Perceptual integration and differentiation of spectral cues for intervocalic stop consonants

The place of articulation of intervocalic stop consonants is conveyed by temporally distributed spectral information, viz, the formant transitions preceding and following the silent closure interval (VC and CV transitions). Experiment 1 shows that more than 200 msec of silent closure is needed to hear VC and CV formant transitions as separate phonemic events (geminate stops). As closure duration is reduced, these cues are integrated into a single phonemic percept, and the VC transitions become increasingly redundant (Experiments 2 and 3). VC and CV transitions conveying different places of articulation, on the other hand, are heard as separate phonemes at closure durations as short as 100 msec. If closure duration is further reduced, a single stop is heard whose place of articulation corresponds to the CV transitions (Experiment 3). Even in the absence of CV transitions, VC transitions carry little perceptual weight at very short closure durations (Experiment 4). Despite their apparent redundancy, however, the VC transitions exert a positive bias on the perception of CV transitions at very short closure durations. At closure durations beyond 100 msec, on the other hand, VC and CV transitions interact contrastively in perception and tend to be heard as different phonemes (Experiments 5 and 6). The results of these experiments suggest two different processes of temporal integration in phonetic perception, one taking place at a precategorical level, the other combining identical phoneme categories within a certain time span.     

Bidirectional contrast effects in the perception of VC-CV sequences

The perceived places of articulation of two successive stop consonants are not independent: Given some ambiguity in the formant transition cues and a closure duration between 100 and 200 msec, contrastive perceptual interactions in both directions have been observed in identification tasks. Retroactive contrast declines as the closure interval is lengthened and is strongly influenced by the range of closure durations employed, whereas proactive contrast appears to be less sensitive to these factors (Experiment 1). Reduced contrast and no effects of closure duration are obtained in a discrimination task with selective attention to one stimulus portion; this suggests that the effects in identification arise largely at a higher level of (phonetic) perception (Experiment 2). The contrast effects do not seem to represent a perceptual compensation for coarticulatory dependencies between stops produced in sequence, for there appears to be little coarticulation as far as place of articulation is concerned (Experiment 3). The most plausible hypothesis is that the presumed contrast effects do not result from any direct interaction of spectral cues across the closure interval but are due to perceptual information conveyed by the closure itself: Closure durations of 100–200 msec happen to be most appropriate for sequences of two nonhomorganic stops. Here, it seems, is another case in which listeners’ tacit knowledge of canonical speech patterns determines perception.     

The relative weighting of acoustic properties in the perception of [s]+stop clusters by children and adults

We examined the perceptual weighting by children and adults of the acoustic properties specifying complete closure of the vocal tract following a syllable-initial [s]. Experiment 1 was a novel manipulation of previously examined acoustic properties (duration of a silent gap and first formant transition) and showed that children weight the first formant transition more than adults. Experiment 2, an acoustic analysis of naturally producedsay andstay, revealed that, contrary to expectations, a burst can be present instay and that first formant transitions do not necessarily distinguishsay andstay in natural tokens. Experiment 3 manipulated natural speech portions to create stimuli that varied primarily in the duration of the silent gap and in the presence or absence of a stop burst, and showed that children weight these stop bursts less than adults. Taken together, the perception experiments support claims that children integrate multiple acoustic properties as adults do, but that they weight dynamic properties of the signal more than adults and weight static properties less.     

The role of second formant transitions in the stop-semivowel distinction

An experiment was conducted which assessed the relative contributions of three acoustic cues to the distinction between stop consonant and semivowel in syllable initial position. Subjects identified three series of syllables which varied perceptually from [ba] to [wa]. The stimuli differed only in the extent, duration, and rate of the second formant transition. In each series, one of the variables remained constant while the other two changed. Obtained identification ratings were plotted as a function of each variable. The results indicated that second formant transition duration and extent contribute significantly to perception. Short second formant transition extents and durations signal stops, while long second formant transition extents and durations signal semivowels. It was found that second formant transition rate did not contribute significantly to this distinction. Any particular rate could signal either a stop or semivowel. These results are interpreted as arguing against models that incorporate transition rate as a cue to phonetic distinctions. In addition, these results are related to a previous selective adaptation experiment. It is shown that the “phonetic” interpretation of the obtained adaptation results was not justified.     

Auditory and phonetic processes in speech perception: Evidence from a dichotic study

The distinction between auditory and phonetic processes in speech perception was used in the design and analysis of an experiment. Earlier studies had shown that dichotically presented stop consonants are more often identified correctly when they share place of production (e.g., /ba-pa/) or voicing (e.g., /ba-da/) than when neither feature is shared (e.g., /ba-ta/). The present experiment was intended to determine whether the effect has an auditory or a phonetic basis. Increments in performance due to feature-sharing were compared for synthetic stop-vowel syllables in which formant transitions were the sole cues to place of production under two experimental conditions: (1) when the vowel was the same for both syllables in a dichotic pair, as in our earlier studies, and (2) when the vowels differed. Since the increment in performance due to sharing place was not diminished when vowels differed (i.e., when formant transitions did not coincide), it was concluded that the effect has a phonetic rather than an auditory basis. Right ear advantages were also measured and were found to interact with both place of production and vowel conditions. Taken together, the two sets of results suggest that inhibition of the ipsilateral signal in the perception of dichotically presented speech occurs during phonetic analysis.     

Rise time and formant transition duration in the discrimination of speech sounds: the Ba-Wa distinction in developmental dyslexia

Across languages, children with developmental dyslexia have a specific difficulty with the neural representation of the sound structure (phonological structure) of speech. One likely cause of their difficulties with phonology is a perceptual difficulty in auditory temporal processing (Tallal, 1980). Tallal (1980) proposed that basic auditory processing of brief, rapidly successive acoustic changes is compromised in dyslexia, thereby affecting phonetic discrimination (e.g. discriminating /b/ from /d/) via impaired discrimination of formant transitions (rapid acoustic changes in frequency and intensity). However, an alternative auditory temporal hypothesis is that the basic auditory processing of the slower amplitude modulation cues in speech is compromised (Goswami et al., 2002). Here, we contrast children's perception of a synthetic speech contrast (ba/wa) when it is based on the speed of the rate of change of frequency information (formant transition duration) versus the speed of the rate of change of amplitude modulation (rise time). We show that children with dyslexia have excellent phonetic discrimination based on formant transition duration, but poor phonetic discrimination based on envelope cues. The results explain why phonetic discrimination may be allophonic in developmental dyslexia (Serniclaes et al., 2004), and suggest new avenues for the remediation of developmental dyslexia.     

Auditory property detectors and processing place features in stop consonants

The effects of selective adaptation on the perception of consonant-vowel (CV) stimuli varying in place of production was studied under two conditions. In the first condition, repeated presentation of a CV syllable produced an adaptation effect resulting in a shift in the locus of the phonetic boundary between [ba] and [da]. This result replicated previously reported findings. However, in the second condition, an adaptation effect was obtained on this same test series when the critical acoustic information (i.e., formant transitions) was present in final position of a VC speech-like syllable. These latter results support an auditory account of selective adaptation based on the spectral similarity of the adapting stimuli and test series rather than a more abstract linguistic account based on phonetic identity.     

Zebra finches and Dutch adults exhibit the same cue weighting bias in vowel perception

Vocal tract resonances, called formants, are the most important parameters in human speech production and perception. They encode linguistic meaning and have been shown to be perceived by a wide range of species. Songbirds are also sensitive to different formant patterns in human speech. They can categorize words differing only in their vowels based on the formant patterns independent of speaker identity in a way comparable to humans. These results indicate that speech perception mechanisms are more similar between songbirds and humans than realized before. One of the major questions regarding formant perception concerns the weighting of different formants in the speech signal (“acoustic cue weighting”) and whether this process is unique to humans. Using an operant Go/NoGo design, we trained zebra finches to discriminate syllables, whose vowels differed in their first three formants. When subsequently tested with novel vowels, similar in either their first formant or their second and third formants to the familiar vowels, similarity in the higher formants was weighted much more strongly than similarity in the lower formant. Thus, zebra finches indeed exhibit a cue weighting bias. Interestingly, we also found that Dutch speakers when tested with the same paradigm exhibit the same cue weighting bias. This, together with earlier findings, supports the hypothesis that human speech evolution might have exploited general properties of the vertebrate auditory system.     

Auditory evoked potential correlates of speech sound discrimination

An electrophysiological correlate of the discrimination of stop consonants drawn from within and across phonetic categories was investigated by an auditory evoked response (AER) technique. Ss were presented a string of stimuli from the phonetic category [ba] (the standard stimulus) and were asked to detect the occurrence of a stimulus from the same phonetic category (within-category shift), or the occurrence of a stimulus from a different phonetic category [pa] (across-category shift). Both the across- and within-category shift stimuli differed equally from the standard stimulus in the time of onset of the first formant and in the amount of aspiration in the second and third formants. The NIP2 response of the AER was larger to the across-category shift than to the within-category shift. The within-category shift did not differ from a no-shift control. These findings suggest (1) that the AER can reflect the relative discriminability of stop consonants drawn from the same or different phonetic categories in a manner similar to other behavioral measures; (2) that the detailed acoustic representation of stop consonants is transformed into a categorized phonetic representation within 200 msec after stimulus onset.     

Stop-consonant recognition: Release bursts and formant transitions as functionally equivalent,context-dependent cues

Three experiments assessed the roles of release bursts and formant transitions as acoustic cues to place of articulation in syllable-initial voiced stop consonants by systematically removing them from American English /b,d,g/, spoken before nine different vowels by two speakers, and by transposing the bursts across all vowels for each class of stop consonant. The results showed that bursts were largely invariant in their effect, but carried significant perceptual weight in only one syllable out of 27 for Speaker 1, in only 13 syllables out of 27 for Speaker 2. Furthermore, bursts and transitions tended to be reciprocally related: Where the perceptual weight of one increased, the weight of the other declined. They were thus shown to be functionally equivalent, context-dependent cues, each contributing to the rapid spectral changes that follow consonantal release. The results are interpreted as pointing to the possible role of the front-cavity resonance in signaling place of articulation.     

Identification reaction times of voiced/voiceless continua: a right-ear advantage for VOT values near the phonetic boundary

We explored the degree to which the duration of acoustic cues contributes to the respective involvement of the two hemispheres in the perception of speech. To this end, we recorded the reaction time needed to identify monaurally presented natural French plosives with varying VOT values. The results show that a right-ear advantage is significant only when the phonetic boundary is close to the release burst, i.e., when the identification of the two successive acoustical events (the onset of voicing and the release from closure) needed to perceive a phoneme as voiced or voiceless requires rapid information processing. These results are consistent with the recent hypothesis that the left hemisphere is superior in the processing of rapidly changing acoustical information.     

The perceptual similarity of mirror-image acoustic patterns in speech

Prevocalic and postvocalic (unreleased) occurrences of a stop consonant differ in acoustic shape, but are not unrelated. In particular, the formant transitions taking place at release of a stop consonant approximately mirror in time the formant transitions occurring during closure, assuming that the vowel is the ~same. Several experiments have been performed using brief two-component tone burst approximations to the second and third formant transitions’ that occur in prevocalic and postvocalic allophones of/b, d, g/in order to determine whether such mirror-image acoustic patterns are perceptually related. Listener judgments of similarity within triads of these stimuli indicate that mirror-image patterrts representing the same place of articulation are less similar to each other than to patterns representing different places of articulation. Implications for the child who is acquiring language of the fact that mirror-image patterns in speech do not have inherent perceptual similarity are discussed.     

Phonetic disintegration in aphasia: Acoustic analysis of spectral characteristics for place of articulation

In this study, we attempted to determine whether phonetic disintegration of speech in Broca's aphasia affects the spectral characteristics of speech sounds as has been shown for the temporal characteristics of speech. To this end, we investigated the production of place of articulation in Broca's aphasics. Acoustic analysis of the spectral characteristics for stop consonants were conducted. Results indicated that the static aspects of speech production were preserved, as Broca's aphasics seemed to be able to reach the articulatory configuration for the appropriate place of articulation. However, the dynamic aspects of speech production seemed to be impaired, as their productions reflected problems with the source characteristics of speech sounds and with the integration of articulatory movements in the vocal tract. Listener perceptions of the aphasics' productions were compared with acoustic analyses for these same productions. The two measures were related; that is, the spectral characteristics of the utterances provided salient cues for place of articulation perception. An analysis of the occurrences of errors along the dimensions of voicing and place showed that aphasics rarely produce utterances containing both voice and place substitutions.     

Perceptual equivalence of two acoustic cues for stop-consonant manner

The first experiment examined the effects of orthogonal variation in two acoustic cues that are both produced by the articulation of a stop consonant. One component of the articulation produces a temporal cue, the other a spectral cue. In a phonetic identification task, these quite different cues were found to engage in a trading relation: One cue could, within limits, substitute for the other. In a second experiment, the perceptual equivalence implied by that trading relation was put to a stricter test. If the cues are truly equivalent in perception, then they should have their effects on the same perceptual dimension. That being so, it should be possible to combine them in such a way that, working at cross purposes in the perceptual domain, they effectively cancel each other. Pairs of patterns so produced should, then, be harder to discriminate than pairs produced by either cue alone, and still harder than those produced when the (same) two cues are combined in the opposite way and so cooperate. That expectation was confirmed. We suggest that the equivalence thus demonstrated comes about because the two cues are processed by a system specialized to take account of their common origin in speech production. So interpreted, the equivalence may be viewed as an instance of distinctively phonetic perception.     

Some experiments on perceptual learning of mirror-image acoustic patterns

It is well known that the formant transitions of stop consonants in CV and VC syllables are roughly the mirror image of each other in time. These formant motions reflect the acoustic correlates of the articulators as they move rapidly into and out of the period of stop closure. Although acoustically different, these formant transitions are correlated perceptually with similar phonetic segments. Earlier research of Klatt and Shattuck (1975) had suggested that mirror image acoustic patterns resembling formant transitions were not perceived as similar. However, mirror image patterns could still have some underlying similarity which might facilitate learning, recognition, and the establishment of perceptual constancy of phonetic segments across syllable positions. This paper reports the results of four experiments designed to study the perceptual similarity of mirror-image acoustic patterns resembling the formant transitions and steady-state segments of the CV and VC syllables /ba/, /da/, /ab/, and /ad/. Using a perceptual learning paradigm, we found that subjects could learn to assign mirror-image acoustic patterns to arbitrary response categories more consistently than they could do so with similar arrangements of the same patterns based on spectrotemporal commonalities. Subjects respond not only to the individual components or dimensions of these acoustic patterns, but also process entire patterns and make use of the patterns’ internal organization in learning to categorize them consistently according to different classification rules.     

Some effects of acoustic attributes of speech on the processing of phonetic feature information

How do acoustic attributes of the speech signal contribute to feature-processing interactions that occur in phonetic classification? In a series of five experiments addressed to this question, listeners performed speeded classification tasks that explicitly required a phonetic decision for each response. Stimuli were natural consonant-vowel syllables differing by multiple phonetic features, although classification responses were based on a single target feature. In control tasks, no variations in nontarget features occurred, whereas in orthogonal tasks nonrelevant feature variations occurred but had to be ignored. Comparison of classification times demonstrated that feature information may either be processed separately as independent cues for each feature or as a single integral segment that jointly specifies several features. The observed form on processing depended on the acoustic manifestations of feature variation in the signal. Stop-consonant place of articulation and voicing cues, conveyed independently by the pattern and excitation source of the initial formant transitions, may be processed separately. However, information for consonant place of articulation and vowel quality, features that interactively affect the shape of initial formant transitions, are processed as an integral segment. Articulatory correlates of each type of processing are discussed in terms of the distinction between source features that vary discretely in speech production and resonance features that can change smoothly and continuously. Implications for perceptual models that include initial segmentation of an input utterance into a phonetic feature representation are also considered.     

Talker continuity and the use of rate information during phonetic perception

Research has shown that speaking rate provides an important context for the perception of certain acoustic properties of speech. For example, syllable duration, which varies as a function of speaking rate, has been shown to influence the perception of voice onset time (VOT) for syllableinitial stop consonants. The purpose of the present experiments was to examine the influence of syllable duration when the initial portion of the syllable was produced by one talker and the remainder of the syllable was produced by a different talker. A short-duration and a long-duration /bi/-/pi/ continuum were synthesized with pitch and formant values appropriate to a female talker. When presented to listeners for identification, these stimuli demonstrated the typical effect of syllable duration on the voicing boundary: a shorter VOT boundary for the short stimuli than for the long stimuli. An /i/ vowel, synthesized with pitch and formant values appropriate to a male talker, was added to the end of each of the short tokens, producing a new hybrid continuum. Although the overall syllable duration of the hybrid stimuli equaled the original long stimuli, they produced a VOT boundary similar to that for the short stimuli. In a second experiment, two new /i/ vowels were synthesized. One had a pitch appropriate to a female talker with formant values appropriate to a male talker; the other had a pitch appropriate to a male talker and formants appropriate to a female talker. These vowels were used to create two new hybrid continua. In a third experiment, new hybrid continua were created by using more extreme male formant values. The results of both experiments demonstrated that the hybrid tokens with a change in pitch acted like the short stimuli, whereas the tokens with a change in formants acted like the long stimuli. A fourth experiment demonstrated that listeners could hear a change in talker with both sets of hybrid tokens. These results indicate that continuity of pitch but not formant structure appears to be the critical factor in the calculation of speaking rate within a syllable.     

Acoustic feature analysis in the perception of voicing contrasts

Various types of acoustic cues have been shown to signal voicing contrasts in initial prestressed stop consonants. Two of the most important of these cues are the time interval between the release burst and the onset of voicing (voice onset time) and the duration of voiced transitions. In the present experiment, a selective adaptation procedure was used to investigate possible feature extraction mechanisms which underlie the perception of voicing. Subjects first identified either of two series of test stimuli, [ba-p^ha) or [ga-k^ha]. Next they listened to repeated presentations of an adapting stimulus and then again identified the original test series. Adapting stimuli having the same value of voice onset time but different voiced transition durations produced differential (and sometimes opposite) effects on the identification of the test stimuli. In every case, the adapting stimulus with the greater duration of voiced transitions led to larger reduction in voiced responses or to a smaller increase in voiceless responses. These results are incompatible with models of voicing perception based strictly on detectors for voice onset time. The results also suggest that the adaptation effect occurs at a point in the system prior to the actual phonetic decision.     

Coherence masking protection in speech sounds: The role of formant synchrony

Three experiments were performed to examine listeners’ thresholds for identifying stimuli whose spectra were modeled after the vowels /i/ and /ε/, with the differences between these stimuli restricted to the frequency of the first formant. The stimuli were presented in a low-pass masking noise that spectrally overlapped the first formant but not the higher formants. Identification thresholds were lower when the higher formants were present than when they were not, even though the first formant contained the only distinctive information for stimulus identification. This indicates that listeners were more sensitive in identifying the first formant energy through its contribution to the vowel than as an independent percept; this effect is given the namecoherence masking protection. The first experiment showed this effect for synthetic vowels in which the distinctive first formant was supported by a series of harmonics that progressed through the higher formants. In the second two experiments, the harmonics in the first formant region were removed, and the first formant was simulated by a narrow band of noise. This was done so that harmonic relations did not provide a basis for grouping the lower formant with the higher formants; coherence masking protection was still observed. However, when the temporal alignment of the onsets and offsets of the higher and lower formants was disrupted, the effect was eliminated, although the stimuli were still perceived as vowels. These results are interpreted as indicating that general principles of auditory grouping that can exploit regularities in temporal patterns cause acoustic energy belonging to a coherent speech sound to stand out in the auditory scene.     

Phoneme perception in a neonate with a left sylvian infarct

We report the case of a neonate tested three weeks after a neonatal left sylvian infarct. We studied her perception of speech and non-speech stimuli with high-density event-related potentials. The results show that she was able to discriminate not only a change of timbre in tones but also a vowel change, and even a place of articulation contrast in stop consonants. Moreover, a discrimination response to stop consonants was observed even when syllables were produced by different speakers. Her intact right hemisphere was thus able to extract relevant phonetic information in spite of irrelevant acoustic variation. These results suggest that both hemispheres contribute to phoneme perception during the first months of life and confirm our previous findings concerning bilateral responses in normal infants.