Publications

Description: The most fundamental question in neuroscience is "how does the brain work?" Despite decades of research on individual neurons, behaviors, and molecular mechanisms, we lack what physicists call a unifying theory. Here, with Woody Shew, we present evidence that criticality is precisely that unifying principle. We revisit data from 140 papers and resolve a long-standing controversy. Finally, we present eight predictions that emerge when criticality is understood as an optimization principle that maximizes the brain's computational power. These predictions directly inform a new understanding of brain disease, development, homeostasis, and learning.

Description: What is a brain disease? While it may be caused by a protein, it's reasonable to suggest that the disease is only problematic because of failures in computation. Otherwise, the loss of neurons (e.g.), would be irrelevant. Thus, in this work, we reason that disease can only occur when homeostatic mechanisms either fail or are overwhelmed. Prior to that point, perturbations are trivially accounted for by compensatory plasticity --- axiomatically, that's the job of a homeostatic mechanism. James set out to record for the entire lifetime of animals destined to die of neurodegeneration, and track every known homeostatic set-point in neuronal activity. At some point along the way, at least one of these features must be derailed. To our great surprise, James found that, even at the end of life, individual neurons maintain set points in firing rates and other basic properties. In sharp contrast, network criticality was severely disrupted. This movement away from criticality began early in life, and progressively worsened until death. This suggests that criticality might be the dynamical locus of neurodegenerative diseases --- in other words, impaired brain function is not a result of killing neurons outright, but a disruption of the delicate network interactions that account for optimal information processing.

Description: The first major paper out of the Hengen Lab, this is a seminal work by Zhengyu Ma. Zhengyu demonstrated that cortical circuits are homeostatically maintained near criticality in freely behaving animals. Essentially, she tested the core prediction that criticality is the setting on the thermostat --- this is what all the set-points throughout the brain are ultimately trying to acheive. Using chronic, continuous recordings in rats, her study revealed that the brain actively tunes itself to a state that maximizes computational capacity, information transmission, and dynamic range. When Zhengyu pushed the brain away from criticality, it pushed back. Her work bridged theoretical physics and neuroscience by showing that the concept of criticality is not just a mathematical abstraction but a fundamental organizing principle that the brain actively maintains through homeostatic processes.

Description: This was a groundbreaking study --- the aim was to test core principles of homeostatic control where it counts: in the brain outside of experimental constraint. In the end, this manuscript was the first to demonstrate firing rate homeostasis in freely behaving animals, establishing the physiological relevance of homeostatic plasticity mechanisms that had previously only been observed in cultured neurons or anesthetized preparations. By chronically recording from ensembles of single neurons in visual cortex in freely behaving rats, we showed that the brain actively maintains stable activity levels even in the face of sustained challenge (e.g., chronic visual blockade). In retrospect, this paper laid a bit of foundation for the field of in vivo homeostatic plasticity.