Fig. 2: Generating diverse replay sequences in a hippocampal place-cell network modelled by a continuous attractor network (CAN) with FRA.
From: Dynamical modulation of hippocampal replay through firing rate adaptation
a Top: schematic of the CAN used to model the place-cell network during navigation on a linear track. Neurons, shown as coloured dots, are arranged according to the spatial locations of their firing fields. Bottom: single-neuron dynamics, with recurrent input from other neurons and FRA illustrated as negative feedback to the neuron. b Top: activity bump within the CAN. Bottom: spatially localised firing fields for all neurons in the network. c Intrinsic speed of the activity bump as a function of adaptation strength. The orange line indicates the FRA strength threshold below which the bump remains stationary. Grey shading highlights example regimes of weak, moderate, and strong FRA strength. d Replay-like sequences under weak adaptation. Top: schematic showing stationary sequences; each vertical bar represents a spike event, coloured by place-cell identity. Each box denotes a time bin, with decoded position shown on the right. Bottom: population activity in the CAN. The simulated animal remains at the bottom of the linear track (horizontal blue line). e Replay-like sequences under moderate adaptation. f Replay-like sequences under strong adaptation.
