Object dimensions underlying food representations in visual cortex

The human brain is a master of categorization, constantly sorting the endless stream of visual information it receives into useful groups. When we look at the world, our eyes do not just see shapes and colors; they instantly recognize objects like chairs, tools, and food. Scientists have long known that specific regions in the back of the brain, tucked within the folds of the visual cortex, are dedicated to recognizing these categories. Recent discoveries have pinpointed two distinct areas within this visual processing zone that light up specifically when a person sees food. These regions, located on the side and the bottom of the visual processing area, are so specialized that they respond more strongly to images of meals than to almost anything else. Yet, a mystery remained: what exactly are these brain areas looking for? Do they recognize food because of its specific shape, its texture, or perhaps the way it looks in a group? Or do they respond to something deeper, like the way we might grab a piece of fruit or the specific colors that signal ripeness? Understanding this distinction is crucial because it reveals how the brain decides what is edible and how it prepares the body to interact with it.

To solve this puzzle, researchers conducted two detailed experiments using a machine that maps brain activity, asking a group of volunteers to look at various images while their brains were scanned. The team wanted to know if the brain's food-detecting areas were driven by the physical properties of the objects themselves, such as their color or how they are arranged in a pile, or by the actions we perform with them, like grasping or holding. They tested this by showing participants pictures of food alongside other objects that could be held, and by manipulating the images to see if the brain cared more about the object standing alone or the object as part of a colorful group. The results revealed a clear split in how the two food-sensitive areas operate. The region on the side of the visual cortex appeared to be focused on action. It responded strongly to food that could be grasped, treating a piece of fruit as sharing properties with other graspable objects. This area did not seem to care much about whether the image was in color or black and white, nor did it get distracted by the background; it simply recognized the object as something the hand could reach for.

In contrast, the region on the bottom of the visual cortex told a different story. This area was highly sensitive to the surface details of the food, particularly its color and how the items were arranged together. When the researchers showed images of food without a clear background or presented them as ensembles with no distinctive background, this bottom region became much more active. It also showed a strong preference for colored images over greyscale ones, suggesting that it relies on the visual cues of ripeness and variety to identify what is edible. To confirm that this split in function was not just a quirk of human biology but a result of how visual information is organized, the researchers built a computer model designed to mimic the layout of the brain's visual areas. When this artificial network was trained to process images, it naturally developed two separate groups of units that behaved exactly like the two brain regions: one group focused on the ability to grasp objects, and the other focused on the visual statistics of color and arrangement.

These findings suggest that the brain does not have a single, unified way of recognizing food. Instead, it uses two different strategies depending on which part of the visual system is engaged. One strategy is practical and action-oriented, identifying food based on how it can be manipulated by the hand. The other is visual and descriptive, identifying food based on its appearance and how it fits into a larger scene. By separating these functions, the brain can efficiently handle both the immediate need to grab a snack and the broader task of identifying what is safe and nutritious to eat. The study indicates that while the side of the brain helps us understand what we can hold, the bottom helps us understand what we see, and both are essential for the complex act of recognizing a meal.