Gesture - Volume 20, Issue 3, 2021

The research reported here examined the effects of language proficiency, gender, and language dominance on gesture use in the tracking of referents by Persian-English bilinguals. 32 EFL learners were divided into two groups of highly proficient and less proficient speakers with equal number of males and females in each group. They were presented with a video extract and asked to recount the story in both L1 and L2. The sessions were video-recorded and the participants’ speech and gestures were transcribed and coded to compare the total number of word tokens and the proportion of maintained references accompanied by three types of gestures. The results offered an effect for proficiency level as proficient learners produced shorter L2 narratives with fewer spoken referents and co-speech gestures. Regarding language dominance, the results were mixed as both males and females produced longer narratives and more gestures in their dominant language (L1), but fewer spoken references, which could be attributed to the pro-drop phenomenon in Persian. The findings also indicated that females’ narratives were longer with more referring expressions and gestures, suggesting that gender can modulate the relationship between language proficiency/dominance and gesticulation. Finally, abstract deictic gestures were observed to be more frequently used by participants.

While previous studies have shown the importance of visuo-spatial working memory in the processing of co-speech iconic gestures, clear evidence for a potential involvement of the verbal working memory (vWM) is currently lacking. To address this issue, participants in the present study were presented with a dual task paradigm. The main outcome variable was the performance on a Stroop-like gesture task which provides a behavioural index of gesture-speech integration. Participants performed this task under conditions of either high or low concurrent vWM load. Unlike in previous studies, the number of words to remember in the high load condition was determined by their individual verbal span rather than being fixed. Results showed reaction time costs in the form of longer reaction times for semantically incongruent gesture-speech combinations as compared to congruent combinations. However, this semantic congruency effect disappeared when the vWM load increased. This result suggests a causal involvement of verbal working memory capacity in gesture-speech integration.

Gesture and speech are closely associated channels involved in conveying meaning. Previous studies have reported the relationship between gesture use and speech performance for highly proficient speakers. This study extends the investigation of this relationship by correlating the representational gesture rate and complexity, accuracy, and fluency measures for L2 speech produced by lower-intermediate-level EFL learners with Chinese as their L1. Three speech tasks were adopted with two activating concrete concepts and one abstract concepts. The results showed that the representational gesture rate was correlated with meaning-associated but not form-associated speech measures for L2 learners, and that the task type did not influence the relationship between gesture and speech. This study suggests that representational gesture use has a close relationship with meaning-related speech measures for lower proficiency L2 speakers when they talk about both concrete and abstract concepts.

It has been suggested that the gestural accuracy used by speakers of Australian Aboriginal languages like Guugu Yimidhirr and Arrernte to indicate directions and represent topographic features is a consequence of absolute frame of reference being dominant in these languages; and that the lackadaisical points produced by North American English speakers is an outcome of relative frame being dominant in English. We test this claim by comparing locational pointing in contexts of place reference in conversations conducted in two Australian Aboriginal languages, Murrinhpatha and Gija, and in Australian English spoken by non-Aboriginal residents of a small town in north Western Australia. Pointing behaviour is remarkably similar across the three groups and all participants display a capacity to point accurately regardless of linguistic frame of reference options. We suggest that these speakers’ intimate knowledge of the surrounding countryside better explains their capacity to accurately point to distant locations.