The prolonged influence of subsequent context on spoken word recognition

Abstract

Connine, Blasko, and Hall (Journal of Memory and Language 30:234-250, 1991) suggested that within a 1-second temporal window, subsequent biasing information can influence the identification of a previously spoken word. Four experiments further explored this hypothesis. Our participants heard sentences in which an ambiguous target word was followed less than or more than a second later by a word biased in favor of either the target word or another word. Overall, the effects of the contextual biases on responding, measured using phonemic restoration and phoneme identification, were almost as large after 1 second as before 1 second. The implications of these results for defining the window of contextual effects are discussed.Keywords Spoken word recognition . Temporal window . Subsequent context effects . Phoneme restoration . Phoneme identification . Psycholinguistics . Speech perception When hearing running speech, a listener must rapidly transform the continuous acoustic signal into the words that the talker intended. This task is complicated by the fact that the quality of the speech signal can vary in a number of ways, such as when the articulation varies in clarity or environmental noise masks the signal. Despite these challenges, listeners typically have little difficulty identifying the intended words. How is this accomplished?One means of achieving successful word recognition is to utilize information in the surrounding sentential context. For example, in the utterance The feathers were on the = ing, where the = represents noise, feathers can help aid identification of the masked phoneme, since only if the masked phoneme were to be identified as /w/ would the sentence make sense. Ample examples have shown preceding context being used in this way (e.g., Borsky, Tuller, & Shapiro, 1998;Cole, Jakimik, & Cooper, 1980;Gaskell & Marslen-Wilson, 2001;Marslen-Wilson & Welsh, 1978;Samuel, 1981; van Alphen & McQueen, 2001). For example, Samuel (1981) demonstrated that the presence of a semantically biasing word caused listeners to perceptually restore a phoneme that had been intermixed with or replaced by noise (e.g., /b/ in ball) to yield a semantically coherent sentence (e.g., The boy swung his bat at the ball). Likewise, Marslen-Wilson and Welsh (1978) demonstrated that a highly constraining preceding context, as in He still wanted to smoke a..., could cause listeners to misperceive a pseudoword such as shigarette as the word cigarette more often than when the pseudoword was preceded by a low-constraining prior context, as in He noticed there was a shigarette.Far less is known about what influence biasing information has on the identification of a preceding word (e.g., The = ing had feathers). It is possible that in order to maximize the speed with which word recognition occurs, if no prior influential context is available, the processor might delay word identification for a short period of time. If no further biasing information is perceived within this window of time, the processor simply commits ...

Inferring causes during speech perception

“…This idea is compatible with episodic accounts and exemplar-based accounts of speech perception that assume rich storage of information (Goldinger, 1998; Johnson, 1997; Pierrehumbert, 2002), i.e. episodic trades of each percept experienced (see also Szostak & Pitt, 2013). An open question is whether other relevant information that influences the listener’s uncertainty about the cause of the pronunciation, such as how likely a given incidental cause is to affect a particular speech sound, is stored alongside these episodes, or only inferred later on during the categorization process.…”

In Spoken Word Recognition, the Future Predicts the Past

“…Here, we focus on much lesser explored postdictive processes, which allow subsequent context to bias perception. This phenomenon has been demonstrated behaviorally (Ganong, 1980;Connine et al, 1991;McQueen, 1991;Samuel, 1991;Gordon et al, 1993;McMurray et al, 2009;Szostak and Pitt, 2013) and has been explained in terms of commitment delay. The system waits to accumulate lexical evidence before settling on an interpretation of a phoneme and maintains subphonemic information until the commitment is made.…”

In spoken word recognition the future predicts the past

“…Here we focus on much lesser explored postdictive processes, which allow subsequent context to bias perception. This phenomenon has been demonstrated behaviourally (Ganong, 1980;Connine et al, 1991;McQueen, 1991;Samuel, 1991;Gordon et al, 1993;McMurray et al, 2009;Szostak and Pitt, 2013), and has been explained in terms of commitment delay: The system waits to accumulate lexical evidence before settling on an interpretation of the phoneme, and maintains sub-phonemic information until the commitment is made. Precisely how the brain implements sub-phonemic maintenance and commitment processes is currently unestablished, but previous research has indicated some likely regions involved.…”

“…Finding maintained sensitivity to subphonemic detail in parallel to phonological commitment is very important for the interpretation of psychophysical research, which has implicitly equated insensitivity to within-category variation with phonological commitment (Connine et al, 1991;Szostak and Pitt, 2013;Bicknell et al, 2015). This previous work has largely converged on a processing model whereby phonological commitment can be delayed for over one second after onset.…”