The inheritance of the dendritic temporal input discrimination

Current developments in neurobiology clarify novel roles for the dendritic tree of neuronal cells. Experiments have shown mechanisms of non­linear synaptic N­Methyl­D­aspartic acid (NMDA) dependent activations, which enable to discriminate temporal input patterns through the amplitudes of the excitatory postsynaptic potentials. Essentially, dendritic branches work as independent discriminating units of specific temporal inputs, thus granting for an enormous increase of the single neuron computational capabilities. The aims of this work are: i) to investigate the discrimination capabilities of neurons by computational modeling of different morphological features of dendritic trees in order to establish the storage power of neurons; ii) to compare neuron storage capacities from different species, brain regions and neuron types. Although several repositories disseminate detailed neuronal morphologies distributed across neuron types and animal species, deep computational investigations require suitable tools for flexible manipulations of dendritic trees. To this purpose, we used the TREES toolbox to reproduce a wide range of dendrite morphologies. The branches of these synthetic tree morphologies were provided by an independent pattern recognizer to simulate the temporal input discrimination by NMDA dependent activations. We observed that such a model was able to learn distinct input patterns by expressing unequivocal output codes. Furthermore, by investigating the relationship between the number of dendritic branches and the number of learnable patterns, we found that these quantities were related by a quadratic law. Comparative analyses revealed remarkable results about the storage power which significantly differed across species, brain regions and neuron types. In conclusion, the results obtained in this work suggest that beyond the dendritic temporal input discrimination allowing, as shown above, for a significant neuronal storage power, a notable direct proportional relationship between the total length of the dendritic tree and the inherent neuronal storage power may exist. Hence, the dendritic morphological features acquire important and direct functional interpretations. Finally, the proposed law suggests relevant anatomical evolutionary considerations about the brain cytoarchitecture among species.