A minimum-error, energy-constrained neural code is an instantaneous-rate code.

Area of Science:

• Computational Neuroscience
• Neural Coding
• Sensory Systems

Background:

• Sensory neurons encode stimuli via action potentials (spikes).
• Spike generation is metabolically costly, implying a trade-off between encoding accuracy and energy expenditure.
• Existing models propose energy-constrained neural coding, optimizing spike timing to minimize error and energy use.

Purpose of the Study:

• To investigate the relationship between an optimal, energy-constrained neural coder and established coding schemes (rate and temporal codes).
• To derive and validate an instantaneous rate coder that balances energy expenditure and encoding error.
• To unify principles of neural encoding for sensory signals.

Main Methods:

• Derivation of an instantaneous rate coder based on signal properties and their derivatives.
• Analysis of the coder's behavior in high spike rate limits and comparison with an existing optimal coding neuron.
• Validation against experimental data from weakly electric fish afferents and in vitro cortical pyramidal neurons.

Main Results:

• The derived instantaneous rate coder's spike rate depends on the stimulus signal and its derivative.
• In high spike rate regimes, the coder maximizes fidelity under an energy constraint, matching predictions of the optimal coding neuron.
• The instantaneous rate coder accurately predicts spike rates in electric fish and approximates responses to DC step inputs in cortical neurons, aligning with rate and temporal coding measures.

Conclusions:

• The instantaneous rate coder provides a link between energy-constrained optimal encoding and traditional rate and temporal coding schemes.
• This model offers a unifying principle for understanding how sensory neurons encode information efficiently.
• The findings suggest that neural codes are optimized to balance metabolic costs with the fidelity of sensory information transmission.