Inhibitory plasticity supports replay generalization in the hippocampus

Area of Science:

• Neuroscience
• Computational Neuroscience
• Synaptic Plasticity

Background:

• Memory consolidation integrates new experiences into long-term memory through sequence replay.
• The precise content and statistical properties of these replays, particularly concerning salient stimuli, are not fully understood.
• Existing models struggle to explain deviations in replay statistics from actual experience.

Purpose of the Study:

• To investigate the role of inhibitory synapse plasticity in shaping memory replay during consolidation.
• To develop a biologically plausible model explaining how synaptic rules influence memory generalization.
• To experimentally validate model predictions in awake behaving mice.

Main Methods:

• Utilized three levels of computational models: leaky integrate-and-fire, biophysically detailed, and abstract binary.
• Incorporated a Hebbian spike-time-dependent plasticity rule at inhibitory synapses.
• Employed optogenetics in awake behaving mice to manipulate neural representations and observe network dynamics.

Main Results:

• Demonstrated that a Hebbian plasticity rule at inhibitory synapses parsimoniously explains deviations in memory replay statistics.
• Showed that this rule enables efficient generalization of learned sequences.
• Confirmed experimentally that artificially implanted non-generalizable representations accumulate inhibition during sharp-wave ripples, validating model predictions.

Conclusions:

• A Hebbian spike-time-dependent plasticity rule at inhibitory synapses provides a unified explanation for memory replay statistics and generalization.
• This synaptic mechanism offers a direct link between synaptic function and cognitive processes in memory consolidation.
• Findings have significant implications for understanding normal learning and neurological disorders affecting memory.