Skip to main content
. Author manuscript; available in PMC: 2020 Mar 1.
Published in final edited form as: Hippocampus. 2018 Nov 13;29(3):260–274. doi: 10.1002/hipo.22994

Figure 6. Our model produces scale-invariant time cells that resemble neural data.

Figure 6.

a. The output layer consists of 5 leaky integrate and fire neurons which we identify as time cells. b. From top to bottom: output layer firing rates for the five time cells generated by this network. Rescaled version of the firing rates for the five neurons. It is clear that the firing rates coincide with each other when rescaled, showing that the firing rates for the time cells are indeed scale-invariant. Firing rates are averaged over 100 trials. c. Heatmap generated from 100 simulated time cells. Compare to Figures 1, 2c, and 3c.