New research from North­eastern pro­fessor of psy­chology Iris Berent and her col­leagues indi­cates that lan­guage and motor sys­tems are intri­cately linked—though not in the way that has been widely believed.

Spoken lan­guages express words by sound pat­terns, some of which are pre­ferred to others. For instance, the sound pat­tern “blog” is pre­ferred to “lbog” in Eng­lish as well as many other lan­guages. The researchers wanted to know what accounts for such preferences—specifically, whether they reflect abstract rules of lan­guage in the brain, or if upon hearing speech people attempt to sim­u­late how those sounds are pro­duced by the speech motor system.

Their find­ings sup­port pre­vious research indi­cating the con­nec­tion between people’s knowl­edge of lan­guage and the motor system; how­ever, that con­nec­tion is dif­ferent than what has been pre­vi­ously assumed. The motor system doesn’t drive lin­guistic pref­er­ence directly, they found. Rather, abstract rules of lan­guage guide lin­guistic pref­er­ence, and these abstract rules can trigger motor action. In other words, motor action is a con­se­quence of—not the cause of—linguistic preference.

Sound pat­terns like “blog” are pre­ferred over those like “lbog” not because they are easy to pro­duce; rather, these syl­la­bles are pre­ferred because they con­form to lin­guistic rules, and con­se­quently they tend to acti­vate the motor system, she said.

What’s more, Berent said these find­ings could have impli­ca­tions in studying language-​​related dis­or­ders that are linked to the motor system. One of those areas is dyslexia, which Berent has been studying for years.

This has huge the­o­ret­ical impli­ca­tions,” said Berent, a cog­ni­tive sci­en­tist whose research exam­ines the nature of lin­guistic com­pe­tence. “The idea that lin­guistic knowl­edge is fully embodied in motor action is a hot topic in neu­ro­science right now. Our study shows that motor action is still very impor­tant in lan­guage pro­cessing, but we show a new twist on the mind-​​body connection.”

The research was pub­lished Monday after­noon in the journal Pro­ceed­ings of the National Academy of Sci­ences. Among Berent’s col­lab­o­ra­tors was Alvaro Pascual-​​Leone, an inter­na­tion­ally renowned neu­rol­o­gist at Beth Israel Dea­coness Med­ical Center in Boston and Har­vard Med­ical School and whose exper­tise in tran­scra­nial mag­netic stim­u­la­tion, or TMS, played a key role in the research. Xu Zhao, PhD’15, a doc­toral stu­dent in Northeastern’s Depart­ment of Psy­chology, and other researchers affil­i­ated with the Beth-​​Israel Dea­coness Med­ical Center, Har­vard Med­ical School, Brigham and Women’s Hos­pital, and Uni­ver­sity of Oxford co-​​authored the paper.

Albert Gal­aburda, a co-​​author on the paper and a pre­em­i­nent neu­rol­o­gist at BIDMC, said, “This study helps to solve a long­standing debate in the lit­er­a­ture: What part of speech depends on expe­ri­ence and what part depends on rel­a­tively experience-​​independent gram­mat­ical rules, or some kind of logic system? Since my pri­mary interest is in language-​​based learning dis­or­ders, par­tic­u­larly dyslexia, this ques­tion can be trans­formed to ask whether dyslexics have a pri­mary dis­order of grammar, or a pri­mary dis­order of the motor system or the poor per­cep­tion of speech reaching their ears when babies.”

The researchers’ find­ings are based on a study in which they sought to gauge the sen­si­tivity of English-​​speaking adults to syl­lable struc­ture. Across lan­guages, syl­la­bles like “blif” are more common than “lbif,” and past research from Berent’s lab found that syl­la­bles like “blif” are easier to process, sug­gesting that these syl­la­bles are pre­ferred. The researchers sought to dis­cover the reason for this pref­er­ence: do ill-​​formed syl­la­bles like “lbif” vio­late abstract rules, or do people have dif­fi­culty in their pro­cessing because these syl­la­bles are hard to produce?

To examine this ques­tion, the researchers used TMS, a non­in­va­sive tech­nique that induces focal cor­tical cur­rent via electro-​​magnetic induc­tion to tem­porarily inhibit spe­cific brain regions. The goal was to find out if dis­rupting par­tic­i­pants’ lip motor regions using TMS would elim­i­nate the pref­er­ence for “blif.”

In the exper­i­ment, par­tic­i­pants were pre­sented with an audi­tory stimulus—either a mono­syl­lable or disyl­lable, for example “blif” or “belif”—and asked to indi­cate if that stim­ulus included one or two syl­la­bles. Two hun­dred mil­lisec­onds before hearing that sound, TMS pulses were admin­is­tered to tem­porarily dis­rupt the lip motor region. The crit­ical com­par­ison con­cerned well-​​formed syl­la­bles (e.g., “blif”) vs. ill-​​formed ones (e.g., “lbif”). The researchers asked whether the dis­rup­tion of the motor system would dis­rupt the dis­ad­van­tage of “lbif.” If people dis­like “lbif” because this pat­tern is dif­fi­cult to artic­u­late, then syl­la­bles like “lbif” should be more sus­cep­tible to TMS, and there­fore once people receive the TMS, their dis­like for “lbif” should be lessened.

They found that TMS pulses did impair par­tic­i­pants’ ability to accu­rately deter­mine the number of syl­la­bles. How­ever, the results flew in the face of the embod­i­ment motor hypoth­esis. The researchers found that ill-​​formed syl­la­bles like “lbif” were least likely to be impaired by TMS, and a sub­se­quent func­tional MRI exper­i­ment found that these syl­la­bles were also least likely to engage the lip motor area in the brain.

The results show that speech per­cep­tion auto­mat­i­cally engages the artic­u­la­tory motor system, but lin­guistic pref­er­ences per­sist even when the lan­guage motor system is dis­rupted. These find­ings sug­gest that, despite their inti­mate links, the lan­guage and motor sys­tems are distinct.

Lan­guage is designed to opti­mize motor action, but its knowl­edge con­sists of prin­ci­ples that are dis­em­bodied and poten­tially abstract,” the researchers concluded.