Sentence Context Effects Research Articles

Assessing perceived listening difficulty using behavioral gaze patterns for audiovisual speech

Listeners alternate their gaze between the talker’s eyes, nose, and mouth during conversation, spending more time gazing directly at the mouth in more difficult listening conditions. Perceived listening difficulty and speech perception performance vary independently, demonstrated with indirect measures of perceived listening difficulty/effort such as pupil dilation and dual task paradigms. This study took a more direct approach, evaluating whether perceived listening difficulty could be assessed by observing a listener’s natural gaze pattern in a challenging speech-perception task. The proportion of time spent gazing at the talker’s mouth was measured for auditory-visual speech perception in conditions that produced equal percentage-correct scores, but where one condition was designed to produce greater perceived difficulty. Experiment 1 compared gaze proportion for a speech-shaped noise versus two same-gender interfering talkers. Experiment 2 examined the effect of sentence context using low and high-probability sentences. The results are discussed in terms of the potential application of the eye-gaze metric to evaluate rehabilitative interventions in cases where subjectively reported benefits do not manifest in improved speech-understanding scores. [The views expressed in this abstract are those of the authors and do not reflect the official policy of the Department of Army/Navy/Air Force, Department of Defense, or U.S. Government.]

Understanding environmental sounds in sentence context

There is debate about how individuals use context to successfully predict and recognize words. One view argues that context supports neural predictions that make use of the speech motor system, whereas other views argue for a sensory or conceptual level of prediction. While environmental sounds can convey clear referential meaning, they are not linguistic signals, and are thus neither produced with the vocal tract nor typically encountered in sentence context. We compared the effect of spoken sentence context on recognition and comprehension of spoken words versus nonspeech, environmental sounds. In Experiment 1, sentence context decreased the amount of signal needed for recognition of spoken words and environmental sounds in similar fashion. In Experiment 2, listeners judged sentence meaning in both high and low contextually constraining sentence frames, when the final word was present or replaced with a matching environmental sound. Results showed that sentence constraint affected decision time similarly for speech and nonspeech, such that high constraint sentences (i.e., frame plus completion) were processed faster than low constraint sentences for speech and nonspeech. Linguistic context facilitates the recognition and understanding of nonspeech sounds in much the same way as for spoken words. This argues against a simple form of a speech-motor explanation of predictive coding in spoken language understanding, and suggests support for conceptual-level predictions.

The sentence superiority effect revisited

A sentence superiority effect was investigated using post-cued word-in-sequence identification with the rapid parallel visual presentation (RPVP) of four horizontally aligned words. The four words were presented for 200ms followed by a post-mask and cue for partial report. They could form a grammatically correct sentence or were formed of the same words in a scrambled agrammatical sequence. Word identification was higher in the syntactically correct sequences, and crucially, this sentence superiority effect did not vary as a function of the target’s position in the sequence. Cloze probability measures for words at the final, arguably most predictable position, revealed overall low values that did not interact with the effects of sentence context, suggesting that these effects were not driven by word predictability. The results point to a level of parallel processing across multiple words that enables rapid extraction of their syntactic categories. These generate a sentence-level representation that constrains the recognition process for individual words, thus facilitating parallel word processing when the sequence is grammatically sound.

The effect of F0 contours on the intelligibility of speech against interfering sounds for Mandarin Chinese

To study the role of F0 contour on speech intelligibility for Mandarin Chinese, speech reception thresholds (SRTs) of mono-tone speech and multi-tone speech were measured both in steady speech-spectrum noise (SSN) and in two-talkers’ speech (TTS). The effect of informational masking (IM) in TTS masking was also examined. The mono-tone speech was produced by mono-tone sentences, which only consisted of syllables with tone 1 or tone 0. All sentences used in this study were grammatically correct but semantically anomalous to avoid the effect of sentence context. The type of TTS maskers was also manipulated as mono-tone or multi-tone. The results revealed: 1) For SSN masking, there was no significant difference between the mono-tone speech and the multi-tone speech (-7.8 dB vs. -7.3 dB on SRTs), indicating a weak effect of altering F0 contour at the sentence level. 2) However, for TTS masking, the intelligibility of multi-tone speech in multi-tone masking was significantly difficult (2 dB increase for SRT) than other conditions, indicating an important role of F0 contour in IM. These results were different from the corresponding reports for English speech, and the characteristics of F0 contour and language were analyzed and discussed. [supported by a NSFC grant 61473008.]

Learning to Read Words on Flashcards: Effects of Sentence Contexts and Word Class in Native and Nonnative English-Speaking Kindergartners

This study investigated three questions: 1) Does clarifying the meanings of words during flashcard reading better support word learning over not providing meanings? 2) Does grammatical word class affect word learning? 3) Do these manipulations affect word learning differently in native and nonnative speakers of English? Native (n=40) and nonnative (n=41) English-speaking kindergartners practiced learning to read nouns and function words on flashcards in a repeated measures, counterbalanced design with random assignment. In one condition words were taught in meaningful sentences. In the other condition words were taught in isolation. Results of ANOVAs revealed that both native and nonnative speakers of English were better able to read words when they were taught in isolation than in sentences, and native speakers were better able to spell the words as well. However, both groups showed better ability to express their meanings when words were read in semantically rich sentences than when read in isolation. Children learned to read, spell, and produce sentences for content words more easily than for function words. Results indicate that the way children practice reading words and the type of words being read influence what they learn about the words.

Top-Down Processes in Simulated Electric-Acoustic Hearing: The Effect of Linguistic Context on Bimodal Benefit for Temporally Interrupted Speech.

Previous studies have documented the benefits of bimodal hearing as compared with a cochlear implant alone, but most have focused on the importance of bottom-up, low-frequency cues. The purpose of the present study was to evaluate the role of top-down processing in bimodal hearing by measuring the effect of sentence context on bimodal benefit for temporally interrupted sentences. It was hypothesized that low-frequency acoustic cues would facilitate the use of contextual information in the interrupted sentences, resulting in greater bimodal benefit for the higher context (CUNY) sentences than for the lower context (IEEE) sentences. Young normal-hearing listeners were tested in simulated bimodal listening conditions in which noise band vocoded sentences were presented to one ear with or without low-pass (LP) filtered speech or LP harmonic complexes (LPHCs) presented to the contralateral ear. Speech recognition scores were measured in three listening conditions: vocoder-alone, vocoder combined with LP speech, and vocoder combined with LPHCs. Temporally interrupted versions of the CUNY and IEEE sentences were used to assess listeners' ability to fill in missing segments of speech by using top-down linguistic processing. Sentences were square-wave gated at a rate of 5 Hz with a 50% duty cycle. Three vocoder channel conditions were tested for each type of sentence (8, 12, and 16 channels for CUNY; 12, 16, and 32 channels for IEEE) and bimodal benefit was compared for similar amounts of spectral degradation (matched-channel comparisons) and similar ranges of baseline performance. Two gain measures, percentage-point gain and normalized gain, were examined. Significant effects of context on bimodal benefit were observed when LP speech was presented to the residual-hearing ear. For the matched-channel comparisons, CUNY sentences showed significantly higher normalized gains than IEEE sentences for both the 12-channel (20 points higher) and 16-channel (18 points higher) conditions. For the individual gain comparisons that used a similar range of baseline performance, CUNY sentences showed bimodal benefits that were significantly higher (7% points, or 15 points normalized gain) than those for IEEE sentences. The bimodal benefits observed here for temporally interrupted speech were considerably smaller than those observed in an earlier study that used continuous speech. Furthermore, unlike previous findings for continuous speech, no bimodal benefit was observed when LPHCs were presented to the LP ear. Findings indicate that linguistic context has a significant influence on bimodal benefit for temporally interrupted speech and support the hypothesis that low-frequency acoustic information presented to the residual-hearing ear facilitates the use of top-down linguistic processing in bimodal hearing. However, bimodal benefit is reduced for temporally interrupted speech as compared with continuous speech, suggesting that listeners' ability to restore missing speech information depends not only on top-down linguistic knowledge but also on the quality of the bottom-up sensory input.

Bilingual non-selective lexical access in sentence contexts: A meta-analytic review

Abstract Research on bilingual sentence processing demonstrates effects of cross-language activation during lexical access. However, there are mixed findings regarding the ability of semantically-constraining sentences to eliminate non-selective effects. In a quantitative meta-analysis the magnitude of cognate facilitation was examined as a function of sentence constraint, task and language of the sentence [native language (L1) versus second (L2)] as moderator variables. Twenty-six studies met criteria for measuring cognate facilitation in low-constraint sentence contexts and 18 experiments for high-constraint sentence contexts. The weighted average effect size for both contexts was significant, but significantly smaller for high-constraint contexts. This provides evidence that semantic information from a sentence constrains cross-language lexical activation. Effect sizes were moderated by task and language of the sentence. Findings are discussed in terms of models of bilingual lexical access and a new framework for understanding the mechanism of sentence context effects is proposed.

Sentential influences on acoustic-phonetic processing: a Granger causality analysis of multimodal imaging data

ABSTRACTSentential context influences the way that listeners identify phonetically ambiguous or perceptual degraded speech sounds. Unfortunately, inherent inferential limitations on the interpretation of behavioural or BOLD imaging results make it unclear whether context influences perceptual processing directly, or acts at a post-perceptual decision stage. In this paper, we use Kalman-filter enabled Granger causation analysis of MRI-constrained MEG/EEG data to distinguish between these possibilities. Using a retrospective probe verification task, we found that sentential context strongly affected the interpretation of words with ambiguous initial voicing (e.g. DUSK-TUSK). This behavioural context effect coincided with increased influence by brain regions associated with lexical representation on regions associated with acoustic-phonetic processing. These results support an interactive view of sentence context effects on speech perception.

Erratum

Hedrick, M. S., Lee, J. Y., Harkrider, A., & von Hapsburg, D. (2014) Effects of sentence context on phonemic categorization of natural and synthetic consonant-vowel tokens. Perceptual & Motor Skills: Perception, 118, 1, 210-224. DOI: 10.2466/24.22.PMS.118k14w6 The second author wishes to add a third footnote to the bottom of page 210: 3Ji Young Lee is currently at the Department of Audiology and Speech-Language Pathology, Catholic University of Daegu, 13-13 Hayang-ro, Hayang-eup, Gyeongsan-si, Gyeongsang-buk-do 712-702 REPUBLIC OF KOREA.

Effects of masker type, sentence context, and listener age on speech recognition performance in 1-back listening tasks.

Studies have shown that older listeners with normal hearing have greater difficulty understanding speech in noisy environments than younger listeners even during simple assessments where listeners respond to auditory stimuli immediately after presentation. Older listeners may have increased difficulty understanding speech in challenging listening situations that require the recall of prior sentences during the presentation of new auditory stimuli. This study compared the performance of older and younger normal-hearing listeners in 0-back trials, which required listeners to respond to the most recent sentence, and 1-back trials, which required the recall of the sentence preceding the most recent. Speech stimuli were high-context and anomalous sentences with four types of maskers. The results show that older listeners have greater difficulty in the 1-back task than younger listeners with all masker types, even when SNR was adjusted to produce 80% correct performance in the 0-back task for both groups. The differences between the groups in the 1-back task may be explained by differences in working memory for the noise and spatially separated speech maskers but not in the conditions with co-located speech maskers, suggesting that older listeners have increased difficulty in memory-intensive speech perception tasks involving high levels of informational masking.

Sentence context effects in bilingual word recognition: Cognate status, sentence language, and semantic constraint

In two lexical decision experiments, we investigated how sentence language affects the bilingual's recognition of target words from the same or a different language. Dutch–English bilinguals read Dutch (L1) or English (L2) sentences, presented word by word, followed by English (Experiment 1) or Dutch (Experiment 2) target words. Targets were Dutch–English cognates or non-cognates in isolation or preceded by sentences providing a high or a low semantic constraint. English cognates were facilitated irrespective of whether they were preceded by high or low constraining English sentences (no language switch) or Dutch sentences (switch). For Dutch cognates, inhibition effects arose in low constraining sentences (irrespective of Dutch or English) and in English (switch) sentences (irrespective of semantic constraint). Thus, under mixed language conditions, sentence constraint modulates target word processing but does not always completely eliminate cross-linguistic effects. The results are interpreted in a BIA+ model that extends monolingual views on sentence comprehension.

Independent and interacting effects of sentential context and phonological neighborhood structure in spoken word production

Models of speech production typically invoke at least two cognitive systems: a structurally static lexicon and a dynamic, compositional system. A speaker must have access to a network of lexical representations that is vested with certain intrinsically associated information (e.g., words' phonological structure), but typical speech involves sequential, online production of words to construct meaningful sentences. Thus, a complete model of speech communication must overlay the lexicon's static architecture with a dynamic, context-driven system. The present study examines effects of these two systems on spoken words' phonetic realizations. Participants produced monosyllabic words beginning with voiceless stop consonants (e.g., coat) that varied in their phonological neighborhood density. Targets were produced aloud after subjects silently read sentence contexts that were either highly predictive (e.g., She sported her stylish new fur...) or relatively neutral (e.g., The artist had trouble drawing the...). R...

Effects of sentence context on phonemic categorization of natural and synthetic consonant-vowel tokens.

The purpose was to assess if phonemic categorization in sentential context is best explained by autonomous feedforward processing or by top-down feedback processing that affects phonemic representation. 11 listeners with normal hearing, ages 20-50 years, were asked to label consonants in /pi/-/ti/ consonant-vowel (CV) stimuli in 9-step continua. One continuum was derived from natural tokens and the other was synthetically generated. The CV stimuli were presented in isolation and in three sentential contexts: a neutral context, a context favoring /p/, and a context favoring /t/. For both natural and synthetic stimuli, the isolated and neutral context sentences yielded significantly more /t/ responses than sentence contexts primed for either /p/ or /t/. No other conditions were significantly different. Results did not show easily explainable semantic context effects. Instead, data clustering was more readily explained by top-down feedback processing affecting phonemic representation.

Elaboración de una versión reducida de las listas de frases en español (vr-LFE) para evaluar la percepción del habla con ruido

This article presents, a short version of the ‘listas de frases en español’ (vr-LFE; “Lists of Phrases in Spanish”) to evaluate speech perception in noise. The original version is an adaptation of the SPIN (Speech Perception in Noise) test for the English language. The original LFE consists of 6 lists, each of them including 25 high-predictability sentences and 25 low-predictability sentences. The difference between the correct recognition scores on the two types of sentences provides an estimation of the effective use of linguistic context (semantic, syntactic, and pragmatic information) by the listener in the speech perception process. In the present work, 60 sentences were selected from an initial pool of 300 sentences. These 60 sentences were arranged in 5 lists of 12 sentences each (6 high-predictability and 6 low-predictability sentences). This version preserves the same characteristics of reliability (equivalence among the lists) and validity (sensibility to the effects of sentence context and to the signal-to-noise ratio) as the original version, but considerably reduces the time needed for its administration. These lists can be used to evaluate speech perception difficulties in noise in older listeners with mild-to-moderate hearing loss (presbycusis), with or without mild cognitive impairment, and in non-native listeners with different degrees of linguistic competence.

When Language Switching has No Apparent Cost: Lexical Access in Sentence Context.

We report two experiments that investigate the effects of sentence context on bilingual lexical access in Spanish and English. Highly proficient Spanish-English bilinguals read sentences in Spanish and English that included a marked word to be named. The word was either a cognate with similar orthography and/or phonology in the two languages, or a matched non-cognate control. Sentences appeared in one language alone (i.e., Spanish or English) and target words were not predictable on the basis of the preceding semantic context. In Experiment 1, we mixed the language of the sentence within a block such that sentences appeared in an alternating run in Spanish or in English. These conditions partly resemble normally occurring inter-sentential code-switching. In these mixed-language sequences, cognates were named faster than non-cognates in both languages. There were no effects of switching the language of the sentence. In Experiment 2, with Spanish-English bilinguals matched closely to those who participated in the first experiment, we blocked the language of the sentences to encourage language-specific processes. The results were virtually identical to those of the mixed-language experiment. In both cases, target cognates were named faster than non-cognates, and the magnitude of the effect did not change according to the broader context. Taken together, the results support the predictions of the Bilingual Interactive Activation + Model (Dijkstra and van Heuven, 2002) in demonstrating that bilingual lexical access is language non-selective even under conditions in which language-specific cues should enable selective processing. They also demonstrate that, in contrast to lexical switching from one language to the other, inter-sentential code-switching of the sort in which bilinguals frequently engage, imposes no significant costs to lexical processing.

SPaM: A combined self-paced reading and masked-priming paradigm

In this article, we validate an experimental paradigm, SPaM, that we first described elsewhere (Luke & Christianson, Memory & Cognition 40:628-641, 2012). SPaM is a synthesis of self-paced reading and masked priming. The primary purpose of SPaM is to permit the study of sentence context effects on early word recognition. In the experiment reported here, we show that SPaM successfully reproduces results from both the self-paced reading and masked-priming literatures. We also outline the advantages and potential uses of this paradigm. For users of E-Prime, the experimental program can be downloaded from our lab website, http://epl.beckman.illinois.edu/.

To predict or not to predict: Age-related differences in the use of sentential context.

Older adults (as a group) are less likely than younger adults to engage in an anticipatory mode of language comprehension, failing to successfully preactivate information about upcoming likely (predictable) words during online processing. To assess (within one set of materials) age-related changes in the use of sentential context to affect processing of predictable words and in the consequences of violating predictions, event-related brain potentials were recorded while older adults read sentences that varied in sentence-level constraint and expectancy of sentence-final words. Strongly constraining sentences were completed by their most expected, predictable words and weakly constraining sentences were completed by their most expected, less predictable words. Both types of sentences also were completed by unexpected (but plausible) words. Older adults showed reduced and delayed effects of sentential context on processing predictable words. Whereas younger adults elicit an enhanced positive ERP (starting around 500 ms poststimulus onset, largest over prefrontal electrode sites), specifically for unexpected words that violate strong expectancies for a different word, older adults as a group did not exhibit this neural consequence of disconfirmed predictions. Older adults were instead more likely to show a left-lateralized frontal negativity for predictable items. This ERP response has been attributed to processes needed to revisit contextual material in forming an interpretation of message-level meaning, which may be more likely when anticipatory modes of comprehension are not engaged. Taken together, the results suggest that normal aging can affect allocation of resources to different cognitive and neural pathways in achieving comprehension outcomes.

The phonemic restoration effect reveals pre-N400 effect of supportive sentence context in speech perception

The phonemic restoration effect refers to the tendency for people to hallucinate a phoneme replaced by a non-speech sound (e.g., a tone) in a word. This illusion can be influenced by preceding sentential context providing information about the likelihood of the missing phoneme. The saliency of the illusion suggests that supportive context can affect relatively low (phonemic or lower) levels of speech processing. Indeed, a previous event-related brain potential (ERP) investigation of the phonemic restoration effect found that the processing of coughs replacing high versus low probability phonemes in sentential words differed from each other as early as the auditory N1 (120–180 ms post-stimulus); this result, however, was confounded by physical differences between the high and low probability speech stimuli, thus it could have been caused by factors such as habituation and not by supportive context. We conducted a similar ERP experiment avoiding this confound by using the same auditory stimuli preceded by text that made critical phonemes more or less probable. We too found the robust N400 effect of phoneme/word probability, but did not observe the early N1 effect. We did however observe a left posterior effect of phoneme/word probability around 192–224 ms—clear evidence of a relatively early effect of supportive sentence context in speech comprehension distinct from the N400.

Early ERF responses to final words in a sentence context during reading in individuals with autism spectrum disorders and typically developing individuals: an MEG study

Event Abstract Back to Event Early ERF responses to final words in a sentence context during reading in individuals with autism spectrum disorders and typically developing individuals: an MEG study Banu Ahtam1*, Sven Braeutigam1 and Anthony Bailey1 1 University of Oxford, United Kingdom This study used magnetoencephalography (MEG) to study the neural basis of abnormalities in sentence context effects in individuals with an autism spectrum disorder (ASD). 22 individuals with ASD and 22 typically developing (TD) individuals participated in the study. Participants were matched on age, gender, and IQ. All measurements were taken at the Oxford Neurodevelopmental Magnetoencephalography Centre using a Neuromag-306 VectorViewTM system, providing a helmet-shaped array of 102 pairs of gradiometers. Participants read sentences ending either with a homonym [dominant (DH) vs. subordinate (SH) meanings] or an unambiguous word (UW). The sentences were followed by a probe word that was semantically related or unrelated to the meaning of the sentence. Participants were asked to indicate whether the probe word was related or unrelated to the meaning of the sentence that it followed and to give their responses with a key press. This study has been approved by the local NHS (UK) Ethics Committee. All participants gave written informed consent before the experiment. The final word results showed that the ERF responses around 150ms were weaker and a little delayed in the ASD group. The rank ordering of amplitude strengths for ~150ms response to final words was the same in the TD and ASD groups (DH > UW > SH). Activity was observed over parietal and left occipito-temporal regions in both groups, but the left posterior temporal activity was reduced in ASD. The argument that semantic access can take place as early as 150-180ms [1] is supported by the findings of this study. The reduced activity in SH was more pronounced in the ASD group suggesting different semantic representations for more and less frequent meanings of the homonyms in the ASD than the TD group. Additionally, delayed responses to final words at ~150ms in ASD may suggest differences in the time course of semantic context influences on the semantic activation of the final words between TD and ASD.

Effects of sentence context on lexical ambiguity resolution in patients with schizophrenia

Previous research has suggested that a failure in processing contextual information may account for the heterogeneous clinical manifestations and cognitive impairments observed in schizophrenia. In the domain of language, context processing in schizophrenia has been investigated mostly with single-word semantic priming paradigms; however, natural language comprehension depends on more than semantic relations between words. The present study aimed to systematically assess sentence context effects in homonym meaning activation in patients with schizophrenia.Fourteen patients with schizophrenia and 14 normal controls matched to the patients on sex, age, education and parental education, were examined using a cross-modal priming paradigm. Primes were sentences biasing the first, second, or neither meaning of a sentence-final equibiased homonym; targets were related to either the first or the second meaning of the homonym and appeared after an interstimulus interval (ISI) of 0ms or 750ms.Patients with schizophrenia exhibited a trend towards facilitation of both target types following unbiased sentences at ISI=0ms, similar to controls. However, in contrast to the pattern of selective target facilitation exhibited by control subjects following first- or second meaning-biased sentences, no significant target facilitation was observed in patients in the same condition. At ISI=750ms, patients did no longer exhibit significant target facilitation in any sentence context condition. This pattern of results is compatible with the assumption of a combined impairment in lexical (automatic spreading of activation within the semantic network) and extralexical (working memory) processes in patients with schizophrenia.