The Representation of Affective Internal States in the Hypothalamus: A Neural Manifold View

Our understanding of hypothalamic circuits that control social and other innate survival behaviors has exploded in recent years, thanks to the application of powerful genetically based tools for marking, mapping, manipulating, and monitoring specific neuronal subpopulations. Yet there remains an important set of unanswered questions that must be addressed to fully understand this complex system, including how the hypothalamus encodes affective internal states underlying these behaviors. In systems neuroscience, neural representations of cognitive variables have been revealed by analyzing the activity and dynamics of neural manifolds hidden within high-dimensional population neural activity. These signals are not detectable in the activity of individual neurons or neuron types. Because the hypothalamus has traditionally been viewed as the least "cognitive" part of the brain, the so-called "population doctrine" in cognitive neuroscience has not been widely applied to this system. Here, we discuss recent data that suggest that the application of this manifold-level approach may be useful for identifying representations and mechanisms of internal affective state coding in the hypothalamus. In turn, these findings argue that cognition and emotion may share computational mechanisms to a greater degree than previously assumed.