giannelli, f. 1, 2 & de diego balaguer, r. 1, 2
1 Department of Cognition, Development and Educational Psychology, University of Barcelona, Barcelona, Spain
2 Cognition and Brain Plasticity Unit, IDIBELL, L?Hospitalet de Ll., Spain
During language acquisition listeners can segment speech and extract its rules by using prediction mechanisms based on general statistical learning. Prosody is a convergent cue that adds an additional periodic cue and enhances rule learning in natural speech. We investigated whether the effects of prosodic information are due to its periodic convergent characteristics manipulating rhythmicity in both audio and visual domains. We created artificial language streams containing trisyllabic meaningless words where the first element predicted the third one; and the second element was variable. The experiment had two conditions. In the first, participants listened to a language stream characterized by a frequency-modulated pitch gradually decreasing and increasing again. Every syllable had a different pitch, so that informative syllables had the highest or the lowest pitch. In the second condition, pitch was replaced by visual information. Participant listened to a language without pitch modulation while a bar moving up and down appeared on a screen. The appearance of the bar was synchronized with the syllables, so that on listening informative syllables the bar either appeared at the top or at the bottom of the screen. In addition, participants went through the same test while performing a synchronized-tapping task. They had to tap when pitch was at highest or lowest, or when the bar hit the top or bottom of its trajectory. Our hypothesis was that synchronized tapping might improve language rule learning, by enhancing prediction mechanisms. Synchronization with syntactic information should facilitate the extraction of the rules embedded in speech. Immediately after each language exposure, participants were tested on rule learning using two-alternative forced choice tests with isolated test items presented in pairs. Participants had to choose between words and non-words or between rule-words and non-words. Preliminary results will be presented and discussed.