Comparing dendritic trees with actual trees

TL;DR

This study compares the structural features of dendritic trees and actual biological trees, revealing greater self-similarity in real trees and highlighting both similarities and differences in their morphology.

Contribution

It provides a quantitative comparison between neuron and tree structures, testing the hypothesis of their similarity and identifying key differences in self-similarity.

Findings

Real trees exhibit more self-similarity than neurons.

Many morphological features are comparable across trees and neurons.

The study opens new avenues for understanding form-function relationships.

Abstract

Since they became observable, neuron morphologies have been informally compared with biological trees but they are studied by distinct communities, neuroscientists, and ecologists. The apparent structural similarity suggests there may be common quantitative rules and constraints. However, there are also reasons to believe they should be different. For example, while the environments of trees may be relatively simple, neurons are constructed by a complex iterative program where synapses are made and pruned. This complexity may make neurons less self-similar than trees. Here we test this hypothesis by comparing the features of segmented sub-trees with those of the whole tree. We indeed find more self-similarity within actual trees than neurons. At the same time, we find that many other features are somewhat comparable across the two. Investigation of shapes and behaviors promises new ways of conceptualizing the form-function link.