Topic: attractor network

Learning enables animals to construct flexible knowledge structures, allowing them to adapt to new tasks by reconfiguring prior experiences into novel strategies. This compositional nature of learning suggests that fundamental building blocks from past experiences are essential for future problem-solving. To investigate how experience shapes learning, we trained Recurrent Neural Networks (RNNs) to model a complex, context-dependent odor timing task, using constraints derived from prior reports detailing mouse behavior and shaping procedures. The task, a temporal variant of the Delay Non-match

The head-direction (HD) system is an essential component of the spatial navigation system and is believed to provide key signals during sleep to support memory formation in the downstream parahippocampal regions. The HD signal is processed by a mammillary-thalamic-subicular loop within the Papez circuits. Specifically, the lateral mammillary nucleus (LMN) receives an angular velocity signal from the brainstem and is believed to form an attractor network supporting the generation of the HD signal. By recording from ensembles of LMN neurons in freely moving mice during wakefulness and sleep, we