Auditory priming releases Chinese speech from informational masking

Abstract

Before an English speech sentence is presented, hearing or reading the sentence without the last key word improves recognition of the last key word if the full-length speech sentence is presented under speech masking but not under noise masking. This phenomenon suggests a content priming effect on releasing speech from informational masking. To determine whether the priming effect extends to tonal Chinese speech, and, in particular, whether it can be induced by the target talkers voice, in the present study, listeners were presented with either same-voice/different-sentence primes or same-voice/same-sentence primes before hearing the target sentence in either two-talker-speech masking or noise masking. Under speech masking, each of the two prime types significantly improved recognition of the last key word in the full-length target sentence, but the content priming is stronger than the voice priming. Under noise masking, same-voice/same-sentence primes had a weak but significant priming effect, but same-voice/different-sentence primes had only a negligible priming effect. These results suggest that both content and voice cues can be used by listeners to release Chinese speech from informational masking, but only content cues are useful for releasing Chinese speech from energetic masking. [Work supported by China NSF and Canadian IHR.]