A New Framework for How the Brain Compresses Our Noisy World

In every moment of our lives, we are exposed to countless sensory signals. These include photons that strike our retinas, sound waves that hit our eardrums, volatile molecules that bind to receptors in our nostrils, and chemicals that glide over our taste buds. Our skin is also equipped with nerve endings that respond to pressure and heat.

Despite the overwhelming amount of data, we are able to navigate this sensory overload thanks to the brain's remarkable data compression capabilities. This compression process, known as categorization, transforms the noisy and information-rich world into objects, people, concepts, and emotions that we can comprehend and respond to at the level of our experiences.

Traditionally, neuroscience has viewed categorization as a passive process that occurs at the very end of sensory processing. Sensory details are received by the brain, their features are decoded, and they are then matched to templates stored in memory, like a clerk sorting through a neural filing cabinet. However, this approach struggles to explain the remarkable flexibility with which we assign labels to various features in the world.

For instance, a sudden rhythmic patter on a clear day might be a pigeon taking flight, while the same sound could be the shuffle of a stranger's footsteps when walking down a dimly lit alley at night.

Two leading neuroscientists, Lisa Feldman Barrett from Northeastern University and Earl Miller from the Massachusetts Institute of Technology, have collaborated to present a new perspective on categorization. In their paper published in Nature Reviews Neuroscience, they describe how the brain constantly reconstructs its categories in real-time, based not only on sensory input and memory, but also on the body's immediate physiological needs.

According to their framework, the brain projects categories onto the world in response to the body's survival needs, even before we become consciously aware of our sensory impressions.

This counterintuitive insight is central to Barrett and Miller's framework. The researchers argue that the brain's predictions, rather than the cumulative effects of sensory information, ultimately shape and limit how we categorize objects and features in the world. They term this concept as "the maintenance of the energetic constraints of life as fundamental to how we structure our categories."

Additionally, the duo's framework suggests that categorization doesn't occur in a specific area of the brain, but rather spans the entire organ and beyond. To understand how organisms form categories, they argue, neuroscientists must look all over the nervous system, from head to toe.

The collaboration between Barrett and Miller, who had previously worked independently, marks a significant development in the field of categorization research. Miller's expertise in predictive coding, which regards our perceptions as products of the brain's predictions, complements Barrett's work on emotions and the concept of allostasis.

Allostasis refers to how an organism predictively regulates its energy use, and Barrett has shown that emotion categories such as fear, happiness, and anger resemble predictive "action plans" that the nervous system generates to activate behaviors that have benefited us in the past.

In their new framework, Barrett and Miller view emotions as predictive action plans that activate behaviors to maintain the body's energy resources. For example, a heightened physiological state with increased heart rate, fast breathing, and tense muscles in the context of being chased by a dog would indicate the emotion category "fear" and trigger behaviors to fight or flee, thus conserving energy.

Traditionally, emotions were thought to be hardwired in specific circuits present from birth. However, Barrett's work has demonstrated that we construct emotional categories in relation to both signals from within the body and external context. When Barrett reached out to Miller with the idea of collaborating, he immediately understood the value of their combined perspectives. Together, they aim to move beyond the traditional filing-cabinet model of categorization to incorporate their ideas about prediction and allostasis.