Framework to identify food selectivity origins in the brain

"It's crazy how many different aspects of life food touches and how many different factors might contribute to creating any particular category of food," said Leila Wehbe, an associate professor in the School of Computer Science's Machine Learning Department and CMU's Neuroscience Institute. "If you think about it, food is even more primal than social cognition. If you don't have food, you die. It's necessary for survival in all species. Even the ones that don't have that much social interaction need food. And so, I like to think of it as the brain has evolved to either help you find food or keep you from being food."

Wehbe was part of a team that recently published a paper in Trends in Neurosciences that outlines how visual and nonvisual signals contribute to the brain's cortical responses when people were shown images of food. Team members include Michael Tarr, the Kav?i?-Moura Professor of Cognitive and Brain Science in the Dietrich College of Humanities and Social Sciences Department of Psychology and the Neuroscience Institute; and Maggie Henderson, an assistant professor in the Psychology Department and the Neuroscience Institute.

Tarr said there are regions of the brain that show some specialization for behaviorally critical domains, such as processing images of faces or tools. Until recently, researchers hadn't identified a similar region involved in processing images of food.

Recent studies using the Natural Scenes Dataset, however, have identified a food region in the brain. The Natural Scenes Dataset is a large-scale fMRI dataset comprising scans taken of people's brains while they were looking at images of objects, such as food, in their natural context or environment -- like dishes at a buffet, or a hot dog and soda at a picnic table. The dataset's growth has made research into the brain's food region possible.

"In the past, standard approaches tended to average across participants, which obscured more nuanced findings," Tarr said. "Large, naturalistic data allowed us to better study individual participants, which, in turn, revealed a pattern of food specialization present in every subject."

Researchers also found that images of food in context elicited additional responses in the brain, as compared to images without context. One explanation for this finding is that seeing these objects in their natural context might invoke the varied experiences of interacting with food or other information related to identifying food as just that: food. For example, signals that could influence food selectivity in the visual cortex include color, social cues or the motor actions of eating.

Henderson said while this work examined food selectivity in the brain's visual system, the scope isn't limited to food.

"This finding of food selectivity reveals general principles of high-level information processing in the brain," Henderson said. "How far can we get by thinking about the visual statistics of the environment, and how does that interact with other brain systems, like reward circuitry, social cognition and action recognition? We can think of food as one particular case study for how a range of visual and nonvisual factors might interact to give rise to how the brain is ultimately organized."