Supplementary Figure 10: Possible enrichment mechanisms.

Comparison of three theoretical methods by which place cells could enrich a goal location. a. Place fields could be generally stable, but place cells near the reward are more likely to reoccur as place cells from session-to-session. (top) Recurrence probability as a function of distance from reward. (middle) Mean place field centroid shift as a function of distance from reward. (bottom) Mean place field shift variance as a function of distance from reward. b. Place cells could reoccur at equal probability along the belt, but place fields shift towards the reward location such that fields before the reward shift forward and fields after the reward shift backwards. Plots as in a. Place fields shifting towards the reward also leads to enrichment in our model. c. Place fields might not shift uniformly towards the reward, but if fields are generally stable, the ones near the reward could shift less than ones farther away. Plots as in a. d-f. Using the parameters from a-c, our enrichment model suggests all three hypothetical models could lead to enrichments: (d) increased place cell recurrence at reward position, (e) place fields shifting towards the reward location, or (f) place fields near the reward shifting less than ones away from the reward.