Uncovering temporal structure in hippocampal output patterns

The place cell activity of hippocampal pyramidal cells has been described as the cognitive map substrate of spatial memory. Replay is observed during hippocampal sharp-wave ripple-associated population burst events and is critical for consolidation and recall-guided behaviors. To present, population burst event (PBE) activity has been analyzed as a phenomenon subordinate to the place code. Here, we use hidden Markov models to study PBEs observed during exploration of both linear mazes and open fields. We demonstrate that estimated models are consistent with temporal replay sequences and that the latent states correspond to a spatial map of the environment. Moreover, we demonstrate the identification of hippocampal replay without recourse to the place code, using only PBE model congruence. These results suggest that downstream regions may rely on PBEs to form a substrate for memory. Additionally, by forming models independent of animal behavior, we lay the groundwork for studies of non-spatial memory. Introduction Large populations of neurons fire in tandem during hippocampal sharp-waves and their accompanying CA1 layer ripple oscillations (Buzsáki, 1986). By now, multiple studies have shown that during many sharp-wave ripple-associated population burst events (PBEs), hippocampal “place cells” (O’Keefe, 1976) fire in compressed sequences that reflect the firing order determined by the sequential locations of their individual place fields (Diba and Buzsáki, 2007; Foster and Wilson, 2006; Lee and Wilson, 2002; Nádasdy et al., 1999). While the firing patterns during active exploration are considered to represent the brain’s global positioning system and provide a substrate for spatial and episodic memory, instead it is the synchronized activity during PBEs that is most likely to affect cortical activity beyond the hippocampus (Buzsáki, 1989; Carr et al., 2011; Diekelmann and Born, 2010; Siapas and Wilson, 1998). Likewise, widespread activity modulation is seen throughout the brain following these sharp-wave ripple population bursts (Logothetis et al., 2012). The literature on PBEs has largely focused on developing templates of firing patterns during active behavior and evaluating the extent to which these templates’ patterns are reprised during subsequent PBEs. But what if the fundamental mode of the hippocampus is not the re-expression of place fields, but rather the PBE sequences during sharp-wave ripples (SWRs)? PBE sequences are enhanced during exploration of novel environments (Cheng and Frank, 2008; Foster and Wilson, 2006), they presage learning-related changes in place fields (Dupret et al., 2010), and appear to be critical to task learning (Ego-Stengel andWilson, 2010; Girardeau et al., 2009; Jadhav et al., 2012). Here, we examine the information provided by CA1 and CA3 pyramidal neurons, the output nodes of the hippocampus, through the looking glass of PBE firing patterns. We developed a technique to build models of PBE sequences strictly outside of active exploration and independent of place fields and demonstrate that this nevertheless allows us to uncover spatial maps. Furthermore, these models can be used to detect congruent events that are consistent with replay but without any explicit place cell template. Our technique therefore provides new possibilities for evaluating hippocampal output patterns in single-trial and other fast learning paradigms, where a reliable sequential template pattern is not readily available. Overall, our work suggests that a 1 of 17 . CC-BY-NC-ND 4.0 International license peer-reviewed) is the author/funder. It is made available under a The copyright holder for this preprint (which was not . http://dx.doi.org/10.1101/242594 doi: bioRxiv preprint first posted online Jan. 4, 2018;