Publications

Abstract

Across many languages from unrelated families, spoken-word recognition is subject to a constraint whereby potential word candidates must contain a vowel. This constraint minimizes competition from embedded words (e.g., in English, disfavoring win in twin because t cannot be a word). However, the constraint would be counter-productive in certain languages that allow stand-alone vowelless open-class words. One such language is Berber (where t is indeed a word). Berber listeners here detected words affixed to nonsense contexts with or without vowels. Length effects seen in other languages replicated in Berber, but in contrast to prior findings, word detection was not hindered by vowelless contexts. When words can be vowelless, otherwise universal constraints disfavoring vowelless words do not feature in spoken-word recognition.

Abstract

The present study used electrophysiological and behavioral measures to investigate the perception of an English stop consonant contrast by native English listeners and by native Dutch listeners who were highly proficient in English. A /ba/-/pa/ continuum was created from a naturally produced /pa/ token by removing successive periods of aspiration, thus reducing the voice onset time. Although aspiration is a relevant cue for distinguishing voiced and unvoiced labial stop consonants (/b/ and /p/) in English, prevoicing is the primary cue used to distinguish between these categories in Dutch. In the electrophysiological experiment, participants listened to oddball sequences containing the standard /pa/ stimulus and one of three deviant stimuli while the mismatch-negativity response was measured. Participants then completed an identification task on the same stimuli. The results showed that native English participants were more sensitive to reductions in aspiration than native Dutch participants, as indicated by shifts in the category boundary, by differing within-group patterns of mismatch-negativity responses, and by larger mean evoked potential amplitudes in the native English group for two of the three deviant stimuli. This between-group difference in the sensorineural processing of aspiration cues indicates that native language experience alters the way in which the acoustic features of speech are processed in the auditory brain, even following extensive second-language training.

Abstract

How do Dutch and Korean listeners use acoustic–phonetic information when learning words in an artificial language? Dutch has a voiceless ‘unaspirated’ stop, produced with shortened Voice Onset Time (VOT) in prosodic strengthening environments (e.g., in domain-initial position and under prominence), enhancing the feature {−spread glottis}; Korean has a voiceless ‘aspirated’ stop produced with lengthened VOT in similar environments, enhancing the feature {+spread glottis}. Given this cross-linguistic difference, two competing hypotheses were tested. The phonological-superiority hypothesis predicts that Dutch and Korean listeners should utilize shortened and lengthened VOTs, respectively, as cues in artificial-language segmentation. The phonetic-superiority hypothesis predicts that both groups should take advantage of the phonetic richness of longer VOTs (i.e., their enhanced auditory–perceptual robustness). Dutch and Korean listeners learned the words of an artificial language better when word-initial stops had longer VOTs than when they had shorter VOTs. It appears that language-specific phonological knowledge can be overridden by phonetic richness in processing an unfamiliar language. Listeners nonetheless performed better when the stimuli were based on the speech of their native languages, suggesting that the use of richer phonetic information was modulated by listeners' familiarity with the stimuli.

Abstract

An eye-tracking experiment examined contextual flexibility in speech processing in response to distortions in spoken input. Dutch participants heard Dutch sentences containing critical words and saw four-picture displays. The name of one picture either had the same onset phonemes as the critical word or had a different first phoneme and rhymed. Participants fixated onset-overlap more than rhyme-overlap pictures, but this tendency varied with speech quality. Relative to a baseline with noise-free sentences, participants looked less at onset-overlap and more at rhyme-overlap pictures when phonemes in the sentences (but not in the critical words) were replaced by noises like those heard on a badly-tuned AM radio. The position of the noises (word-initial or word-medial) had no effect. Noises elsewhere in the sentences apparently made evidence about the critical word less reliable: Listeners became less confident of having heard the onset-overlap name but also less sure of having not heard the rhyme-overlap name. The same acoustic information has different effects on spoken-word recognition as the probability of distortion changes.

Abstract

Children hear new words from many different talkers; to learn words most efficiently, they should be able to represent them independently of talker-specific pronunciation detail. However, do children know what the component sounds of words should be, and can they use that knowledge to deal with different talkers' phonetic realizations? Experiment 1 replicated prior studies on lexically guided retuning of speech perception in adults, with a picture-verification methodology suitable for children. One participant group heard an ambiguous fricative ([s/f]) replacing /f/ (e.g., in words like giraffe); another group heard [s/f] replacing /s/ (e.g., in platypus). The first group subsequently identified more tokens on a Simpie-[s/f]impie-Fimpie toy-name continuum as Fimpie. Experiments 2 and 3 found equivalent lexically guided retuning effects in 12- and 6-year-olds. Children aged 6 have all that is needed for adjusting to talker variation in speech: detailed and abstract phonological representations and the ability to apply them during spoken-word recognition.

Abstract

Two eye-tracking experiments tested whether native listeners can adapt
to reductions in casual Dutch speech. Listeners were exposed to segmental
([b] > [m]), syllabic (full-vowel-deletion), or no reductions. In a subsequent
test phase, all three listener groups were tested on how efficiently they could
recognize both types of reduced words. In the first Experiment’s exposure
phase, the (un)reduced target words were predictable. The segmental reductions
were completely consistent (i.e., involved the same input sequences).
Learning about them was found to be pattern-specific and generalized in the
test phase to new reduced /b/-words. The syllabic reductions were not consistent
(i.e., involved variable input sequences). Learning about them was
weak and not pattern-specific. Experiment 2 examined effects of word repetition
and predictability. The (un-)reduced test words appeared in the exposure
phase and were not predictable. There was no evidence of learning for
the segmental reductions, probably because they were not predictable during
exposure. But there was word-specific learning for the vowel-deleted words.
The results suggest that learning about reductions is pattern-specific and
generalizes to new words if the input is consistent and predictable. With
variable input, there is more likely to be adaptation to a general speaking
style and word-specific learning.

Abstract

Listeners perceive speech sounds relative to context. Contextual influences might differ over hemispheres if different types of auditory processing are lateralized. Hemispheric differences in contextual influences on vowel perception were investigated by presenting speech targets and both speech and non-speech contexts to listeners’ right or left ears (contexts and targets either to the same or to opposite ears). Listeners performed a discrimination task. Vowel perception was influenced by acoustic properties of the context signals. The strength of this influence depended on laterality of target presentation, and on the speech/non-speech status of the context signal. We conclude that contrastive contextual influences on vowel perception are stronger when targets are processed predominately by the right hemisphere. In the left hemisphere, contrastive effects are smaller and largely restricted to speech contexts.

Abstract

When perceiving vowels, listeners adjust to speaker-specific vocal-tract characteristics (such as F1) through "extrinsic vowel normalization". This effect is observed as a shift in the location of categorization boundaries of vowel continua. Similar effects have been found with non-speech. Non-speech materials, however, have consistently led to smaller effect-sizes, perhaps because of a lack of attention to non-speech. The present study investigated this possibility. Non-speech materials that had previously been shown to elicit reduced normalization effects were tested again, with the addition of an attention manipulation. The results show that increased attention does not lead to increased normalization effects, suggesting that vowel normalization is mainly determined by bottom-up signal characteristics.

Abstract

In two eye-tracking experiments in Italian, we investigated how acoustic information and stored knowledge about lexical stress are used during the recognition of tri-syllabic spoken words. Experiment 1 showed that Italians use acoustic cues to a word’s stress pattern rapidly in word recognition, but only for words with antepenultimate stress. Words with penultimate stress – the most common pattern – appeared to be recognized by default. In Experiment 2, listeners had to learn new words from which some stress cues had been removed, and then recognize reduced- and full-cue versions of those words. The acoustic manipulation affected recognition only of newly-learnt words with antepenultimate stress: Full-cue versions, even though they were never heard during training, were recognized earlier than reduced-cue versions. Newly-learnt words with penultimate stress were recognized earlier overall, but recognition of the two versions of these words did not differ. Abstract knowledge (i.e., knowledge generalized over the lexicon) about lexical stress – which pattern is the default and which cues signal the non-default pattern – appears to be used during the recognition of known and newly-learnt Italian words.

Abstract

This paper presents a corpus study that investigates the co-occurrence of reduced word forms in natural speech. We extracted Dutch past participles from three different speech registers and investigated the influence of several predictor variables on the presence and duration of schwas in prefixes and /t/s in suffixes. Our results suggest that reduced word forms tend to co-occur even if we partial out the effect of speech rate. The implications of our findings for episodic and abstractionist models of lexical representation are discussed.

Abstract

Information in speech does not unfold discretely over time; perceptual cues are gradient and overlapped. However, this varies greatly across segments and environments: listeners cannot identify the affricate in /ptS/ until the frication, but information about the vowel in /li/ begins early. Unlike most prior studies, which have concentrated on subsets of language sounds, this study tests perception of every English segment in every phonetic environment, sampling perceptual identification at six points in time (13,470 stimuli/listener; 20 listeners). Results show that information about consonants after another segment is most localized for affricates (almost entirely in the release), and most gradual for voiced stops. In comparison to stressed vowels, unstressed vowels have less information spreading to
neighboring segments and are less well identified. Indeed, many vowels,
especially lax ones, are poorly identified even by the end of the following segment. This may partly reflect listeners’ familiarity with English vowels’ dialectal variability. Diphthongs and diphthongal tense vowels show the most sudden improvement in identification, similar to affricates among the consonants, suggesting that information about segments defined by acoustic change is highly localized. This large dataset provides insights into speech perception and data for probabilistic modeling of spoken word recognition.

Abstract

This study aimed first to determine whether there is a delay associated with processing words in an unfamiliar regional accent compared to words in a familiar regional accent, and second to establish whether short-term exposure to an unfamiliar accent affects the speed and accuracy of comprehension of words spoken in that accent. Listeners performed an animacy decision task for words spoken in their own and in an unfamiliar accent. Next, they were exposed to approximately 20 minutes of speech in one of these two accents. After exposure, they repeated the animacy decision task. Results showed a considerable delay in word processing for the unfamiliar accent, but no effect of short-term exposure.

Abstract

We present results from an experiment which shows that voice perception is influenced by the phonetic content of speech. Dutch listeners were presented with thirteen speakers pronouncing CVC words with systematically varying segmental content, and they had to discriminate the speakers’ voices. Results show that certain segments help listeners discriminate voices more than other segments do. Voice information can be extracted from every segmental position of a monosyllabic word and is processed rapidly. We also show that although relative discriminability within a closed set of voices appears to be a stable property of a voice, it is also influenced by segmental cues – that is, perceived uniqueness of a voice depends on what that voice says.

Abstract

We explore the role of the acoustic consequences of domain-initial strengthening in spoken-word recognition. In two cross-modal identity-priming experiments, listeners heard sentences and made lexical decisions to visual targets, presented at the onset of the second word in two-word sequences containing lexical ambiguities (e.g., bus tickets, with the competitor bust). These sequences contained Intonational Phrase (IP) or Prosodic Word (Wd) boundaries, and the second word's initial Consonant and Vowel (CV, e.g., [tI]) was spliced from another token of the sequence in IP- or Wd-initial position. Acoustic analyses showed that IP-initial consonants were articulated more strongly than Wd-initial consonants. In Experiment 1, related targets were post-boundary words (e.g., tickets). No strengthening effect was observed (i.e., identity priming effects did not vary across splicing conditions). In Experiment 2, related targets were pre-boundary words (e.g., bus). There was a strengthening effect (stronger priming when the post-boundary CVs were spliced from IP-initial than from Wd-initial position), but only in Wd-boundary contexts. These were the conditions where phonetic detail associated with domain-initial strengthening could assist listeners most in lexical disambiguation. We discuss how speakers may strengthen domain-initial segments during production and how listeners may use the resulting acoustic correlates of prosodic strengthening during word recognition.

Abstract

Experiments 1 and 2 examined the time-course of retrieval of phonological, visual-shape and semantic knowledge as Dutch participants listened to sentences and looked at displays of four pictures. Given a sentence with beker, `beaker', for example, the display contained phonological (a beaver, bever), shape (a bobbin, klos), and semantic (a fork, vork) competitors. When the display appeared at sentence onset, fixations to phonological competitors preceded fixations to shape and semantic competitors. When display onset was 200 ms before (e.g.) beker, fixations were directed to shape and then semantic competitors, but not phonological competitors. In Experiments 3 and 4, displays contained the printed names of the previously-pictured entities; only phonological competitors were fixated preferentially. These findings suggest that retrieval of phonological, shape and semantic knowledge in the spoken-word and picture-recognition systems is cascaded, and that visual attention shifts are co-determined by the time-course of retrieval of all three knowledge types and by the nature of the information in the visual environment.

Abstract

Listeners use lexical knowledge to adjust their prelexical representations of speech sounds in response to the idiosyncratic pronunciations of particular speakers. We used an exposure-test paradigm to investigate whether this type of perceptual learning transfers across syllabic positions. No significant learning effect was found in Experiment 1, where exposure sounds were onsets and test sounds were codas. Experiments 2-4 showed that there was no learning even when both exposure and test sounds were onsets. But a trend was found when exposure sounds were codas and test sounds were onsets (Experiment 5). This trend was smaller than the robust effect previously found for the coda-to-coda case. These findings suggest that knowledge about idiosyncratic pronunciations may be position specific: Knowledge about how a speaker produces sounds in one position, if it can be acquired at all, influences perception of sounds in that position more strongly than of sounds in another position.

Abstract

Words repeated in the same voice are better recognized than when they are repeated in a different voice. Such findings have been taken as evidence for the storage of talker-specific lexical episodes. But results on perceptual learning suggest that talker-specific adjustments concern sublexical representations. This study thus investigates whether voice-specific repetition effects in auditory lexical decision are lexical or sublexical. The same critical set of items in Block 2 were, depending on materials in Block 1, either same-voice or different-voice word repetitions, new words comprising re-orderings of phonemes used in the same voice in Block 1, or new words with previously unused phonemes. Results show a benefit for words repeated by the same talker, and a smaller benefit for words consisting of phonemes repeated by the same talker. Talker-specific information thus appears to influence word recognition at multiple representational levels.

Abstract

Listeners use suprasegmental auditory lexical stress information to resolve the competition words engage in during spoken-word recognition. The present study investigated whether (a) visual speech provides lexical stress information, and, more importantly, (b) whether this visual lexical stress information is used to resolve lexical competition. Dutch word pairs that differ in the lexical stress realization of their first two syllables, but not segmentally (e.g., 'OCtopus' and 'okTOber'; capitals marking primary stress) served as auditory-only, visual-only, and audiovisual speech primes. These primes either matched (e.g., 'OCto-'), mismatched (e.g., 'okTO-'), or were unrelated to (e.g., 'maCHI-') a subsequent printed target (octopus), which participants had to make a lexical decision to. To the degree that visual speech contains lexical stress information, lexical decisions to printed targets should be modulated through the addition of visual speech. Results show, however, no evidence for a role of visual lexical stress information in audiovisual spoken-word recognition.

Abstract

Eye movements of Dutch participants were tracked as they looked at arrays of four words on a computer screen and followed spoken instructions (e.g., "Klik op het woord buffel": Click on the word buffalo). The arrays included the target (e.g., buffel), a phonological competitor (e.g., buffer, buffer), and two unrelated distractors. Targets were monosyllabic or bisyllabic, and competitors mismatched targets only on either their onset or offset phoneme and only by one distinctive feature. Participants looked at competitors more than at distractors, but this effect was much stronger for offset-mismatch than onset-mismatch competitors. Fixations to competitors started to decrease as soon as phonetic evidence disfavouring those competitors could influence behaviour. These results confirm that listeners continuously update their interpretation of words as the evidence in the speech signal unfolds and hence establish the viability of the methodology of using eye movements to arrays of printed words to track spoken-word recognition.

Abstract

This chapter is a review of the literature in experimental psycholinguistics on spoken word recognition. It is organized around eight questions. 1. Why are psycholinguists interested in spoken word recognition? 2. What information in the speech signal is used in word recognition? 3. Where are the words in the continuous speech stream? 4. Which words did the speaker intend? 5. When, as the speech signal unfolds over time, are the phonological forms of words recognized? 6. How are words recognized? 7. Whither spoken word recognition? 8. Who are the researchers in the field?

Abstract

We investigated perception of words with reduced word-final /t/ using an adapted eyetracking paradigm. Dutch listeners followed spoken instructions to click on printed words which were accompanied on a computer screen by simple shapes (e.g., a circle). Targets were either above or next to their shapes, and the shapes uniquely identified the targets when the spoken forms were ambiguous between words with or without final /t/ (e.g., bult, bump, vs. bul, diploma). Analysis of listeners’ eye-movements revealed, in contrast to earlier results, that listeners use the following segmental context when compensating for /t/-reduction. Reflecting that /t/-reduction is more likely to occur before bilabials, listeners were more likely to look at the /t/-final words if the next word’s first segment was bilabial. This result supports models of speech perception in which prelexical phonological processes use segmental context to modulate word recognition.

Abstract

Listeners can make perceptual adjustments to phoneme categories in response to a talker who consistently produces a specific phoneme ambiguously. We investigate here whether this type of perceptual learning is also used to adapt to regional accent differences. Listeners were exposed to words produced by a Flemish talker whose realization of [x℄or [h℄ was ambiguous (producing [x℄like [h℄is a property of the West-Flanders regional accent). Before and after exposure they categorized a [x℄-[h℄continuum. For both Dutch and Flemish listeners there was no shift of the categorization boundary after exposure to ambiguous sounds in [x℄- or [h℄-biasing contexts. The absence of a lexically-driven learning effect for this contrast may be because [h℄is strongly influenced by coarticulation. As is not stable across contexts, it may be futile to adapt its representation when new realizations are heard

Abstract

Prosodic influences on phonetic realizations of four Dutch consonants (/t d s z/) were examined. Sentences were constructed containing these consonants in word-initial position; the factors lexical stress, phrasal accent and prosodic boundary were manipulated between sentences. Eleven Dutch speakers read these sentences aloud. The patterns found in acoustic measurements of these utterances (e.g., voice onset time (VOT), consonant duration, voicing during closure, spectral center of gravity, burst energy) indicate that the low-level phonetic implementation of all four consonants is modulated by prosodic structure. Boundary effects on domain-initial segments were observed in stressed and unstressed syllables, extending previous findings which have been on stressed syllables alone. Three aspects of the data are highlighted. First, shorter VOTs were found for /t/ in prosodically stronger locations (stressed, accented and domain-initial), as opposed to longer VOTs in these positions in English. This suggests that prosodically driven phonetic realization is bounded by language-specific constraints on how phonetic features are specified with phonetic content: Shortened VOT in Dutch reflects enhancement of the phonetic feature {−spread glottis}, while lengthened VOT in English reflects enhancement of {+spread glottis}. Prosodic strengthening therefore appears to operate primarily at the phonetic level, such that prosodically driven enhancement of phonological contrast is determined by phonetic implementation of these (language-specific) phonetic features. Second, an accent effect was observed in stressed and unstressed syllables, and was independent of prosodic boundary size. The domain of accentuation in Dutch is thus larger than the foot. Third, within a prosodic category consisting of those utterances with a boundary tone but no pause, tokens with syntactically defined Phonological Phrase boundaries could be differentiated from the other tokens. This syntactic influence on prosodic phrasing implies the existence of an intermediate-level phrase in the prosodic hierarchy of Dutch.

Abstract

We conducted four experiments to investigate the specificity of perceptual adjustments made to unusual speech sounds. Dutch listeners heard a female talker produce an ambiguous fricative [?] (between [f] and [s]) in [f]- or [s]-biased lexical contexts. Listeners with [f]-biased exposure (e.g., [witlo?]; from witlof, “chicory”; witlos is meaningless) subsequently categorized more sounds on an [εf]–[εs] continuum as [f] than did listeners with [s]-biased exposure. This occurred when the continuum was based on the exposure talker's speech (Experiment 1), and when the same test fricatives appeared after vowels spoken by novel female and male talkers (Experiments 1 and 2). When the continuum was made entirely from a novel talker's speech, there was no exposure effect (Experiment 3) unless fricatives from that talker had been spliced into the exposure talker's speech during exposure (Experiment 4). We conclude that perceptual learning about idiosyncratic speech is applied at a segmental level and is, under these exposure conditions, talker specific.

Abstract

In a phonological priming experiment using spoken Dutch words, Dutch listeners were taught varying expectancies and relatedness relations about the phonological form of target words, given particular primes. They learned to expect that, after a particular prime, if the target was a word, it would be from a specific phonological category. The expectancy either involved phonological overlap (e.g., honk-vonk, “base-spark”; expected related) or did not (e.g., nest-galm, “nest-boom”; expected unrelated, where the learned expectation after hearing nest was a word rhyming in -alm). Targets were occasionally inconsistent with expectations. In these inconsistent expectancy trials, targets were either unrelated (e.g., honk-mest, “base-manure”; unexpected unrelated), where the listener was expecting a related target, or related (e.g., nest-pest, “nest-plague”; unexpected related), where the listener was expecting an unrelated target. Participant expectations and phonological relatedness were thus manipulated factorially for three types of phonological overlap (rhyme, one onset phoneme, and three onset phonemes) at three interstimulus intervals (ISIs; 50, 500, and 2,000 msec). Lexical decisions to targets revealed evidence of expectancy-based strategies for all three types of overlap (e.g., faster responses to expected than to unexpected targets, irrespective of phonological relatedness) and evidence of automatic phonological processes, but only for the rhyme and three-phoneme onset overlap conditions and, most strongly, at the shortest ISI (e.g., faster responses to related than to unrelated targets, irrespective of expectations). Although phonological priming thus has both automatic and strategic components, it is possible to cleave them apart.

Abstract

Although researchers studying human speech recognition (HSR) and automatic speech recognition (ASR) share a common interest in how information processing systems (human or machine) recognize spoken language, there is little communication between the two disciplines. We suggest that this lack of communication follows largely from the fact that research in these related fields has focused on the mechanics of how speech can be recognized. In Marr's (1982) terms, emphasis has been on the algorithmic and implementational levels rather than on the computational level. In this article, we provide a computational-level analysis of the task of speech recognition, which reveals the close parallels between research concerned with HSR and ASR. We illustrate this relation by presenting a new computational model of human spoken-word recognition, built using techniques from the field of ASR that, in contrast to current existing models of HSR, recognizes words from real speech input.

Abstract

We report detailed analyses of a very large database on timing of speech perception collected by Smits et al. (Smits, R., Warner, N., McQueen, J.M., Cutler, A., 2003. Unfolding of phonetic information over time: A database of Dutch diphone perception. J. Acoust. Soc. Am. 113, 563–574). Eighteen listeners heard all possible diphones of Dutch, gated in portions of varying size and presented without background noise. The present report analyzes listeners’ responses across gates in terms of phonological features (voicing, place, and manner for consonants; height, backness, and length for vowels). The resulting patterns for feature perception differ from patterns reported when speech is presented in noise. The data are also analyzed for effects of stress and of phonological context (neighboring vowel vs. consonant); effects of these factors are observed to be surprisingly limited. Finally, statistical effects, such as overall phoneme frequency and transitional probabilities, along with response biases, are examined; these too exercise only limited effects on response patterns. The results suggest highly accurate speech perception on the basis of acoustic information alone.