Knowledge-Lean Studies
In the early 1980s, my colleagues and I wanted to identify the generic strate-
gies used by effective readers as they process written text (e.g., Alexander,
Hare, & Garner, 1984; Garner, Hare, Alexander, Haynes, & Winograd,
1984). As with the first generation of expert–novice researchers, we sought
out those strategies that made a difference in learning without regard to the
topic or domain knowledge of the reader. That approach was somewhat suc-
cessful in that we and others identified general strategies that were potent fac-
tors in separating good readers from poor readers (Paris, Lipson, & Wixson,
1983; Pressley, Goodchild, Fleet, Zajchowski, & Evans, 1989). We also worked
diligently to convert those general strategies into procedures that could be ef-
fectively taught to students, and did so with some degree of success (e.g., Alex-
ander, White, Haensly, & Crimmins-Jeanes, 1987; Judy et al., 1988). However,
the transformation from poor reader to good reader was elusive and the pic-
ture of reading development remained blurry and fragmented.
Knowledge-Rich Studies
When the emphasis on general strategic processing did not sufficiently explain
students’ success (or lack thereof ) at text-based learning, my colleagues and I
became more interested in the relationship between topic or domain knowl-
edge and strategy use (Alexander & Judy, 1988). Specifically, Judith Judy and I
(1988) reviewed the literature to ascertain the potential relations between
knowledge and strategy use. Based on that review, we posited that those with
less knowledge were more dependent on general cognitive strategies, but that
some level of relevant knowledge was required for the efficient and effective use
of those strategies. Further, we hypothesized that both general and domain-
specific strategies were critical to learning, albeit in different ways.
Later, my colleagues and I (e.g., Alexander & Kulikowich, 1991; Alexan-
der, Pate, Kulikowich, Farrell, & Wright, 1989) put some of those emerging
hypotheses to the test in an extensive study of knowledge and strategic proc-
essing in history and human biology. Those knowledge–strategy studies were
important catalysts for the formulation of the MDL for several reasons.
First, those investigations, which were often cross-age, cross-topic, and cross-
domain, highlighted the strong but shifting relation between individuals’
knowledge and their successful use of strategies. A simple linear path could
not adequately capture this knowledge–strategy relation. Whereas too much
relevant knowledge meant that certain strategies were of limited value, too
little relevant knowledge meant that strategies were often poorly executed.
Second, while differences in topic or domain resulted in varied patterns
within individuals, performance patterns were quite consistent for topics and
- MODEL OF DOMAIN LEARNING 281