a between-subjects factor. Next, syntactic per-
formance was analyzed through a LMM on
RTs withTraining(tool use versus free hand
versus video in Experiment 2 and tool use
versus free hand versus constrained hand in
Experiment 3) as the between-subjects fac-
tor andSentence(coordinated clauses versus
subject-relative clauses versus object-relative
clauses) andTime(pretest versus posttest) as
within-subjects factors.Subjects,Sentence, and
Timewere added to account for random effects.
rmANOVA was run ond′, withTraining(tool
use versus free hand versus video in Experi-
ment 2 and tool use versus free hand versus
constrained hand in Experiment 3) as the
between-subjects factor andSentence(coor-
dinated clauses versus subject-relative clauses
versus object-relative clauses) andTime(pre-
test versus posttest) as within-subjects factors.
To quantify the robustness of the syntactic
benefits after tool-use training across Experi-
ments 2 and 3, we finally computed the effect
size of the pretest-to-posttest improvement in
the syntactic task ( 41 , 43 ). This was done by
calculating the difference between pretest and
posttest performance divided by the pooled
SD in the pretest (for the entire sample of
participants). The effect size in the object rela-
tive clauses condition was then analyzed in an
rmANOVA withTraining(tool use versus free
hand) andExperiment(Experiment 2 versus
Experiment 3) as between-subjects factors.
Experiments 4 and 5: Statistics
In Experiment 4, to assess the progress in
performance during syntactic training, an
rmANOVA ond′and an LMM on RTs were
conducted. The rmANOVA was performed
withBlock(six blocks) as the within-subjects
factor andTraining(object-relative clauses
versus subject-relative clauses) as the between-
subjects factor. The LMM performed on RTs
included the same within-subject factor, with
SubjectsandBlockas random factors.
Next, motor performance with the tool was
analyzed through an LMM on the total num-
ber of inserted pegs withTraining(object-
relative clauses versus subject-relative clauses)
as the between-subjects factor andTime(pre-
test versus posttest) andBlock(four blocks) as
within-subjects factors.Subjects,Time, andBlock
were added to account for random effects.
Finally, to corroborate our results, motor per-
formance was analyzed using an rmANOVA
on the individual improvement slope (b) with
Training(object-relative clauses versus subject-
relative clauses) as the between-subjects fac-
tor andTime(pretest versus posttest) as the
within-subjects factor. Improvement slope was
obtained by performing a linear regression over
the number of inserted pegs for each partici-
pant before and after syntactic training sep-
arately. Slopes were also compared against
zero with one-samplettests for object relatives
and subject relatives separately. No difference
against zero predicted no improvement (i.e.,
flat slope), whereas a significant difference pre-
dicted a change in performance, as indexed
by a positive (i.e., increased performance) or
negative (i.e., decreased performance) slope.
Bonferroni correction was applied to correct
Pvalues for multiple comparisons.
In Experiment 5, similar models to Experi-
ment 4 were applied. To assess the progress
in performance during syntactic training, we
conducted an rmANOVA ond′and an LMM
on RTs. The rmANOVA was performed with
Block(six blocks) as the within-subjects fac-
tor andMotor Test(tool versus constrained
hand) as the between-subjects factor. The
LMM performed on RTs included the same
within-subject factor, withSubjectsandBlock
as random factors. To evaluate motor per-
formance, an rmANOVA was performed on
the number of pegs inserted in pretest and
posttest. For pretest, we considered the num-
ber of pegs inserted in the last block, namely
when the participants reached the required
motor threshold (eight pegs). Conversely, for
posttest, we calculated the mean across the
four blocks performed. The rmANOVA was
performed withTime(pretest versus post-
test) as the within-subjects factor andMotor
Test(tool versus constrained hand) as the
between-subjects factor. An rmANOVA was
also conducted by considering only the post-
test performance across the four blocks per-
formed. The rmANOVA includedBlock(four
blocks) as the within-subjects factor andMotor
Test(tool versus constrained hand) as the
between-subjects factor. To corroborate the
effects found in the Experiment 4, paired-
samplettests were performed between the
last pretest block (i.e., when the threshold
was reached) and each of the posttest blocks
separately for the tool-use and constrained-
hand conditions. Two-samplettests were
additionally conducted to compare the number
of pegs inserted between the two motor con-
ditions for each posttest block. Bonferroni
correction was applied to account for multiple
comparisons. As in Experiment 4, we ana-
lyzed the improvement slope (b). This was
obtained by performing a linear regression
over the number of inserted pegs for each par-
ticipant after syntactic training. Slopes were
also compared against zero with one-sample
ttests for tool use and constrained hand sepa-
rately. Bonferroni’s correction was applied to
correctPvalues for multiple comparisons.REFERENCESANDNOTES- L. E. Milleret al., Sensing with tools extends somatosensory
processing beyond the body.Nature 561 , 239–242 (2018).
doi:10.1038/s41586-018-0460-0; pmid: 30209365 - S. H. Johnson-Frey, The neural bases of complex tool use in
humans.Trends Cogn. Sci. 8 , 71–78 (2004). doi:10.1016/
j.tics.2003.12.002; pmid: 15588811 - P. M. Greenfield, Language, tools and brain: The ontogeny and
phylogeny of hierarchically organized sequential behavior.
Behav. Brain Sci. 14 , 531–551 (1991). doi:10.1017/
S0140525X00071235- K. Pastra, Y. Aloimonos, The minimalist grammar of action.
Philos. Trans. R. Soc. Lond. B Biol. Sci. 367 , 103–117 (2012).
doi:10.1098/rstb.2011.0123; pmid: 22106430 - J. Steele, P. F. Ferrari, L. Fogassi, From action to language:
Comparative perspectives on primate tool use, gesture and the
evolution of human language.Philos. Trans. R. Soc. Lond.
B Biol. Sci. 367 ,4–9 (2012). doi:10.1098/rstb.2011.0295;
pmid: 22106422 - M. Ben-Shachar, T. Hendler, I. Kahn, D. Ben-Bashat,
Y. Grodzinsky, The neural reality of syntactic transformations:
Evidence from functional magnetic resonance imaging.
Psychol. Sci. 14 , 433–440 (2003). doi:10.1111/
1467-9280.01459; pmid: 12930473 - L. W. Barsalou, Grounded cognition.Annu. Rev. Psychol. 59 ,
617 – 645 (2008). doi:10.1146/annurev.
psych.59.103006.093639; pmid: 17705682 - F. Pulvermüller, L. Fadiga, Active perception: Sensorimotor
circuits as a cortical basis for language.Nat. Rev. Neurosci. 11 ,
351 – 360 (2010). doi:10.1038/nrn2811; pmid: 20383203 - L. Craighero, L. Fadiga, G. Rizzolatti, C. Umiltà, Action for
perception: A motor-visual attentional effect.J. Exp. Psychol.
Hum. Percept. Perform. 25 , 1673–1692 (1999). doi:10.1037/
0096-1523.25.6.1673; pmid: 10641315 - C. Brozzoliet al., Touch perception reveals the dominance of
spatial over digital representation of numbers.Proc. Natl. Acad.
Sci. U.S.A. 105 , 5644–5648 (2008). doi:10.1073/
pnas.0708414105; pmid: 18385382 - M. H. Fischer, A. D. Castel, M. D. Dodd, J. Pratt, Perceiving
numbers causes spatial shifts of attention.Nat. Neurosci. 6 ,
555 – 556 (2003). doi:10.1038/nn1066; pmid: 12754517 - O. Hauk, I. Johnsrude, F. Pulvermüller, Somatotopic
representation of action words in human motor and premotor
cortex.Neuron 41 , 301–307 (2004). doi:10.1016/S0896-6273
(03)00838-9; pmid: 14741110 - S. M. Wilson, A. P. Saygin, M. I. Sereno, M. Iacoboni, Listening
to speech activates motor areas involved in speech production.
Nat. Neurosci. 7 , 701–702 (2004). doi:10.1038/nn1263;
pmid: 15184903 - M. A. Just, P. A. Carpenter, T. A. Keller, W. F. Eddy,
K. R. Thulborn, Brain activation modulated by sentence
comprehension.Science 274 , 114–116 (1996). doi:10.1126/
science.274.5284.114; pmid: 8810246 - M. Makuuchi, J. Bahlmann, A. Anwander, A. D. Friederici,
Segregating the core computational faculty of human language
from working memory.Proc. Natl. Acad. Sci. U.S.A. 106 ,
8362 – 8367 (2009). doi:10.1073/pnas.0810928106;
pmid: 19416819 - A. Santi, Y. Grodzinsky, Working memory and syntax interact in
Broca’s area.Neuroimage 37 ,8–17 (2007). doi:10.1016/
j.neuroimage.2007.04.047; pmid: 17560794 - P. Fazioet al., Encoding of human action in Broca’s area.
Brain 132 , 1980–1988 (2009). doi:10.1093/brain/awp118;
pmid: 19443630 - A. C. Royet al., Syntax at hand: Common syntactic structures
for actions and language.PLOS ONE 8 , e72677 (2013).
doi:10.1371/journal.pone.0072677; pmid: 23991140 - F. Vargha-Khadem, K. Watkins, K. Alcock, P. Fletcher,
R. Passingham, Praxic and nonverbal cognitive deficits in a
large family with a genetically transmitted speech and
language disorder.Proc. Natl. Acad. Sci. U.S.A. 92 , 930– 933
(1995). doi:10.1073/pnas.92.3.930; pmid: 7846081 - F. Pulvermüller, The syntax of action.Trends Cogn. Sci. 18 ,
219 – 220 (2014). doi:10.1016/j.tics.2014.01.001;
pmid: 24582824 - M. T. Ullman, A neurocognitive perspective on language: The
declarative/procedural model.Nat. Rev. Neurosci. 2 , 717– 726
(2001). doi:10.1038/35094573; pmid: 11584309 - E. Koechlin, T. Jubault, Broca’s Area and the Hierarchical
Organization of Human Behavior, Broca’s area and the
hierarchical organization of human behavior.Neuron 50 ,
963 – 974 (2006). doi:10.1016/j.neuron.2006.05.017;
pmid: 16772176 - L. Maffongelliet al., Distinct brain signatures of content and
structure violation during action observation.Neuropsychologia
75 , 30–39 (2015). doi:10.1016/j.neuropsychologia.2015.05.020;
pmid: 26004058 - M. J. D. Martins, R. Bianco, D. Sammler, A. Villringer, Recursion in
action: An fMRI study on the generation of new hierarchical levels
in motor sequences.Hum. Brain Mapp. 40 , 2623–2638 (2019).
doi:10.1002/hbm.24549; pmid: 30834624 - A. C. Roy, M. A. Arbib,The Syntactic Motor System
(John Benjamins Publishing, 2005).
Thibaultet al.,Science 374 , eabe0874 (2021) 12 November 2021 13 of 14
RESEARCH | RESEARCH ARTICLE