Schematic exemplars as items in multiple-item recognition learning

Abstract

Schematically related dot patterns (i.e., distortions of a common prototype) were employed as either the right items or the wrong items of a multiple-item recognition learning task. Relative to a control list containing unrelated items, the presence of related right items yielded facilitated performance, but only early in practice. By contrast, the presence of related wrong items yielded an inhibitory effect early in practice. The results suggested that shared features are more likely to be extracted from the information present in right items than from the information present in wrong items.