Bidirectional synaptic plasticity rapidly modifies hippocampal representations.

Area of Science:

• Neuroscience
• Synaptic Plasticity
• Computational Neuroscience

Background:

• Neural adaptations underlying learning are primarily mediated by activity-dependent synaptic plasticity.
• Dendritic calcium spikes, or plateau potentials, drive a non-standard form of synaptic plasticity linked to place field formation in hippocampal CA1 neurons.

Purpose of the Study:

• To investigate the role of behavioral timescale synaptic plasticity (BTSP) in reshaping existing place fields.
• To elucidate the dependence of BTSP on postsynaptic activation and synaptic weight.
• To model the impact of BTSP on neural adaptations within a network context.

Main Methods:

• Evoked plateau potentials near existing place fields in rodent hippocampal CA1 neurons.
• Manipulations of place cell membrane potential.
• Computational modeling and network simulations implementing a bidirectional synaptic learning rule.

Main Results:

• Plateau potentials induced less potentiation and more depression near existing place fields, suggesting an inverse dependence on postsynaptic activation.
• Further analysis revealed this effect is driven by a dependence on current synaptic weight: weak inputs potentiate, while strong inputs depress.
• A network model demonstrated that BTSP, governed by this weight-dependent rule, facilitates neural adaptations driven by population activity.

Conclusions:

• Behavioral timescale synaptic plasticity (BTSP) exhibits bidirectional synaptic weight changes that reshape existing neural representations.
• The plasticity rule is dependent on current synaptic weight, not solely postsynaptic activation.
• BTSP enables neural networks to adapt to experience through population activity, highlighting a distinct mechanism from pairwise neuronal correlations.