Hippocampal Representations of Event Structure and Temporal Context during Episodic Temporal Order Memory

Abstract

The hippocampus plays an important role in representing spatial locations and sequences and in transforming representations. How these representational structures and operations support memory for the temporal order of random items is still poorly understood. We addressed this question by leveraging the method of loci, a powerful mnemonic strategy for temporal order memory that particularly recruits hippocampus-dependent computations of spatial locations and associations. Applying representational similarity analysis to functional magnetic resonance imaging activation patterns revealed that hippocampal subfields contained representations of multiple features of sequence structure, including spatial locations, location distance, and sequence boundaries, as well as episodic-like temporal context. Critically, the hippocampal CA1 exhibited spatial transformation of representational patterns, showing lower pattern similarity for items in same locations than closely matched different locations during retrieval, whereas the CA23DG exhibited sequential transformation of representational patterns, showing lower pattern similarity for items in near locations than in far locations during encoding. These transformations enabled the encoding of multiple items in the same location and disambiguation of adjacent items. Our results suggest that the hippocampus can flexibly reconfigure multiplexed event structure representations to support accurate temporal order memory.

Event integration and temporal differentiation: how hierarchical knowledge emerges in hippocampal subfields through learning

“…or close in narrated, but not actual, time (Bellmund et al, 2022; but see Deuker et al, 2016). Additional studies have reported decreased similarity in CA3/DG between objects that share spatiotemporal context (Copara et al, 2014; Dimsdale-Zucker et al, 2018; Kyle, Smuda, et al, 2015; Liu et al, 2022), and specifically in DG (Berron et al, 2016; see also Baker et al, 2016). Our results extend this prior work in two critical ways: first, we show that DG pattern separation can occur in the temporal domain, thus advancing our knowledge of how temporally extended experiences are represented in the brain (Davachi & DuBrow, 2015; Eichenbaum, 2014).…”

“…Putatively higher similarity at the boundary could reflect prediction of the coming event (Brown et al, 2016; Hindy et al, 2016; Kok & Turk-Browne, 2018; Schapiro et al, 2012), accompanied by reactivation at the end of the event (Ben-yakov & Dudai, 2011; Cohn-Sheehy et al, 2021; Griffiths & Fuentemilla, 2020; Silva et al, 2019; Sols et al, 2017). One speculation is that there is a shift in hippocampal sequence representations through learning (Ben-yakov et al, 2014; Davachi & DuBrow, 2015), whereby contextual stability during initial learning promote similarity of event representations (DuBrow & Davachi, 2014; Hasselmo & Eichenbaum, 2005; Manns et al, 2007), and these similar representations later facilitate reactivation of learned event components through pattern completion (Grande et al, 2019; Horner et al, 2015; Liu et al, 2022).…”

“…The copyright holder for this preprint (which this version posted July 20, 2022. ; https://doi.org/10.1101/2022.07.18.500527 doi: bioRxiv preprint Buzsáki & Tingley, 2018;Davachi & DuBrow, 2015;Deuker et al, 2016;Eichenbaum, 2017;Heusser et al, 2016;Liu et al, 2022;O'Keefe & Dostrovsky, 1971;O'Keefe & Nadel, 1978;Pastalkova et al, 2008;Paz et al, 2010;Ranganath & Hsieh, 2016;Schendan et al, 2003;Thavabalasingam et al, 2019;Umbach et al, 2020;J. Zheng et al, 2022).…”

“…Computations supported by distinct hippocampal subfields might help to solve these challenges (Norman & O’Reilly, 2003; O’Reilly & McClelland, 1994; Treves & Rolls, 1994; Yassa & Stark, 2011). Across humans and animal models, the hippocampus has been implicated in the representation of context, time, and sequential events (Bellmund et al, 2020; Brown et al, 2014; Buzsáki & Tingley, 2018; Davachi & DuBrow, 2015; Deuker et al, 2016; Eichenbaum, 2017; Heusser et al, 2016; Liu et al, 2022; O’Keefe & Dostrovsky, 1971; O’Keefe & Nadel, 1978; Pastalkova et al, 2008; Paz et al, 2010; Ranganath & Hsieh, 2016; Schendan et al, 2003; Thavabalasingam et al, 2019; Umbach et al, 2020; J. Zheng et al, 2022).…”

“…At event boundaries, the changing perceptual input drives CA3 representations towards a different activity pattern representing the new context (Hasselmo & Eichenbaum, 2005; Kesner & Rolls, 2015). Thus, we propose that contextual stability within the same event together with a shift to a different activity pattern between events can be a mechanism by which attractor dynamics lead to event representations in CA3 (Clewett et al, 2019; Davachi & DuBrow, 2015; Horner et al, 2015; Horner & Burgess, 2014; Liu et al, 2022). Indeed, our prior work has shown that stability of hippocampal representations correlates with temporal memory for event sequences (DuBrow & Davachi, 2014; Ezzyat & Davachi, 2014).…”