Navigating through space in speech and gesture: Effects of speaker proficiency, language type, and event type

Speakers of different languages follow language-specific patterns when describing motion events in their first language (L1)-patterns that are also reflected in their co-speech gestures. In this study, we asked whether these patterns extend to second language (L2) learners, focusing on the speech and gestures produced by adult Persian (L1)-English (L2) bilinguals-with either high or low English proficiency-and by English monolinguals (n = 22/group) when describing different events (self-motion, caused-motion) in their two languages (English, Persian). First looking at productions in English, we found close alignment between gesture and speech across proficiency levels and event types, with a preference for the conflated pattern (i.e., the target English pattern) that synthesizes manner and path components of motion into a single clause or a single gesture. At the same time, the preference for the conflated strategy tended to be stronger among speakers with native or high proficiency than bilinguals with low proficiency. Turning next to productions in both English and Persian only among bilinguals, we found interactions between proficiency level, event type, and language type. Overall bilinguals with high proficiency produced more conflated speech responses than bilinguals with low proficiency in their L2 English (but not L1 Persian) and particularly in caused-motion descriptions; this was a pattern that was also largely reflected in gesture. Our findings highlight event type and proficiency as two key factors that affect patterns of speech and gesture production among bilinguals and provide evidence for close coupling between gesture and speech in L2 production.