Categorization and discrimination of nonspeech sounds: Differences between steady-state and rapidly-changing acoustic cues

Abstract

Different patterns of performance across vowels and consonants in tests of categorization and discrimination indicate that vowels tend to be perceived more continuously, or less categorically, than consonants. The present experiments examined whether analogous differences in perception would arise in nonspeech sounds that share critical transient acoustic cues of consonants and steady-state spectral cues of simplified synthetic vowels. Listeners were trained to categorize novel nonspeech sounds varying along a continuum defined by a steady-state cue, a rapidly-changing cue, or both cues. Listeners' categorization of stimuli varying on the rapidly changing cue showed a sharp category boundary and posttraining discrimination was well predicted from the assumption of categorical perception. Listeners more accurately discriminated but less accurately categorized steady-state nonspeech stimuli. When listeners categorized stimuli defined by both rapidly-changing and steady-state cues, discrimination performance was accurate and the categorization function exhibited a sharp boundary. These data are similar to those found in experiments with dynamic vowels, which are defined by both steady-state and rapidly-changing acoustic cues. A general account for the speech and nonspeech patterns is proposed based on the supposition that the perceptual trace of rapidly-changing sounds decays faster than the trace of steady-state sounds.

A unified account of categorical effects in phonetic perception

“…This parallels ideas from Pisoni (1975), who considers two modes of perception: acoustic and phonetic. If static spectral cues trigger acoustic processing and dynamic spectral and static temporal cues trigger phonetic processing, then we get a pattern that corresponds closely to our findings and those of Mirman et al (2004). The match is not perfect: although fricative perception is typically sensitive to dynamic cues, the Lago et al (2015) experiment modeled here used only static spectral cues to fricative identity.…”

“…Fricatives have the widest array of critical cues (see (McMurray and Jongman, 2011) for an overview), including dynamic spectral cues in the form of locus equations (Jongman et al 2000) and changes in moments (Forrest et al 1988), static temporal noise duration (Strevens, 1960;Jassem, 1962), and static spectral cues including F2 onset frequency, spectral moments, and spectral peak location (Jongman et al 2000). Mirman et al (2004) showed that the type of acoustic cue used to distinguish a contrast affects listeners' discrimination behavior, using evidence from listeners' identification and discrimination of non-speech stimuli. They trained listeners to categorize stimuli along a continuum that was differentiated either by steady state spectral cues (similar to isolated vowels), or by rapidly changing acoustic information (similar to stop consonants).…”

“…The model makes a prediction that if listeners are trained on data in which the meaningful and unexplained noise variance are varied, then they should eventually exhibit the predicted categorical perception effects. One example of such an experiment is Mirman et al (2004), who showed that familiarization with varying acoustic cues correlated with identification and discrimination behavior. Another way to see the effect of varying the variances on τ is to manipulate the presence of noise variance in the signal during testing.…”

“…Researchers have proposed a number of qualitative explanations for these differences. For example, the differences have been claimed to stem from the way each type of sound is stored in memory (Pisoni, 1973), to be related to innate auditory discontinuities that could influence stop consonant perception (Pisoni, 1977;Eimas et al, 1971), and to result from different processing mechanisms for steady state and rapidly changing spectral cues (Mirman et al, 2004). However, qualitatively the effects are very similar between consonants and vowels, with a sharp shift in the identification function and a peak in discrimination near the category boundary.…”

Impaired categorical perception of lexical tones in Mandarin-speaking congenital amusics

“…It has been suggested that 4-to 6-month-old infants discriminate the sound categories in both Mandarin tones and music (Lynch & Eilers, 1992;Mattock, Molnar, Polka & Burnham, 2008), but if they are not exposed to Mandarin language, or a given musical context, these abilities are lost by 9 months for speech tones (Mattock et al, 2008), and by adulthood for music (Lynch & Eilers, 1992). Furthermore, listeners with music training outperformed those without music training in categorical perception studies involving Mandarin tones (Wu & Lin, 2008) or nonspeech frequency continua (e.g., Howard, Rosen & Broad, 1992;Mirman, Holt & McClelland, 2004;Pisoni, 1977). These studies indicated that experience with tonal language and music may affect the categorical perception of speech and nonlinguistic frequency continua.…”