The brain may use a shared code for gender across faces, bodies, and objects

A recent neuroimaging study reveals that the brain may utilize a unified code to interpret gender cues from faces, bodies, and objects. Researchers at the Institute of Psychology of the Chinese Academy of Sciences employed functional magnetic resonance imaging (fMRI) to monitor participants' brain activity while they observed images of men and women, as well as objects associated with each gender.

The study aimed to determine whether any brain region encodes gender irrespective of the visual input. The participants, 22 healthy adults aged 19 to 28, were shown 30 images each of male and female faces, bodies, and objects. The images were displayed in 12-second blocks, with three repetitions of each condition, and participants pressed a button whenever the same image appeared consecutively to ensure engagement.

The researchers used a machine-learning algorithm to determine if brain activity patterns could differentiate male images from female images. Gender information was detected in various brain regions depending on the type of visual input. For faces, gender cues were found in several visual areas on both sides of the brain, as well as in the right superior parietal lobule and the left fusiform gyrus.

Body images revealed gender-related activity in the occipitotemporal cortex and other regions on both sides of the brain. Objects associated with men or women showed gender-related activation in the cuneus, calcarine cortex, lingual gyrus, fusiform gyrus, and inferior temporal gyrus, among other areas on both sides of the brain.

The researchers then investigated whether gender information from one type of image could be transferred to another. The only region that successfully encoded gender across all three image types was a small cluster in the right middle temporal gyrus, with a second analysis supporting this finding. This region, however, was not exclusively dedicated to gender processing, as it also encoded information about the type of visual input (face, body, or object).

To understand the type of visual information this region used, the researchers compared its activity with an artificial neural network trained to classify images by gender. The activity in the right middle temporal gyrus closely matched the middle layers of the network, indicating that it relies on mid-level visual features such as contours, textures, shapes, and spatial arrangements.

The study also examined the communication between gender-related brain regions. The pattern of connections was similar when participants viewed faces and bodies but distinct when objects were viewed. The findings suggest a category-general neural hub for gender perception, highlighting how mid-level visual features and distributed networks enable the abstraction of gender from diverse visual inputs.