Schema Acquisition and Sources of Cognitive Load

Kalyuga, S. (2010). Schema Acquisition and Sources of Cognitive Load. In J.L. Plass, R. Moreno, & R. Brünken (Eds.), Cognitive Load Theory, ch. 3. New York: Cambridge.


Schemas represent knowledge as patterns of relationship between elements
schema: chunk of information 
novice - expert difference: experts have largely interconnected set of domain specific schematic knowledge structures. These structures become guidance during high level cognitive processing. Without this guidance learners have to go through random search which are cognitively inefficient, time consuming and causes heavy load on working memory which would interfere with construction of schemas
- new information is encoded in terms of existing schemas
- preexisting schemas resist change for example schemas acquired during everyday experiences may reject to comprehend scientific information 
- instructional methods should tune up to student's existing schemas in order to reduce the cognitive load.


goal of learning: acquisition and automation of schematic knowledge structures in long term memory


direct initial instruction principle: direct instructional explanations and guidance is like schema base executive for a novice learner (who doesn't have schemas related to the concept). 
Problem solving or discovery learning methods are inefficient executive that results in extraneous cognitive load.


expertise principle: ensures that at each stage of instruction, learner's level of expertise is taken into account to reduce the extraneous cognitive load. For example, by giving appropriate guidance and eliminating redundant instruction for an intermediate learner. 


small step size of knowledge change principle: if there is no schema based on prior knowledge, and the instructions are too far away from learner's level, too many new elements will be introduced. And too many new interacting elements will cause high intrinsic load, which will become extraneous load. To reduce this, the instructions can be given in small step sizes with manageable load within each step. Too many new elements cannot be encoded as small chunks, but instead can be given as limited step sizes increments of knowledge change, not a rapid alteration.