A Visual Recall Paradigm to Assess Implicit Statistical Learning Poster

Implicit statistical learning is, by definition, learning that occurs without conscious awareness. However, measures that putatively assess implicit statistical learning often require explicit reflection. ‘Processing-based’ tasks can measure learning without requiring conscious reflection, by measuring processes that are facilitated by implicit statistical learning. In three experiments, we used visual serial recall tasks to demonstrate implicit statistical learning of two different artificial grammars. When stimuli consistently co-occur, it is efficient to ‘chunk’ them into a single cognitive unit, thus reducing working memory demands. Experiments 1 and 2 used a novel artificial grammar designed to facilitate this chunking effect. In these experiments participants showed better recall for predictable sequences than random sequences, demonstrating implicit learning. In experiment 3, we extend this finding to a more traditional artificial grammar, previously used to assess statistical learning in infants, adults and even nonhuman animals, and found strong learning effects. These experiments demonstrate that serial recall tasks are a valuable approach to measure implicit statistical learning, without reliance on conscious decision making or explicit processing.