Knowledge construction relies on active mental processing of perceptions. It results in understanding, which results from
generative processing. That is knowledge construction is a generative learning process (Wittrock, 1974). Generative processing involves
relating new information to prior knowledge in order to build more elaborate knowledge structures. These knowledge structures are
necessary for interpreting new information, reasoning from what is known, and for solving problems. Modern theory describes generative
learning in terms of deeper levels of processing (Craik & Lockhart,1972). Normally only the results of the deeper analyses can be regarded
as learning, the by-products of preliminary or'surface' analysis are discarded. What is needed is meaning, and the extraction of meaning
involves the deeper levels of processing. Depth of processing may be better thought of as the kind, and number, of elaborations generated
by the learner (Eysenck & Keane, 1990). Indeed, the superior learning that results from material that has been acted upon, and the
finding that material is better remembered if it is actually generated by the learner (rather than equivalent material merely presented to
the learner) has been termed the generation effect (eg Gardiner, 1989). 
<p>
From <a href=http://www.icbl.hw.ac.uk/ctl/mayes/paper11.html>David Jonassen, Terry Mayes, and Ray McAleese</a> (1993) A Manifesto for a Constructivist Approach to Technology in Higher Education. in Duffy, Jonassen and Lowyck (Eds), <em>Designing constructivist learning environments</em>