Kremer Y, Léger JF, Goodman DF, Brette R, Bourdieu L (2011).Late Emergence of the Whisker Direction Selectivity Map in the Rat Barrel Cortex. J Neurosci 31(29):10689-700.
Abstract. In the neocortex, neuronal selectivities for multiple sensorimotor modalities are often distributed in topical maps thought to emerge during a restricted period in early postnatal development. Rodent barrel cortex contains a somatotopic map for whisker identity, but the existence of maps representing other tactile features has not been clearly demonstrated. We addressed the issue of the existence in the rat cortex of an intra-barrel map for whisker movement direction using in vivo two photon imaging. We discovered that the emergence of a direction map occurs long after all known critical periods in the somatosensory system. This map is remarkably specific, taking a pinwheel form centered near the barrel center and aligned to the barrel cortex somatotopy. We suggest that this map may arise from intra-cortical mechanisms and demonstrate by simulation that the combination of spike-timing-dependent plasticity at synapses between layer 4 and layer 2/3 and realistic pad stimulation is sufficient to produce such a map. Its late emergence long after other classical maps suggests that experience-dependent map formation and refinement continue throughout adult life.
Supplementary movie (20 MB). Spikes of L2/3 neurons and evolution of their direction selectivity during the stimulation with random moving bars. The movie shows the spikes of all neurons in L2/3 during the stimulation with random moving bars (left) and the evolution of their direction selectivity (right), as estimated from the average selectivity of their presynaptic neurons in L4, weighted by the synaptic weights. Initial conditions and parameters are identical to those used in the Figure 6. In these conditions, the simulation lasted 1500 seconds. Only 3% of all frames are included in the movie. Note that when a moving bar activates a sequence of whiskers, the first barrel column reached by the bar is almost homogeneously activated (at least at the beginning of the simulation) whereas subsequent barrel columns already exhibit inhomogeneous activity. This inhomogeneous activation of the second, third, etc barrel columns is not a consequence of the existence of a radial direction selectivity map, as this map does not have enough time to develop for instance at the first presentation of the moving bar. It is just due to horizontal connections within layer 2/3, which propagate the activation of the first column to the next adjacent columns, inhibiting predominantly the closest regions to the first column. This inhomogeneous activation of barrel columns becomes reinforced as the direction selectivity map develops, but this contribution only comes in second.
Code. In the examples of the Brian simulator, there is a script that stimulates a plastic model of the barrel cortex with randomly moving bars, and shows the selectivity map after learning.