Synchronized excitability in a network enables generation of internal neuronal sequences

Abstract

Hippocampal place field sequences are supported by sensory cues and network internal mechanisms. In contrast, sharp-wave (SPW) sequences, theta sequences, and episode field sequences are internally generated. The relationship of these sequences to memory is unclear. SPW sequences have been shown to support learning and have been assumed to also support episodic memory. Conversely, we demonstrate these SPW sequences were present in trained rats even after episodic memory was impaired and after other internal sequences – episode field and theta sequences – were eliminated. SPW sequences did not support memory despite continuing to ‘replay’ all task-related sequences – place- field and episode field sequences. Sequence replay occurred selectively during synchronous increases of population excitability -- SPWs. Similarly, theta sequences depended on the presence of repeated synchronized waves of excitability – theta oscillations. Thus, we suggest that either intermittent or rhythmic synchronized changes of excitability trigger sequential firing of neurons, which in turn supports learning and/or memory.DOI: http://dx.doi.org/10.7554/eLife.20697.001

Reactivation of time cell sequences in the hippocampus

“…This finding parallels previous reports of sequence reactivation in hippocampal neurons representing spatial trajectories, but for the temporal domain (Diba & Buzsáki, 2007;Foster & Wilson, 2006). Although a recent report has shown that the formation of place and time cell sequences differs (Wang, Roth, & Pastalkova, 2016), the reactivation of temporal and spatial sequence during sharp-wave ripples depends on similar mechanisms. In addition, our results suggest that the occurrence of reactivations containing only time cells supports the independent processing of temporal experiences in blocks of selective reactivation events (Gupta, van der Meer, Touretzky, & Redish, 2012).…”

“…S5). Similar to previous reports (Lee & Wilson, 2002;Wang, Roth, & Pastalkova, 2016), the reactivation of time cell sequences were temporally compressed, lasting around 200 ms (see Supporting Information Fig. S6).…”

“…The reactivation of place cells may occur in forward or reverse sequential order as during spatial navigation, but in a temporally compressed manner through short bursts of neuronal activity frequently associated to hippocampal sharp-wave ripples (100-250 Hz) (Buzsáki, 1998). Although sequence reactivation has been proposed as a general mechanism for memory consolidation, the reactivation of neuronal sequences coding the temporal order of an experience remains much less characterized (but see Wang, Roth, & Pastalkova, 2016).…”

Three brain states in the hippocampus and cortex

“…This suggests that the theta rhythm, a striking pattern of brain activity in its own right (Buzsaki, ; Buzsaki, ; Buzsáki, ; Green & Arduini, ; Hasselmo, ), might be a marker of the hippocampal state in which place firing occurs. Indeed, numerous results suggest that the theta rhythm has a functional role in the hippocampus (Ranck Jr, ; Winson, ; Buzsaki et al, ; O'Keefe & Recce, ; O'Keefe, ; Wang, Romani, Lustig, Leonardo, & Pastalkova, ; Redish, ; Wang, Roth, & Pastalkova, ; also see Brandon, Koenig, Leutgeb, & Leutgeb, ).…”

Cognitive swarming in complex environments with attractor dynamics and oscillatory computing

“…There are two reasons why neural swarming control must decouple each agent's physical location from its internal selflocalization. First, the multiplicity of agents is a qualitative difference with brain circuits; every hippocampal neuron in a biological network corresponds to the same single agent (e.g., a rat) and has particular inputs from internal processing or sensory inputs (Wang et al, 2016) that contribute to the appearance and location of the cell's place field. Given the analogy of agents to neurons ( Fig.…”