Exploring the role of sublexical information in speech perception and misperception

Listeners are able to use previously acquired knowledge to predict the content of upcoming speech when the latter signal is distorted or degraded. However, this process can also result in misperceptions. One paradigm used to investigate comprehension of degraded speech is perceptual priming, in which a previously unintelligible signal suddenly becomes intelligible for listeners following presentation of matching information. Here we investigated the role of sublexical phonotactic information in eliciting misperceptions during perceptual priming. In a series of experiments using a prime-probe design with congruent, incongruent, and neutral conditions, we presented participants with three different types of spectrotemporally degraded probe sentences: real English, nonsense (containing phonological neighbour words but semantically empty), and pseudo sentences (containing nonwords). In the nonsense and pseudo-sentence experiments designed to elicit misperceptions, words and nonwords were phonotactically matched to the reference words using homophonic transformation. For real sentences, accuracy was 95% in the congruent condition. Crucially, “accuracy” rates for congruent nonsense and pseudo-sentences were also high (70% and 74%, respectively), indicating participants were able to assemble lexical information from the context of the clear prime sentence to produce ‘misperceptions’ of the degraded probe. Further, when presented with clear nonsense or pseudo-sentences as primes, participants were able to recognize up to 45% of degraded real probe sentences. These findings show that listeners use sublexical phonotactic information to predict the content of degraded speech. We discuss implications for current accounts of speech perception that have largely ignored a role for phonotactics.