Washington [US], July 17 (ANI): A new study finds that incoming speech sounds are linked by our brains to our knowledge of grammar, which is quite abstract in nature. But the big question is how does the brain process complex grammatical structures?
A group of researchers from the Max Planck Institute for Psycholinguistics and Radboud University in Nijmegen discovered that the brain encodes the structure of sentences (“the vase is red”) and sentences (“the red vase”) into different neural firing patterns .
Also read | Wall Street’s former Netflix darling on the ropes, losing subscribers sends shares down 70 percent.
The results of the neuroimaging study were published in PLOS Biology.
How does the brain represent sentences? This is one of the fundamental questions of neuroscience, because sentences are an example of abstract structural knowledge that is not directly observable from language. While all sentences are made up of smaller building blocks like words and phrases, not all combinations of words or phrases result in sentences.
Also read | SL vs PAK 1st Test: Babar Azam’s Gritty Hundred keeps Pakistan alive after Prabath Jayasuriya’s five-wicket haul on Day 2.
In fact, listeners need more than just knowing which words go together: they need abstract knowledge of language structure to understand a sentence. So how does the brain encode the structural relationships that make up a sentence?
Lise Meitner group leader Andrea Martin already had a theory about how the brain computes linguistic structures, based on evidence from computer simulations. To further test this ‘time-based’ model of language structure, developed alongside Edinburgh University’s Leonidas Doumas, Martin and colleagues used EEG (electroencephalography) to measure neuronal responses through the scalp.
In a collaboration with first author and doctoral student Fan Bai and MPI director Antje Meyer, she wanted to investigate whether the brain reacts differently to sentences and sentences and whether this could provide clues as to how the brain encodes abstract structures.
The researchers created sets of spoken Dutch phrases (like de rode vaas “the red vase”) and sentences (like de vaas is rood “the vase is red”) that were identical in duration and number of syllables and very similar in meaning. They also created paintings with objects (e.g. a vase) in five different colors. 15 adult native Dutch speakers took part in the experiment.
For each spoken stimulus, they were asked to perform one of three tasks in a random order. The first task was structure-related, as participants had to decide whether they had heard a sentence or a push-button sentence. The second and third task meanings were related, as participants had to decide whether the color or object of the spoken stimulus matched the following image.
As expected from computer simulations, the activation patterns of neurons in the brain for phrases and sentences were different, both in terms of timing and the strength of neural connections. “Our results show how the brain breaks down language into linguistic structures by exploiting the timing and connectivity of neural firing patterns. These signals from the brain provide a novel basis for future research into how our brains generate language,” says Martin.
“Furthermore, the time-based mechanism could in principle be used for machine learning systems that interface to understanding spoken language to represent abstract structures, which machine systems are currently struggling with.” We will conduct further studies on how knowledge is abstract structures and countable Statistical information, such as transition probabilities between language units, is used by the brain in understanding spoken language.” (ANI)
(This is an unedited and auto-generated story from the Syndicated News feed, LatestLY staff may not have modified or edited the content.)
#Science #News #Study #Sentences #connection #brain #latest