Excited to share our new preprint on how network dynamics and structured connectivity jointly define the spatial and temporal profiles of neural correlations. Work w/ @roxana_zeraati, @SelfOrgAnna, @EngelTatiana: arxiv.org/abs/2207.07930 Here is a #tweeprint: 
1/8 Correlated fluctuations in the activity of neural populations occur across multiple temporal and spatial scales, which relate to computations in many cognitive tasks.
2/8 While temporal and spatial correlations are usually studied separately, they emerge from the same spatiotemporal dynamics. So how are they related to each other?
3/8 To answer this question, we derived analytical expressions for spatiotemporal correlations in networks of binary units with connectivity arranged in one and two dimensions.
4/8 We showed that multiple timescales in auto- and cross-correlations arise from interactions among network units defined by spatial connectivity.
5/8 Specifically, the autocorrelation contains an intrinsic timescale and a set of interaction timescales generated by spatial interactions among units. The interaction timescales are shared between auto- and cross-correlations.
6/8 Using Fourier transformation, we showed that each interaction timescale arises from fluctuations at a different spatial frequency mode, contributing hierarchically to the overall patterns of correlations.
7/8 Both spatial and temporal scales of correlations depend on the spatial connectivity range because interaction timescales decrease heterogeneously with broader spatial connectivity.
8/8 Finally, external inputs can modulate the correlation timescales when spatial interactions are nonlinear, and the modulation effect depends on the operating regime of network dynamics.
• • •
Missing some Tweet in this thread? You can try to
force a refresh
