Transfer of statistical learning from speech perception to production generalizes to reading

Area of Science:

• Psycholinguistics
• Speech Science
• Phonetics

Background:

• Short-term exposure to specific speech acoustics can robustly influence speech production.
• Previous studies relied on auditory repetition tasks, leaving the scope of perception-production transfer unclear for tasks without explicit auditory models.

Purpose of the Study:

• To investigate if perception-production transfer in the acoustic-phonetic domain occurs in tasks lacking an auditory model.
• To determine if this transfer generalizes to novel speech stimuli.

Main Methods:

• Two experiments were conducted involving participants exposed to auditory speech samples with varying acoustic statistics.
• Participants then performed reading aloud tasks without auditory feedback.
• Stimuli included real words (BEER/PEER, BEACH/PEACH) and nonwords (BEETH/PEETH).

Main Results:

• Participants produced words differently based on prior exposure to standard versus accented speech samples.
• This effect was observed in direct repetition tasks and generalized to new word and nonword pairs.
• The findings indicate transfer to production even without an auditory model.

Conclusions:

• Perception-production transfer in the acoustic-phonetic domain is not limited to auditory repetition tasks.
• This transfer extends to production tasks that do not involve an explicit auditory model.
• The observed transfer effect generalizes to novel syllables and words, highlighting robust cross-modal interactions in speech.