Fig. 1

Experimental stimuli consisted in six abstract objects (top row), six human bodies (mid row), and six real objects (bottom row)

Fig. 2

Example of experimental trial. Participants were asked to mentally rotate the first stimulus in the direction shown by the arrows to the vertical axis of the screen. Then, participants indicated whether the second stimulus (always vertical) represented the same or different item with respect to the first stimulus. In the example reported in the figure, the correct response is “different.”

Fig. 3

(Left) Accuracy: Bonferroni-corrected t-tests revealed that participants were significantly less accurate in MR of abstract objects with respect to human bodies and real objects. (Right) Response time: Bonferroni-corrected t-tests of the significant main effect of stimulus revealed that MR of human bodies was significantly slower with respect to abstract and real objects. (***p < 0.001; **p < 0.01)

Fig. 4

The results of whole-brain analysis are projected onto 3D brains (left column) and axial slices (right column). (A) Gender-specific brain activity associated with MR of different stimuli. The activity clusters represent the brain regions where men had larger activity compared with women for the MR of abstract objects (blue) and human bodies (red). (B) Brain activity-related relationships between gender and comparisons between stimuli. The activity clusters represent the brain regions where women had larger activity compared to men in the contrast between MR of real objects and human bodies (green), and between MR of human bodies and abstract objects (yellow). (C) Correlation between accuracy and brain activity during MR of real objects. The activity clusters represent the regions where accuracy and brain activity correlated, separated for men (blue) and women (magenta). (D) Scatterplots illustrating the correlation between accuracy and brain activity related to MR of real objects in men (left) and women (right). Accuracy and brain activity correlated positively in men (precuneus; left panel). In women there was a negative correlation between accuracy and brain activity in the sensorimotor (pink) and premotor clusters (purple) (right panel). All statistical maps shown in A, B, and C were computed with a cluster-based threshold of z > 2.3 FWE-corrected at p = 0.05