Evidence for precategorical extrinsic vowel normalization

Abstract

Three experiments investigated whether extrinsic vowel normalization takes place largely at a categorical or a precategorical level of processing. Traditional vowel normalization effects in categorization were replicated in Experiment 1: Vowels taken from an [ɪ]-[ε] continuum were more often interpreted as /ɪ/ (which has a low first formant, F 1 ) when the vowels were heard in contexts that had a raised F 1 than when the contexts had a lowered F 1 . This was established with contexts that consisted of only two syllables. These short contexts were necessary for Experiment 2, a discrimination task that encouraged listeners to focus on the perceptual properties of vowels at a precategorical level. Vowel normalization was again found: Ambiguous vowels were more easily discriminated from an endpoint [ε] than from an endpoint [ɪ] in a high-F 1 context, whereas the opposite was true in a low-F 1 context. Experiment 3 measured discriminability between pairs of steps along the [ɪ]-[ε] continuum. Contextual influences were again found, but without discrimination peaks, contrary to what was predicted from the same participants' categorization behavior. Extrinsic vowel normalization therefore appears to be a process that takes place at least in part at a precategorical processing level.

High stimulus variability in nonnative speech learning supports formation of abstract categories: Evidence from Japanese geminates

“…Importantly, however, lowvariability training enhanced discrimination sensitivity just as much as high-variability training. This dissociation between tasks (a group difference for identification but not discrimination) is consistent with the view that, while the identification task taps into categorical representations, the 4I2AFC discrimination task taps into pre-categorical processing (Gerrits and Schouten, 2004;Sjerps et al, 2013). This dissociation indicates that the high-variability group's superior accuracy in identifying natural speech materials was not because of their enhanced discrimination sensitivity to timing features, but rather because of their enhanced categorical representations.…”

“…If so, participants who are trained using high-variability geminates should show better identification accuracy than participants trained with low variability. If high-variability participants do not also show better geminate discrimination accuracy, as measured by means of a discrimination task which focuses on pre-categorical processing (4I2AFC; Gerrits and Schouten, 2004;Sjerps et al, 2013), then the high-variability enhancement is not likely to be due to changes in perceptual sensitivity. In contrast, if high-variability participants were to show better identification and discrimination accuracy, then this would indicate that the benefit of high variability is due, at least in part, to enhanced pre-categorical sensitivity.…”

“…These findings support the view that identification and discrimination tasks can tap into different stages in perception (cf. Gerrits and Schouten, 2004;Sjerps et al, 2013), and suggest that linguistic knowledge is applied only at a later stage in perception than that tapped into by discrimination tasks.…”

The singleton-geminate distinction can be rate dependent: Evidence from Maltese

“…Importantly, this does not question that there are auditory processes contributing to rate normalization; it only suggests that they may be enhanced by language experience. Similar arguments have been made for vowel normalization by Sjerps et al (2013) based on the finding that speaker normalization effects are larger the more speech-like the materials are. Larger shifts for VOT are reported by Summerfield (1981) for initial stop voicing in English.…”

Please say what this word is—Vowel-extrinsic normalization in the sensorimotor control of speech.

“…Their results revealed a clear influence of the carrier-phrase on subsequent vowel identification, indicating that listeners use spectral properties from the immediate context as a frame of reference for their analysis of subsequent vowels. Such normalization processes have since then been replicated in numerous studies (Ainsworth, 1975; Dechovitz, 1977; Nearey, 1989; Johnson, 1990) and support a perspective on speech perception as a context- and talker-contingent process (Dahan et al , 2008; Goldinger, 1998; Nygaard et al ., 1994; Sjerps et al , 2013). …”