Rapid fluctuations in multi-scale correlations of cortical networks encode spontaneous behavior

Experimental work across a variety of species has demonstrated that spontaneously generated behaviors are robustly correlated to variation in neural activity within the cerebral cortex. Indeed, functional magnetic resonance imaging (fMRI) data suggest that functional connectivity in cortical networks varies across distinct behavioral states, providing for the dynamic reorganization of patterned activity. However, these studies generally lack the temporal resolution to establish links between cortical signals and the continuously varying fluctuations in spontaneous behavior typically observed in awake animals. Here, we take advantage of recent developments in wide-field mesoscopic calcium imaging to monitor neural activity across the neocortex and simultaneously record cellular resolution activity using 2-photon microscopy in awake mice. Most analyses of state-dependent connectivity in neuronal networks rely on binning data within identified epochs ignoring the potential for rapid, continuous variation in both behavioral state and network configuration. As neurons can be exquisitely sensitive to patterned or synchronized input, it seems reasonable to hypothesize that rapid changes in network correlations are closely linked to the integrative function of single cells. We present here an analytic formulation for modeling dynamics of behavior that goes beyond linear models while preserving interpretability. Firstly, we show that our proposed model for behavior is indeed superior to the commonly used linear models in wide-cortical scales as well as in cellular data. Secondly, we extract a low-dimensional representation for cortical-cellular correlations capturing their latent dynamics and show that behavior is also encoded within multimodal correlations. Lastly, we show that the latent dynamics of the extracted cross-scale correlations reveals subnetworks that are not evident in traditional anatomical atlas-based parcellation of the cortex. These results provide insight into how behavioral information is represented between the mammalian neocortex and cellular networks and demonstrate a new analytical framework for investigating time-varying cross-modality connectivity in neural networks.