Synaptic plasticity in a cerebellum-like structure depends on temporal order

Area of Science:

• Neuroscience
• Sensory Processing
• Computational Neuroscience

Background:

• Cerebellum-like structures in fish function as adaptive sensory processors.
• These structures generate and subtract learned predictions from sensory input, highlighting novel stimuli.
• Predictive signals, like corollary discharge, paired with sensory input create 'negative image' neural responses.

Purpose of the Study:

• To investigate the cellular mechanisms underlying predictive sensory processing in fish.
• To demonstrate the existence and properties of synaptic plasticity in the electrosensory lobe.
• To validate a model of predictive coding involving Purkinje-like cells and parallel fiber synapses.

Main Methods:

• Electrophysiological recordings in the electrosensory lobe of mormyrid electric fish.
• Investigation of synaptic plasticity at parallel fiber synapses onto Purkinje-like cells.
• Analysis of the temporal dependence and directionality of synaptic plasticity (anti-Hebbian).

Main Results:

• Demonstrated reversible, anti-Hebbian plasticity at parallel fiber synapses.
• Showed that synaptic depression occurs only when a postsynaptic spike follows an EPSP within 60 ms.
• Confirmed that this plasticity has the properties required for predictive coding models.

Conclusions:

• Parallel fiber synapse plasticity in electric fish electrosensory lobes supports predictive coding.
• This anti-Hebbian plasticity enables the subtraction of predicted sensory input, enhancing detection of novelty.
• The findings provide cellular-level evidence for adaptive sensory processing mechanisms in the fish brain.