Dendritic morphology predicts pattern recognition performance in multi-compartmental model neurons with and without...

Area of Science:

• Neuroscience
• Computational Neuroscience
• Biophysics

Background:

• Neuronal structure, particularly dendritic morphology, is crucial for information processing.
• Understanding the relationship between dendritic architecture and neuronal function is key to deciphering neural computation.

Purpose of the Study:

• To investigate the influence of dendritic morphology on a neuron's pattern recognition capabilities.
• To explore how different dendritic tree structures affect the ability of neurons to discriminate input patterns.

Main Methods:

• Utilized algorithms to generate diverse dendritic morphologies with varying numbers of terminal points (22 and 128).
• Constructed multi-compartmental neuron models based on generated dendritic trees.
• Quantified pattern recognition performance by assessing the discrimination of learned and novel input patterns.

Main Results:

• Dendritic morphology significantly affects neuronal pattern recognition performance.
• Neuronal performance shows an inverse correlation with the mean depth of the dendritic tree.
• Dendritic tree asymmetry did not correlate with performance across all morphologies.
• For neurons with dendritic tapering, performance is best predicted by the electrotonic distance of synapses to the soma.

Conclusions:

• Dendritic morphology is a critical determinant of neuronal pattern recognition ability.
• Mean dendritic tree depth is a key factor influencing performance, with shallower trees being more effective.
• These findings hold true for both passive and active neuronal models, highlighting the fundamental role of morphology.