Author response image 3. “line search” approach vs. TDLM on the simulated spiking data (assuming two ground-truth sequences).

(a) The rate map of the simulated place cells (n=40) over a linearized space with 80 positions, smoothed with 2 sample gaussian kernel, to mimic overlapping place fields. (b) We simulated two ground truth sequences with time lag of 2 time samples between successive firings in the ripple event. We also show the decoded position space right next to the ripple event. All replay analyses are performed on this decoded space. The histogram is the sequence distribution of the shuffled data (in blue). The red line is the sequence results in the true data. (c) We randomly shuffled the firing order in the ripple event, so that there are no structured sequences in the simulation. We show there is no temporal structure in the decoded position space. As a result, TDLM reports no significant sequenceness value on the decoded position space.