Statistical learning of auditory-visual associations

Virginie van Wassenhove, Psychology, UCLA - Biology, CalTech

Abstract
Statistical learning refers to the implicit learning of stimulus associations and has been independently reported for visual, auditory and tactile sensory modalities. Here, we tested whether such learning can be observed for synthetic auditory-visual pairings. Joint and conditional probabilities were tested on different groups of participants using a rapid-serial presentation (RSP) paradigm. During the experiment, participants were first exposed to (passively observed) the stimuli. In the second phase of the experiment, participants were presented with ‘singles’ (one auditory or visual stimulus), ‘doublets’ (two stimuli, unimodal or bimodal) or quartets (pairs of bimodal stimuli) in a two-interval forced choice paradigm. They were asked to determine the interval whose stimulus was most frequently occurring during the exposure period. We examined the effects of (i) stimulus duration in the RSP procedure, (ii) joint probability, and (iii) conditional probability on statistical learning. Our results suggest that audio-visual statistical learning occur naturally despite the absence of a task or of an explicit attentional engagement. Additionally, bimodal statistical learning is more efficient than unimodal learning across all stimulus durations.

Not available

Back to Abstract