Presentation
Search Abstracts | Symposia | Slide Sessions | Poster Sessions | Lightning Talks
Initial motor skills determine the benefits of tool-use learning over syntax in language
Poster C49 in Poster Session C, Wednesday, October 25, 10:15 am - 12:00 pm CEST, Espace Vieux-Port
Claudio Brozzoli1,2, Mallory Augier2, Roussey Juliette3, Roy Alice C1,4, Boulenger Véronique1,4; 1These authors contributed equally, 2Equipe IMPACT CRNL - INSERM U1028 - Université Claude Bernard Lyon 1, 3Université Aix-Marseille, 4DDL - CNRS U5596 - Université Lumière Lyon 2
Empirical findings during the last decades revealed that language leverages the sensorimotor circuits handling perception and action to process phonemes and semantics. Yet, evidence was scant for an embodied syntax. Recently, we revealed that planning to use a mechanical tool and understanding sentences with a complex syntax (i.e., object-relatives) activate the same brain structures within the Basal Ganglia (BG) and prompt similar patterns of neural activations.1 If tool use and syntax rely on common neural resources, could one benefit from training the other? In line with this prediction, we unveiled cross-domain learning transfer: using a tool improves syntactic comprehension of complex structures in language. Individuals are however not equally dexterous in using a tool and the initial motor skills might determine the learning curve during motor training. The differences in tool-use dexterity might be crucial for the occurrence of syntax learning transfer between tool use and language. According to the exploration-exploitation model of motor learning, different processes during motor learning differently recruit the BG.2 This can affect the occurrence of benefits from tool use to language. Indeed, the initial exploration phase (linear progress) is associated with activity in the anterior caudate, one of the regions shared by tool use and syntactic processes. Once the motor skill is acquired (in the proximity of the plateau of an asymptotic trend), the activation elicited by motor practice shifts to the posterior caudate, a cluster not shared between the two functions.3 Such neuro-functional dynamics of brain activations accompanying motor learning predicts that less dexterous participants with the tool, who might spend more time in the exploration phase during training (with a linear rather than asymptotic trend of learning), ultimately might also gain more benefits for language. We tested (1) whether the individual initial tool-use dexterity impacts motor learning; (2) whether different profiles of motor learning in using the tool (linear vs. asymptotic) determine different benefits for syntactic processes in language. We first assessed tool dexterity in 40 healthy adults with the Purdue pegboard test. We then evaluated syntactic comprehension in the same participants, before and after tool-use training. Our results show that the majority of participants with a lower score at the initial motor assessment with the tool, displayed a linear rather than asymptotic progression during tool-use training. The frequency of participants displaying linear and asymptotic learning curves was instead similar in the group with higher scores in the initial motor assessment. Crucially, a significantly larger improvement in syntax was found in the group of less dexterous participants with the tool (mostly linear learners) compared to the more dexterous ones (equally distributed in linear and asymptotic learners). In conclusion, our findings support the idea of syntax as a domain general function for action and language and point to the need of adapting the tool complexity to individual abilities to better recruit common resources with language. Thibault et al. Science, 2021 Graybiel Curr Opin Neurobiol, 2005 Choi et al. PNAS, 2020
Topic Areas: Syntax and Combinatorial Semantics, Multisensory or Sensorimotor Integration