Firing Rate Adaptation in Continuous Attractor Neural Networks Explains Theta Phase Shift of Hippocampal Place Cells

แนวคิดหลัก

Firing rate adaptation within a continuous attractor neural network causes the neural activity bump to oscillate around the external input, accounting for theta phase precession and procession of individual neurons.

บทคัดย่อ

The paper presents a continuous attractor neural network (CANN) model that explains the theta phase shift of hippocampal place cells. The key insights are:

1. The interplay between the intrinsic mobility of the network bump (due to firing rate adaptation) and the extrinsic mobility (due to location-dependent sensory inputs) leads to an oscillatory tracking state, where the network bump sweeps back and forth around the external input at theta frequency.

2. The forward and backward sweeps of the network bump account for the theta phase precession and procession observed in individual place cells, respectively.

3. The adaptation strength controls whether a place cell exhibits only predominant phase precession (unimodal cells) or interleaved phase precession and procession (bimodal cells).

4. The model also explains other experimental observations, such as the constant cycling of theta sweeps in a T-maze environment, the speed modulation of place cell firing frequency, and the continued phase shift after transient silencing of the hippocampus.

Overall, the model provides a unified mechanistic explanation for the rich dynamics of hippocampal place cell activity during spatial navigation.

สถิติ

"The smooth tracking state occurs when firing rate adaptation does not exist (m = 0)."
"The travelling wave state occurs when the external drive does not exist (α = 0) and the adaptation strength m exceeds a threshold (m > τ/τv)."
"The oscillation frequency ω increases sublinearly with the external input strength α and the adaptation strength m."

คำพูด

"Due to the intrinsic mobility of the network, the bump tends to move at its own intrinsic speed (which is faster than the external moving input), i.e., the bump tries to escape from the external input. However, due to the strong locking effect of the external input, the bump can not run too far away from the location input, but instead, is attracted back to the location input."
"When the firing rate adaptation is relatively small, the bump oscillation frequency can be analytically solved to be ω = sqrt(m/τv - 1/τ)."
"The firing phase of the probe neuron is also constant during the departure stage."

ข้อมูลเชิงลึกที่สำคัญจาก

Firing rate adaptation affords place cell theta sweeps, phase precession and procession

by Chu,T., Ji,Z... ที่ www.biorxiv.org 11-14-2022

https://www.biorxiv.org/content/10.1101/2022.11.14.516400v4

สอบถามเพิ่มเติม

How would the model predictions change if the network connectivity was not translation-invariant, but instead learned through synaptic plasticity during active navigation

If the network connectivity was not translation-invariant but instead learned through synaptic plasticity during active navigation, the model predictions would likely change in several ways. Firstly, the network's stability and ability to maintain attractor states could be influenced by the specific learning rules governing synaptic plasticity. Changes in connectivity patterns based on experience could lead to a more adaptive and context-specific representation of spatial information in the hippocampus. Additionally, the network's ability to generate theta phase shifts and sweeps may be more dynamic and flexible, allowing for rapid adjustments in response to changing environmental demands. The learning process could also introduce variability in the network's responses, potentially leading to more diverse and nuanced encoding of spatial information.

What are the potential computational advantages of having both forward and reverse theta sequences in the hippocampus, beyond their role in episodic memory

Having both forward and reverse theta sequences in the hippocampus can provide several computational advantages beyond their role in episodic memory. One potential advantage is the ability to support flexible and adaptive decision-making processes. The interleaved forward and reverse theta sequences can allow the brain to consider multiple possible future scenarios in a rapid and continuous manner, enabling quick evaluations of different options and facilitating efficient decision-making. Additionally, the presence of both forward and reverse theta sequences may enhance the brain's ability to integrate information across different time scales, supporting the formation of complex associations and facilitating the retrieval of relevant memories in a variety of contexts. Overall, the coexistence of forward and reverse theta sequences may contribute to the brain's capacity for cognitive flexibility, planning, and goal-directed behavior.

Could the mechanism of firing rate adaptation underlying theta sweeps in this model also explain other types of neural dynamics, such as sharp-wave ripple events or mental exploration, in different brain regions

The mechanism of firing rate adaptation underlying theta sweeps in this model could potentially explain other types of neural dynamics in different brain regions, such as sharp-wave ripple events or mental exploration. Firing rate adaptation plays a crucial role in shaping the dynamics of neural activity and can lead to the generation of oscillatory patterns and sequential activations. In the context of sharp-wave ripple events, the firing rate adaptation mechanism could contribute to the generation of high-frequency oscillations and the replay of previously encoded information during offline states. The interplay between intrinsic and extrinsic mobility in the model could also be relevant to mental exploration, where the brain spontaneously generates and explores different cognitive states and scenarios. By modulating the strength of adaptation and external inputs, the model could potentially capture the dynamics of mental exploration and the generation of diverse cognitive states in the brain. Overall, the firing rate adaptation mechanism proposed in this model may have broader implications for understanding a range of neural dynamics and cognitive processes beyond theta sweeps in the hippocampus.