Simple and complex tool use improve language syntax comprehension


  • Py Raphaël
  • Grosbras Marie-Hélène
  • Brozzoli C.
  • Montant Marie


  • Syntax
  • Language
  • Tool Use
  • Sensorimotor

document type



Do language and motor skills share cognitive processes? It has recently been suggested that a form of hierarchical nesting processing may operate in both skills: motor syntax in manual actions and language syntax in sentence comprehension. Language syntax characterizes the rules governing sentence structures. A sentence is a linear sequence of elements (words) which are dependent one to the other in several levels of nesting. Syntax thus makes it possible to form more or less complex and recursive language structures to generate an infinite diversity of linguistic expressions from a finite number of elements. Goal-directed actions have more recently been described in comparable terms: in an action such as serving coffee, or a simpler action such as moving an object, each element of the operative chain is carried out in a sequential order that similarly includes dependencies between elements and hierarchical nesting. In a previous work relying on the principles of learning transfer, we showed that adult participants trained to enter pegs on a board using a tool significantly improved their performance in a subsequent language syntax comprehension task. In contrast, the control group trained in the same task with the bare hand showed no improvement in the language task. We concluded that using a tool significantly increased the motor syntactic complexity of the operative chain, which resulted in a transfer of syntactic learning from action to language. Here we aimed to investigate further the transfer of syntactic learning by manipulating the hierarchical complexity of action and the presence/absence of a tool. As in the previous study, adult participants performed a language syntax comprehension task, before and after a motor training session. Two groups (n=20 each) were trained with blocks of embedded motor structures (each trial requiring one to three movements), the first group using their hand alone (CH: complex hand), the second group using a tool (CT). Two other groups (n=20 each) were trained with blocks of simple non-embedded actions (only one peg movement), either with their hand (SH: simple hand) or a tool (ST). Our results show that following tool-use training participants improved their overall performance in the language syntax comprehension task, no matter the complexity of the action trained. In addition, we observed (1) a better comprehension of complex syntactic constructions in the CT group; (2) a positive correlation between motor and language performance in the CT group, suggesting that language syntax abilities can predict motor proficiency for complex tool use. Importantly, we did not observe an effect for SH nor CH groups, meaning that motor embedding complexity alone is not sufficient to produce learning transfer. To conclude, our results are in line with the idea that motor skills and language share common domain-general syntactic processes. Also, the learning transfer due to tool use can benefit from the addition of embedded actions.

more information