Biology subjects

Hilgetag, C.

Publications and source records attributed to Hilgetag, C..

5 recordsLinked to original sources

bioRxiv2025

Human cortical networks trade communication efficiency for computational reliability.

Brains are often described as cost-efficient communication networks, optimally balancing the cost of long connections with the benefits of fast communication. Here, inspired by the "use it or lose it" principle, we present a novel game-theoretic model of self-organizing neural units and show that the brain is, in fact, sub-optimal in both regards: First, we demonstrate that regional competition for connectivity under propagative communication dynamics naturally gives rise to network configurations similar to those derived from the human cortex while being even more efficient and economical. Next, we use a reservoir computing framework to compare the information processing capacity of these networks against those of the brain. Although comparable in performance, the more optimal trade-off comes with a tax on computational reliability. Through synthetic lesions, we show that these networks are fragile because, to optimize for communication, they funnel information through a spatially clustered "oligarchy" comprising a tight set of transmodal hubs. In contrast, the human brain uses a more distributed "rich club" backbone that better resists breakdown following targeted attacks, even when it means higher wiring costs and less efficient communication. This reveals a previously overlooked principle: cortical networks trade both cost and efficiency for reliable computation. Thus, our findings highlight computational reliability as another, and even more prominent driver of brain connectivity compared to wiring cost and communication efficiency. TeaserThe brain avoids fragile efficiency: brain networks favor reliable computation, revealing resilience as a hidden design rule.

neuroscience↗

bioRxiv2025

A Computational Perspective on the No-Strong-Loops Principle in Brain Networks

Cerebral cortical networks in the mammalian brain exhibit a non-random organization in which reciprocal projections, although widespread, are systematically asymmetric in strength: feedforward connections are consistently stronger than their feedback counterparts, particularly in sensory cortices. This "no-strong-loops" principle is thought to prevent runaway excitation and maintain stability, yet its actual computational impact remains unclear. Here, we use computational analysis and modeling to show that connectivity asymmetry supports high working-memory capacity, whereas increasing reciprocity reduces memory capacity and representational diversity in reservoir-computing models of recurrent neural networks. We systematically examine synthetic architectures inspired by mammalian cortical connectivity and find that sparse, modular, and hierarchical networks achieve superior performance, relative to random, small-world, or core-periphery graphs, but only when reciprocity is constrained. Validated on directed mammalian (macaque, marmoset, rat, and mouse) connectomes, these results indicate that restricting reciprocal motifs yields functional benefits in sparse networks, consistent with an evolutionary strategy for stable, efficient information processing in the brain. These findings suggest a biologically-inspired design principle for artificial neural systems.

neuroscience↗

bioRxiv2024

Ultrastructural sublaminar-specific diversity of excitatory synaptic boutons in layer 1 of the adult human temporal lobe neocortex

Layer (L)1, beside receiving massive cortico-cortical, commissural and associational projections, is the termination zone of tufted dendrites of pyramidal neurons and the area of Ca2+ spike initiation. However, its synaptic organization in humans is not known. Quantitative 3D-models of excitatory synaptic boutons (SBs) in layer 1 of the human temporal lobe neocortex were generated from neocortical biopsy tissue using transmission electron microscopy, 3D-volume reconstructions and TEM tomography. Particularly, the size of active zones (AZs) and the readily releasable, recycling and resting pool of synaptic vesicles (SVs) were quantified. The majority of excitatory SBs contained numerous mitochondria comprising [~]7% of the total volume, had a large macular, non-perforated AZ ([~]0.20 {micro}m2) and were predominantly located on dendritic spines. Excitatory SBs had a total pool of [~]3500 SVs, a relatively large readily releasable ([~]4 SVs), recycling ([~]470 SVs) and resting ([~]2900 SVs) pool. Astrocytic coverage of excitatory SBs suggests both synaptic cross talk or removal of spilled glutamate by astrocytic processes at synaptic complexes. The structural composition of SBs in L1 may underlie the function of L1 networks that mediate, integrate and synchronize contextual and cross-modal information, enabling flexible and state-dependent processing of feedforward sensory inputs from other layers of the cortical column.

neuroscience↗

bioRxiv2024

A General Framework for Characterizing Optimal Communication in Brain Networks

Communication in brain networks is the foundation of cognitive function and behavior. A multitude of evolutionary pressures, including the minimization of metabolic costs while maximizing communication efficiency, contribute to shaping the structure and dynamics of these networks. However, how communication efficiency is characterized depends on the assumed model of communication dynamics. Traditional models include shortest path signaling, random walker navigation, broadcasting, and diffusive processes. Yet, a general and model-agnostic framework for characterizing optimal neural communication remains to be established. Our study addresses this challenge by assigning communication efficiency through game theory, based on a combination of structural data from human cortical networks with computational models of brain dynamics. We quantified the exact influence exerted by each brain node over every other node using an exhaustive multi-site virtual lesioning scheme, creating optimal influence maps for various models of brain dynamics. These descriptions show how communication patterns unfold in the given brain network if regions maximize their influence over one another. By comparing these influence maps with a large variety of brain communication models, we found that optimal communication most closely resembles a broadcasting model in which regions leverage multiple parallel channels for information dissemination. Moreover, we show that the most influential regions within the cortex are formed by its rich-club. These regions exploit their topological vantage point by broadcasting across numerous pathways, thereby significantly enhancing their effective reach even when the anatomical connections are weak. Our work provides a rigorous and versatile framework for characterizing optimal communication across brain networks and reveals the most influential brain regions and the topological features underlying their optimal communication.

neuroscience↗

bioRxiv2023

Multi-Scale Spiking Network Model of Human Cerebral Cortex

Although the structure of cortical networks provides the necessary substrate for their neuronal activity, the structure alone does not suffice to understand it. Leveraging the increasing availability of human data, we developed a multi-scale, spiking network model of human cortex to investigate the relationship between structure and dynamics. In this model, each area in one hemisphere of the Desikan-Killiany parcellation is represented by a 1 mm2 column with a layered structure. The model aggregates data across multiple modalities, including electron microscopy, electrophysiology, morphological reconstructions, and DTI, into a coherent framework. It predicts activity on all scales from the single-neuron spiking activity to the area-level functional connectivity. We compared the model activity against human electrophysiological data and human resting-state fMRI data. This comparison reveals that the model can reproduce aspects of both spiking statistics and fMRI correlations if the cortico-cortical connections are sufficiently strong. Furthermore, we show that a single-spike perturbation propagates through the network within a time close to the limit imposed by the delays.

neuroscience↗