Interaction of Variation and Expertise in Abstract Concept Learning

• In Event: Poster Session 2
  In Poster Session: Using, Integrating, and Evaluating Information

Abstract

Learning abstract concepts by studying, then comparing, multiple examples has the potential to promote both understanding and generalization. However, little is known about how to select the examples and whether this selection should depend on learner characteristics. Two experiments investigated the effects of variation across examples and learner expertise on learning of a mathematical concept. Better outcomes resulted from low variation for less expert learners, but from high variation for more expert learners. The results suggest that low variation supports comprehension of underlying structure, while high variation facilitates separation of structure from idiosyncratic details. The optimal position in this trade-off depends on learner expertise, with more expert learners coping with, and benefiting from, more variability than less expert learners.

Authors

• David William Braithwaite, Indiana University - Bloomington

• Robert Goldstone, Indiana University