It is no exaggeration to say that genes are essential to nearly every aspect of memory and the process of learning; without them, learning itself wou… - Gary F. Marcus

"How does the body push the comparatively tiny genome so far? Many researchers want to put the weight on learning and experience, apparently believing that the contribution of the genes is relatively unimportant. But though the ability to learn is clearly one of the genome's most important products, such views overemphasize learning and significantly underestimate the extent to which the genome can in fact guide the construction of enormous complexity. If the tools of biological self-assembly are powerful enough to build the intricacies of the circulatory system or the eye without requiring lessons from the outside world, they are also powerful enough to build the initial complexity of the nervous system without relying on external lessons.

The discrepancy melts away as we appreciate the true power of the genome. We could start by considering the fact that the currently accepted figure of 30,000 could well prove to be too low. Thirty thousand (or thereabouts) is, at press time, the best estimate for how many protein-coding genes are in the human genome. But not all genes code for proteins; some, not counted in the 30,000 estimate, code for small pieces of RNA that are not converted into proteins (called microRNA), of "pseudogenes," stretches of DNA, apparently relics of evolution, that do not properly encode proteins. Neither entity is fully understood, but recent reports (from 2002 and 2003) suggest that both may play some role in the all-important process of regulating the IFS that control whether or not genes are expressed. Since the "gene-finding" programs that search the human genome sequence for genes are not attuned to such things-we don't yet know how to identify them reliably-it is quite possible that the genome contains more buried treasure."

There are, of course, many reasons to think that brains operate mostly in parallel. Individual neurons are too slow to allow brains to operate in strict serial von Neumann fashion, and ample data suggest that in any given laboratory task (and by extension, any real-world situation) many different parts of the brain are engaged simultaneously.

"A major push is under way to figure out the molecular basis of those "critical" or "sensitive" periods, to figure out how the brain changes as certain learning abilities come and go. In some, if not all, of those mammals that have the alternating stripes in the visual cortex known as ocular dominance columns, those columns can be adjusted early in development, but not in adulthood. A juvenile monkey that has one eye covered for an extended period of time can gradually readjust its brain wiring to favor the open eye; an adult monkey cannot adjust its wiring. At the end of a critical period, a set of sticky sugar-protein hybrids known as proteoglycans condenses into a tight net around the dendrites and cell bodies of some of the relevant neurons, and in so doing those proteoglycans appear to impede axons that would otherwise be wriggling around as part of the process of readjusting the ocular dominance columns; no wriggling, no learning. In a 2002 study with rats, Italian neuroscientist Tommaso Pizzorusso and his colleagues dissolved the excess proteoglycans with an antiproteoglycan enzyme known as "chABC," and in so doing managed to reopen the critical period. After the chABC treatment, even adult rats could recalibrate their ocular dominance columns. ChABC probably won't help us learn second languages anytime soon, but its antiproteoglycan function may have important medical implications in the not-too-distant future. Another 2002 study, also with rats, showed that chABC can also promote functional recovery after spinal cord injury."