How language familiarity facilitates speaker identification: behavioral and multivariate EEG evidence

Area of Science:

• Cognitive Neuroscience
• Speech Perception
• Psychoacoustics

Background:

• The language-familiarity effect (LFE) demonstrates superior voice identification in native languages compared to foreign languages.
• The cognitive and neural underpinnings of LFE, particularly the specific information processed, remain unclear.

Purpose of the Study:

• To investigate the specific acoustic-phonetic information extracted during speech perception that contributes to the LFE.
• To determine the temporal dynamics of the LFE using electroencephalography (EEG).

Main Methods:

• Two experiments were conducted: behavioral voice identification tasks (Experiment 1) and electroencephalography (EEG) with multivariate decoding (Experiment 2).
• Native Mandarin speakers learned voices from Mandarin and German speech, then identified voices in various conditions (trained/untrained words, untrained language, reversed speech).

Main Results:

• A significant LFE was observed behaviorally, with higher voice identification accuracy for native Mandarin speech across meaningful linguistic conditions, but not for reversed speech.
• EEG analysis revealed that the LFE emerged earlier when processing the native language, indicating faster extraction of relevant features.
• Familiar language facilitated the extraction of both language-specific and language-general acoustic-phonetic features.

Conclusions:

• Language familiarity enhances voice identification by improving the extraction of acoustic-phonetic information.
• The LFE is supported by the efficient processing of familiar linguistic cues, influencing speaker recognition.
• This study elucidates the role of language familiarity in speech perception and speaker identification mechanisms.