Figure 1.

Steps of information transmission in neuronal networks. Convergent feed-forward connected neurons (A) are a basic motif for many theories of cortical information processing. A certain degree of convergence is necessary, due to the fact that postsynaptic potentials from single cortical connections are usually too weak to elicit an action potential in the postsynaptic cell. For theories that build on temporally precise signal integration and action potential generation in single neurons, three main steps in activity transfer are critical: synaptic transmission (B), which can show considerable amplitude variability, high failure rates, and temporal jitter (see main text), dendritic integration (C), which has been found to be highly non-linear under certain conditions, and action potential generation (D), the temporal precision of which has been shown to depend on rise time and amplitude of membrane potential deflections, prior to the action potential initiation.