Parametrically dissociating speech and nonspeech perception in the brain using fMRI.

Area of Science:

• Neuroscience
• Cognitive Neuroscience
• Auditory Perception

Background:

• Understanding the neural basis of speech perception is crucial for cognitive neuroscience.
• Preattentive processing, occurring before conscious awareness, plays a vital role in how the brain decodes complex auditory information like speech.
• Identifying specific brain networks involved in phonetic perception can elucidate fundamental mechanisms of human communication.

Purpose of the Study:

• To identify candidate brain regions forming a neural network for preattentive phonetic perception.
• To investigate how acoustic and phonetic complexity influence brain activity during speech perception.
• To determine if specific brain regions differentiate between natural and synthetic speech.

Main Methods:

• Functional magnetic resonance imaging (fMRI) was employed to measure brain activity.
• Multivariate multiple regression analysis was used to analyze imaging data.
• Stimuli varied across speech/nonspeech, acoustic complexity, phonetic complexity, and natural/synthetic dimensions.

Main Results:

• Seven distributed brain regions showed activity correlated with speech and speech complexity.
• Key regions included left-sided posterior superior temporal gyrus (STG), angular gyrus, ventral occipitotemporal cortex, inferior/posterior supramarginal gyrus, and middle frontal gyrus (MFG).
• Right-sided posterior STG and anterior insula were also identified; only the left MFG distinguished natural from synthetic speech.

Conclusions:

• The findings suggest a distributed neural network for preattentive phonetic perception.
• The left middle frontal gyrus (MFG) plays a specific role in discriminating natural versus synthetic speech.
• The results support a parallel processing model for auditory speech and nonspeech perception.