How words ripple through bilingual hands: Motor-language coupling during L1 and L2 writing

Abstract

The speed of our hand movements can be affected by concurrent processing of manual action verbs (MaVs). Whereas this phenomenon is well established for native languages (L1s), it remains underexplored in late foreign languages (L2s), especially during highly automatized tasks. Here we timed keystroke activity while Spanish-English bilinguals copied MaVs, non-manual action verbs, and non-action verbs in their L1 and L2. Motor planning and execution dynamics were indexed by first-letter lag (the time-lapse between word presentation and first keystroke) and whole-word lag (the time-lapse between first and last keystroke), respectively. Despite yielding no effects on motor planning, MaVs facilitated typing execution in L1 but delayed it in L2, irrespective of the subjects' typing skills, age of L2 learning, and L2 competence. Therefore, motor-language coupling effects seem to be present in both languages though they can arise differently in each. These results extend language grounding models, illuminating the role of embodied mechanisms throughout life.

Keywords: Action verbs; Bilingualism; Embodied cognition; Motor-language coupling; Written production.

Fig. 1.

Structure of a trial in the verb-copying task. Each trial began with a 500 ms ocular fixation cross at the center of the screen before the display of each test. A grammatical context appeared (Estoy, in the L1-Spanish task; I am, in the L2-English task) on the screen during 500 ms until the target verb became visible. The target verb appeared as a gerundio in L1-Spanish (e.g., planchando) and as a present participle in L2-English (e.g., ironing). Participants were instructed to type the target verb as fast and accurately as possible in a single uninterrupted gesture. The verb remained on the screen until the participant gave a complete response. New trials were triggered by pressing the spacebar. Trial-onset asynchrony varied randomly between 300 and 500 ms. The same scheme was employed in both experiments. The figure illustrates a single trial from the L2-English task in the manual action verb condition. FLL: first-letter lag (lapse between target onset and first keystroke). WWL: whole-word lag (lapse between first and last keystroke).

Fig. 2.

First-letter lag (FLL) and whole-word lag (WWL) results in Ll-Spanish and L2-English. (A) Normalized FLL for L1 and L2, reflecting motor planning latencies. The panel shows significantly faster performance for L1 than L2, with no significant differences between verb types within each language. (B) Normalized WLL for L1 (Spanish) and L2 (English), reflecting motor execution latencies. The panel shows significantly faster performance for L1 than L2, with different between-condition patterns in each language: in L1, WLL was shorter for MaVs than nMaVs and similar to that of NaVs; in L2, this measure yielded longer latencies for MaVs than nMaVs and NaVs. Error bars represent standard errors. Asterisks indicate statistically significant differences between conditions. MaVs: manual action verbs; nMaVs: non-manual action verbs; NaVs: non-action verbs.