Fig. 2.

The structure of spontaneous activity reflects functional organization and is influenced by experience: (a) an activation pattern obtained from a single frame from spontaneous activity VSD recording (right) matches the orientation map obtained by averaging responses to full-field gratings of vertical orientation (left), in the visual cortex of an anesthetized cat. Adapted from Ref. 43 with permission. (b) Spontaneous waves recorded using VSD in an anesthetized rat barrel cortex, immediately before and after training with a flashing sequence that evoked the wave template represented on the left. Waves are represented by their first frame and the trajectory of their center over $\sim 160\mathrm{ms}$. Spontaneous waves well matched to the template are indicated by a single arrowhead [correlation coefficient $(CC)>0.6$] or by double arrowheads ($CC>0.7$) and are more frequent after training. Adapted from Ref. 45 with permission. (c) Multiunit activity recorded in V1 of awake, freely viewing ferrets either receiving no stimulus (bottom) or viewing natural (top) or artificial stimuli (not shown in this adapted figure) is used to construct neural activity distributions in young and adult animals. Distributions of evoked activities averaged over different stimuli are compared with the distribution of spontaneous activities, assumed to represent the prior expectations about visual features. The internal model of young animals (left) is expected to show little adaptation to the natural environment and thus show a mismatch between spontaneous and evoked distributions. On the contrary, adult animals (right) are expected to be adapted to natural scenes and thus to exhibit a high degree of similarity. Adapted from Ref. 46 with permission.