Retinal Waves Stirring Up a Storm

exploratory bouts in a novel environment? The only way to find out will be to record chronically across multiple units and cortical layers during the actual behavior and plasticity process. Finally, how such a shift in excitatory–inhibitory balance might be achieved is of interest. The brevity of the exploratory behavior required for bidirectional map changes is suggestive. Remarkably, just 4 min/week (of which only a fraction is spent actively scanning) in a In many parts of the developing nervous system, neu-novel environment over a month of deprivation is suffi-ronal connectivity becomes more precise as some in-cient to shrink the spared whisker representation. Focal puts are eliminated while others are maintained. This attention to a stimulus can enhance a cell's peak reprocess of synaptic refinement requires neurotransmis-sponse (Gilbert, 1998) and has most often been linked sion (Lichtman et al., 1998). The temporal properties of with state-dependent activity of neuromodulatory sys-this activity, rather than its mere presence, are thought tems (Weinberger, 1995). Increases in behaviorally rele-to be necessary to drive segregation of competing in-vant inputs require cholinergic input from the basal fore-puts. In particular, much modeling work built on the brain, whose stimulation can also modify cortical maps initial theory by Hebb suggests that synchronous inputs (Kilgard and Merzenich, 1998). It is tempting to speculate are costrengthened, whereas asynchronous firing of in-that the decreases reported by Polley et al. (1999) may puts results in weakening of one set of connections (see, be similarly regulated by diffuse systems that selectively e.g., Miller, 1996). This theory is supported by experi-modulate subsets of inhibitory interneurons (Kawaguchi mental findings in which coactivated inputs are potenti-and Shindou, 1998; Xiang et al., 1998). Would it be possi-ated, whereas inputs are depressed when they are silent ble to replace the very brief behavioral exploration of during postsynaptic activity (Lichtman et al., 1998). Al-a single whisker with exogenous neuromodulators to though the actual withdrawal of synaptic contact due produce similar map changes in nondeprived animals? to asynchronous firing has not yet been observed exper-Is there an advantage to shrinking cortical representa-imentally, this simple theory has provided much impetus tions in the face of rich experience or is this just a by-for a search for endogenous patterned activity during product of competitive mechanisms? Perhaps a smaller, the period of synaptic refinement in the developing ner-tighter representation is more efficient than a widely vous system. distributed one in certain cases. For example, …