Long term brain dynamics extend cognitive neuroscience to timescales relevant for health and physiology

In medicine, we define and treat diseases based on their causes. We classify infections as viral, fungal, parasitic, or bacterial with specialized treatments for each class. We define and treat tumors according to which genetic abnormalities allow them to proliferate uncontrollably. This tenet underlies almost every field of medicine except for the brain where many diseases are largely defined by their symptoms. As a result, with a few notable exceptions, we define treatments of the brain around which symptoms to use them for rather than what root cause they solve. But what if we could peer inside somebody’s brain, identify the pathological circuits and activity that drives a patient’s disease, and then gear our treatment towards that? While past attempts at this have shown initial promise, they have been limited by small sample sizes and difficulty in selecting appropriate study populations. In this thesis, I explore how both these problems can be addressed by a paradigm shift to study neural activity over very long timescales.

In medicine, we define and treat diseases based on their causes. We classify infections as viral, fungal, parasitic, or bacterial with specialized treatments for each class. We define and treat tumors according to which genetic abnormalities allow them to proliferate uncontrollably. This tenet underlies almost every field of medicine except for the brain where many diseases are largely defined by their symptoms. As a result, with a few notable exceptions, we define treatments of the brain around which symptoms to use them for rather than what root cause they solve. But what if we could peer inside somebody’s brain, identify the pathological circuits and activity that drives a patient’s disease, and then gear our treatment towards that? While past attempts at this have shown initial promise, they have been limited by small sample sizes and difficulty in selecting appropriate study populations. In this thesis, I explore how both these problems can be addressed by a paradigm shift to study neural activity over very long timescales.