Learning to Perceive Non-Native Tones via Distributional Training: Effects of Task and Acoustic Cue Weighting

Area of Science:

• Psycholinguistics
• Auditory Neuroscience
• Speech Perception

Background:

• Distributional learning (DL) enables adult listeners to discriminate challenging non-native speech sounds after brief exposure.
• Previous research demonstrated DL's effectiveness for vowel and consonant contrasts using behavioral methods.
• Neurological underpinnings of DL and its application to non-native lexical tone contrasts remain less understood.

Purpose of the Study:

• To investigate the neurological correlates of distributional learning (DL) for non-native lexical tone contrasts.
• To examine the impact of unimodal versus bimodal sound distributions on perceptual learning of Mandarin tones.
• To assess behavioral and neural responses to DL training in Australian-English speakers.

Main Methods:

• Australian-English speakers participated in DL training using a Mandarin tone contrast.
• Training involved exposure to either bimodal or unimodal distributions of tone tokens.
• Behavioral tasks (discrimination, identification) and neural measures (oddball-EEG) were employed.

Main Results:

• Listeners successfully learned to discriminate Mandarin tones after both unimodal and bimodal training.
• Electroencephalography (EEG) responses indicated enhanced perceptual learning in the bimodal distribution group.
• Behavioral and neural outcomes suggest DL is sensitive to phonetic distance and acoustic cue weighting.

Conclusions:

• Perceptual learning via distributional learning extends to non-native lexical tone contrasts.
• Bimodal sound distributions may facilitate more robust neural learning compared to unimodal ones.
• Findings inform models of speech learning, highlighting the roles of auditory and phonetic constraints.