What distinguishes good readers from poor ones is simply . . .

From E.D. Hirsch's Cultural Literacy

Once the relevant knowledge has been acquired, the skill follows. General programs contrived to teach general skills are ineffective. AI research shows that experts perform better than novices not because they have more powerful and better oiled intellectual machinery but because they have more relevant and quickly available information. What distinguishes good readers from poor ones is simply the possession of a lot of diverse, task-specific information.

Probably the most dramatic illustrations of the knowledge-bound character of human skills came from some remarkable experiments conducted by Adriaan de Groot, a Dutch psychologist, who described his findings in a book entitled Het Denken van den Schaker (literally, "the thinking of chess players"). De Groot discovered that chess masters are astonishingly skilled at remembering and reproducing chess positions after a very brief exposure to them. The subjects in his experiments were players of various abilities, as indicated by their official chess rankings. In one experiment, de Groot displayed for five to ten seconds a chess position from an actual game in which twenty-five pieces were left on the board. Grand masters performed this feat with 100 percent accuracy, masters with 90 percent accuracy. Weaker players were lucky if they could correctly place five or six pieces.

Then de Groot varied the conditions of his experiment in one respect. Instead of placing the twenty-five pieces in positions from an actual game, he placed them on the board randomly. The results were unexpected. All his subjects - grand masters, masters, class A players, and class B players- performed the same as novices did, placing only five or six pieces correctly. This experiment has been duplicated in several different laboratories, and structurally in several other fields, including algebra, physics, and medicine, always with the same striking results.