The activity-dependent development of a neocortical network

Developing neuronal networks undergo prominent activity-dependent modifications during late prenatal / neonatal stages and during so-called critical periods in early postnatal development (Fig. 3). During late prenatal and neonatal development (the so-called precritical period), rhythmic network activity can be observed in various neuronal structures, such as the spinal cord (Hanson and Landmesser, 2003;Saint-Amant and Drapeau, 2001), retina (Meister et al., 1991;Galli and Maffei, 1988) and hippocampus (Ben-Ari et al., 2007;Crépel et al., 2007;Lahtinen et al., 2002;Leinekugel et al., 2002). In the newborn rodent cerebral cortex, synchronized oscillatory activity may be driven by sensory inputs from the retina or whiskers (Hanganu et al., 2006;Milh et al., 2007;Yang et al., 2009), generated spontaneously by intracortical mechanisms (Minlebaev et al., 2007;Sun and Luhmann, 2007) or elicited by activation of metabotropic receptors (Wagner and Luhmann, 2006;Kilb and Luhmann, 2003;Calderon et al., 2005) (for review Khazipov and Luhmann, 2006). It has been demonstrated in the newborn mouse visual cortex that the development of precise cortical maps requires patterned spontaneous activity in the retina (Cang et al., 2005). Abnormally patterned retinal activity causes a failure in the refinement of the retinogeniculate circuitry (Demas et al., 2006). It has been further postulated that correlated rhythmic activity plays an instructive role for the development of proper connectivity, cortical modular architecture and cortical maps during earliest stages already clearly before the critical period (Feller and Scanziani, 2005b). During this precritical period, normal activity patterns are also required for the expression of specific guidance molecules (Nicol et al., 2007). However, the exact nature of the in vivo activity patterns, their large-scale spatial and temporal properties, the role of the sensory input and the thalamus in generating or modulating the early neocortical activity patterns are only poorly understood.

            The activity-dependent, but experience-independent formation of early neuronal networks during the precritical period may underlie the formation of crude maps and basic cortical architecture, whereas map refinement takes place during the experience-dependent critical period (for review Feller and Scanziani, 2005c).

Fig. 3: The formation of early neuronal networks relies on genetic information and on electrical activity. During embryonic development, immature neuronal circuits and crude topographic connections are established on the basis of genetic information. With the developmental emergence of gap junctional coupling, voltage-dependent calcium and sodium channels and various neurotransmitter receptors during prenatal and early postnatal stages, neuronal circuits develop highly correlated spontaneous or transmitter evoked electrical activity patterns, which often propagate over long distances within the network (e.g. retinal calcium waves, giant depolarizing potentials or widespread cortical oscillations). During perinatal development (e.g. during "precritical periods"), certain electrical activity patterns induce a specific gene expression. During further postnatal development (e.g. during "critical periods"), the network is modified in an experience dependent manner based on Hebbian learning rules. From: Khazipov, R., Luhmann, H.J. (2006) Early patterns of electrical activity in the developing cerebral cortex of human and rodents. Trends in Neurosciences 29: 414-418.


            The precritical period is followed by a critical period which is characterized by an activity- and experience-dependent structural and functional reorganization of the neocortical network (Fig. 3) (for review (Boisse et al., 2004;Fox, 2002;Van der Loos, 1976). Novel high-speed two-photon microscopy techniques for network imaging have been recently developed in the group of Fritjof Helmchen and will allow the reconstruction of spike train patterns in a neuronal population with near-millisecond precision. Key principles of cortical circuit dynamics during activity- and experience-dependent modifications can be monitored in several hundreds of cells within an identified barrel cortex column.