The study investigates the synaptic connectivity in local cortical circuits, specifically in layer 5 pyramidal neurons of the rat visual cortex. Using simultaneous quadruple whole-cell recordings from hundreds of neurons, the researchers found several nonrandom features in synaptic connectivity. They confirmed that bidirectional connections are more common than expected in a random network and identified several highly clustered three-neuron connectivity patterns. The distribution of synaptic connection strength was found to differ significantly from a Poisson distribution and was better fit by a lognormal distribution, indicating a heavy tail and a concentration of strong connections among a few synaptic connections. Additionally, the strengths of connections sharing pre- or postsynaptic neurons were correlated, suggesting that strong connections are even more clustered than weak ones. These findings suggest that the local cortical network structure can be viewed as a skeleton of stronger connections embedded in a sea of weaker ones, which may play a crucial role in network dynamics. The study highlights the importance of analyzing few-neuron connectivity patterns to understand the network's architecture and dynamics.The study investigates the synaptic connectivity in local cortical circuits, specifically in layer 5 pyramidal neurons of the rat visual cortex. Using simultaneous quadruple whole-cell recordings from hundreds of neurons, the researchers found several nonrandom features in synaptic connectivity. They confirmed that bidirectional connections are more common than expected in a random network and identified several highly clustered three-neuron connectivity patterns. The distribution of synaptic connection strength was found to differ significantly from a Poisson distribution and was better fit by a lognormal distribution, indicating a heavy tail and a concentration of strong connections among a few synaptic connections. Additionally, the strengths of connections sharing pre- or postsynaptic neurons were correlated, suggesting that strong connections are even more clustered than weak ones. These findings suggest that the local cortical network structure can be viewed as a skeleton of stronger connections embedded in a sea of weaker ones, which may play a crucial role in network dynamics. The study highlights the importance of analyzing few-neuron connectivity patterns to understand the network's architecture and dynamics.