Skip to main content
. Author manuscript; available in PMC: 2023 Oct 23.
Published in final edited form as: Cell Rep. 2023 Mar 24;42(4):112254. doi: 10.1016/j.celrep.2023.112254

Figure 2. Method to sample time series constrained by time-resolved correlation.

Figure 2.

(A) Flowchart and visualization of nullspace sampling. See the text for an informal description, and STAR Methods for mathematical detail.

(B) Properties of the windowed phase-randomization model, and the time-resolved correlation, variance, and mean model. Both models performed similarly well, but time-resolved correlation, variance, and mean constraints were fewer in number and easier to interpret. These constraints also generated data that were devoid of spatial correlations and in this way were considerably different from empirical recordings.