A distributed dynamic brain network mediates linguistic tone representation and categorization.

Area of Science:

• Neuroscience
• Psycholinguistics
• Auditory Perception

Background:

• Speech categorization involves processing sensory input and resolving noise for decision-making.
• Neural underpinnings of speech categorization, particularly for lexical tones, are not fully understood.
• Understanding tone categorization is crucial for deciphering linguistic pitch patterns.

Purpose of the Study:

• To investigate the neural representation and categorization of lexical tones in native Mandarin speakers.
• To examine how acoustic and contextual variabilities affect tone categorization.
• To explore the relationship between neural representations and decision-making parameters.

Main Methods:

• Functional magnetic resonance imaging (fMRI) was used to measure brain activity in 31 native Mandarin speakers.
• Multivariate pattern analysis and univariate activation analyses were employed to assess neural representations.
• An evidence accumulation model of decision-making was utilized to analyze behavioral data.

Main Results:

• Acoustically variable-tolerant representations of lexical tone categories were found in the middle portion of the left superior temporal gyrus (STG).
• Frontal and parietal regions showed category-relevant information sensitive to different types of variability.
• Neural representation robustness in a fronto-temporoparietal network correlated with decision-making parameters.

Conclusions:

• A hybrid model is supported, with a representational core in the STG.
• An extensive frontoparietal network dynamically supports the representation and categorization of linguistic pitch.
• This network integrates sensory information and decision-making for successful tone perception.