Neural Representations of Non-native Speech Reflect Proficiency and Interference from Native Language Knowledge.

Learning to process speech in a foreign language involves learning new representations for mapping the auditory signal to linguistic structure. Behavioral experiments suggest that even listeners that are highly proficient in a non-native language experience interference from representations of their native language. However, much of the evidence for such interference comes from tasks that may inadvertently increase the salience of native language competitors. Here we tested for neural evidence of proficiency and native language interference in a naturalistic story listening task. We studied electroencephalography responses of 39 native speakers of Dutch (14 male) to an English short story, spoken by a native speaker of either American English or Dutch. We modeled brain responses with multivariate temporal response functions, using acoustic and language models. We found evidence for activation of Dutch language statistics when listening to English, but only when it was spoken with a Dutch accent. This suggests that a naturalistic, monolingual setting decreases the interference from native language representations, whereas an accent in the listener's own native language may increase native language interference, by increasing the salience of the native language and activating native language phonetic and lexical representations. Brain responses suggest that such interference stems from words from the native language competing with the foreign language in a single word recognition system, rather than being activated in a parallel lexicon. We further found that secondary acoustic representations of speech (after 200 ms latency) decreased with increasing proficiency. This may reflect improved acoustic-phonetic models in more proficient listeners.Significance Statement Behavioral experiments suggest that native language knowledge interferes with foreign language listening, but such effects may be sensitive to task manipulations, as tasks that increase metalinguistic awareness may also increase native language interference. This highlights the need for studying non-native speech processing using naturalistic tasks. We measured neural responses unobtrusively while participants listened for comprehension and characterized the influence of proficiency at multiple levels of representation. We found that salience of the native language, as manipulated through speaker accent, affected activation of native language representations: significant evidence for activation of native language (Dutch) categories was only obtained when the speaker had a Dutch accent, whereas no significant interference was found to a speaker with a native (American) accent.

Recognizing non-native spoken words in background noise increases interference from the native language.

Listeners frequently recognize spoken words in the presence of background noise. Previous research has shown that noise reduces phoneme intelligibility and hampers spoken-word recognition - especially for non-native listeners. In the present study, we investigated how noise influences lexical competition in both the non-native and the native language, reflecting the degree to which both languages are co-activated. We recorded the eye movements of native Dutch participants as they listened to English sentences containing a target word while looking at displays containing four objects. On target-present trials, the visual referent depicting the target word was present, along with three unrelated distractors. On target-absent trials, the target object (e.g., wizard) was absent. Instead, the display contained an English competitor, overlapping with the English target in phonological onset (e.g., window), a Dutch competitor, overlapping with the English target in phonological onset (e.g., wimpel, pennant), and two unrelated distractors. Half of the sentences was masked by speech-shaped noise; the other half was presented in quiet. Compared to speech in quiet, noise delayed fixations to the target objects on target-present trials. For target-absent trials, we observed that the likelihood for fixation biases towards the English and Dutch onset competitors (over the unrelated distractors) was larger in noise than in quiet. Our data thus show that the presence of background noise increases lexical competition in the task-relevant non-native (English) and in the task-irrelevant native (Dutch) language. The latter reflects stronger interference of one's native language during non-native spoken-word recognition under adverse conditions.

The time course of adaptation to distorted speech.

When confronted with unfamiliar or novel forms of speech, listeners' word recognition performance is known to improve with exposure, but data are lacking on the fine-grained time course of adaptation. The current study aims to fill this gap by investigating the time course of adaptation to several different types of distorted speech. Keyword scores as a function of sentence position in a block of 30 sentences were measured in response to eight forms of distorted speech. Listeners recognised twice as many words in the final sentence compared to the initial sentence with around half of the gain appearing in the first three sentences, followed by gradual gains over the rest of the block. Rapid adaptation was apparent for most of the eight distortion types tested with differences mainly in the gradual phase. Adaptation to sine-wave speech improved if listeners had heard other types of distortion prior to exposure, but no similar facilitation occurred for the other types of distortion. Rapid adaptation is unlikely to be due to procedural learning since listeners had been familiarised with the task and sentence format through exposure to undistorted speech. The mechanisms that underlie rapid adaptation are currently unclear.

The differential roles of lexical and sublexical processing during spoken-word recognition in clear and in noise.

Successful spoken-word recognition relies on interplay between lexical and sublexical processing. Previous research demonstrated that listeners readily shift between more lexically-biased and more sublexically-biased modes of processing in response to the situational context in which language comprehension takes place. Recognizing words in the presence of background noise reduces the perceptual evidence for the speech signal and - compared to the clear - results in greater uncertainty. It has been proposed that, when dealing with greater uncertainty, listeners rely more strongly on sublexical processing. The present study tested this proposal using behavioral and electroencephalography (EEG) measures. We reasoned that such an adjustment would be reflected in changes in the effects of variables predicting recognition performance with loci at lexical and sublexical levels, respectively. We presented native speakers of Dutch with words featuring substantial variability in (1) word frequency (locus at lexical level), (2) phonological neighborhood density (loci at lexical and sublexical levels) and (3) phonotactic probability (locus at sublexical level). Each participant heard each word in noise (presented at one of three signal-to-noise ratios) and in the clear and performed a two-stage lexical decision and transcription task while EEG was recorded. Using linear mixed-effects analyses, we observed behavioral evidence that listeners relied more strongly on sublexical processing when speech quality decreased. Mixed-effects modelling of the EEG signal in the clear condition showed that sublexical effects were reflected in early modulations of ERP components (e.g., within the first 300 msec post word onset). In noise, EEG effects occurred later and involved multiple regions activated in parallel. Taken together, we found evidence - especially in the behavioral data - supporting previous accounts that the presence of background noise induces a stronger reliance on sublexical processing.

The Presence of Background Noise Extends the Competitor Space in Native and Non-Native Spoken-Word Recognition: Insights from Computational Modeling.

Oral communication often takes place in noisy environments, which challenge spoken-word recognition. Previous research has suggested that the presence of background noise extends the number of candidate words competing with the target word for recognition and that this extension affects the time course and accuracy of spoken-word recognition. In this study, we further investigated the temporal dynamics of competition processes in the presence of background noise, and how these vary in listeners with different language proficiency (i.e., native and non-native) using computational modeling. We developed ListenIN (Listen-In-Noise), a neural-network model based on an autoencoder architecture, which learns to map phonological forms onto meanings in two languages and simulates native and non-native spoken-word comprehension. We also examined the model's activation states during online spoken-word recognition. These analyses demonstrated that the presence of background noise increases the number of competitor words, which are engaged in phonological competition and that this happens in similar ways intra and interlinguistically and in native and non-native listening. Taken together, our results support accounts positing a "many-additional-competitors scenario" for the effects of noise on spoken-word recognition.

Cross-linguistic Influences on Sentence Accent Detection in Background Noise.

This paper investigates whether sentence accent detection in a non-native language is dependent on (relative) similarity between prosodic cues to accent between the non-native and the native language, and whether cross-linguistic differences in the use of local and more widely distributed (i.e., non-local) cues to sentence accent detection lead to differential effects of the presence of background noise on sentence accent detection in a non-native language. We compared Dutch, Finnish, and French non-native listeners of English, whose cueing and use of prosodic prominence is gradually further removed from English, and compared their results on a phoneme monitoring task in different levels of noise and a quiet condition to those of native listeners. Overall phoneme detection performance was high for the native and the non-native listeners, but deteriorated to the same extent in the presence of background noise. Crucially, relative similarity between the prosodic cues to sentence accent of one's native language compared to that of a non-native language does not determine the ability to perceive and use sentence accent for speech perception in that non-native language. Moreover, proficiency in the non-native language is not a straightforward predictor of sentence accent perception performance, although high proficiency in a non-native language can seemingly overcome certain differences at the prosodic level between the native and non-native language. Instead, performance is determined by the extent to which listeners rely on local cues (English and Dutch) versus cues that are more distributed (Finnish and French), as more distributed cues survive the presence of background noise better.

Talker-familiarity benefit in non-native recognition memory and word identification: The role of listening conditions and proficiency.

Native listeners benefit from talker familiarity in recognition memory and word identification, especially in adverse listening conditions. The present study addresses the talker familiarity benefit in non-native listening, and the role of listening conditions and listeners' lexical proficiency in the emergence of this benefit. Dutch non-native listeners of English were trained to identify four English talkers over 4 days. Talker familiarity benefit in recognition memory was investigated using a recognition memory task with "old" and "new" words produced by familiar and unfamiliar talkers presented either in the clear or in noise. Talker familiarity benefit in word identification was investigated by comparing non-native listeners' performances on the first and the last day in identifying words in different noise levels, produced by either a trained (included in the voice recognition training) or by an untrained talker (not included in the voice recognition training). Non-native listeners demonstrated a talker familiarity benefit in recognition memory, which was modulated by listening conditions and proficiency in the non-native language. No talker familiarity benefit was found in word identification. These results suggest that, similar to native listening, both linguistic and indexical (talker-specific) information influence non-native speech perception. However, this is dependent on the task and type of speech recognition process involved.

The effect of background noise on the word activation process in nonnative spoken-word recognition.

This article investigates 2 questions: (1) does the presence of background noise lead to a differential increase in the number of simultaneously activated candidate words in native and nonnative listening? And (2) do individual differences in listeners' cognitive and linguistic abilities explain the differential effect of background noise on (non-)native speech recognition? English and Dutch students participated in an English word recognition experiment, in which either a word's onset or offset was masked by noise. The native listeners outperformed the nonnative listeners in all listening conditions. Importantly, however, the effect of noise on the multiple activation process was found to be remarkably similar in native and nonnative listening. The presence of noise increased the set of candidate words considered for recognition in both native and nonnative listening. The results indicate that the observed performance differences between the English and Dutch listeners should not be primarily attributed to a differential effect of noise, but rather to the difference between native and nonnative listening. Additional analyses showed that word-initial information was found to be more important than word-final information during spoken-word recognition. When word-initial information was no longer reliably available word recognition accuracy dropped and word frequency information could no longer be used suggesting that word frequency information is strongly tied to the onset of words and the earliest moments of lexical access. Proficiency and inhibition ability were found to influence nonnative spoken-word recognition in noise, with a higher proficiency in the nonnative language and worse inhibition ability leading to improved recognition performance. (PsycINFO Database Record

L2 voice recognition: The role of speaker-, listener-, and stimulus-related factors.

Previous studies examined various factors influencing voice recognition and learning with mixed results. The present study investigates the separate and combined contribution of these various speaker-, stimulus-, and listener-related factors to voice recognition. Dutch listeners, with arguably incomplete phonological and lexical knowledge in the target language, English, learned to recognize the voice of four native English speakers, speaking in English, during four-day training. Training was successful and listeners' accuracy was shown to be influenced by the acoustic characteristics of speakers and the sound composition of the words used in the training, but not by lexical frequency of the words, nor the lexical knowledge of the listeners or their phonological aptitude. Although not conclusive, listeners with a lower working memory capacity seemed to be slower in learning voices than listeners with a higher working memory capacity. The results reveal that speaker-related, listener-related, and stimulus-related factors accumulate in voice recognition, while lexical information turns out not to play a role in successful voice learning and recognition. This implies that voice recognition operates at the prelexical processing level.

Perception of Emotion in Conversational Speech by Younger and Older Listeners.

This study investigated whether age and/or differences in hearing sensitivity influence the perception of the emotion dimensions arousal (calm vs. aroused) and valence (positive vs. negative attitude) in conversational speech. To that end, this study specifically focused on the relationship between participants' ratings of short affective utterances and the utterances' acoustic parameters (pitch, intensity, and articulation rate) known to be associated with the emotion dimensions arousal and valence. Stimuli consisted of short utterances taken from a corpus of conversational speech. In two rating tasks, younger and older adults either rated arousal or valence using a 5-point scale. Mean intensity was found to be the main cue participants used in the arousal task (i.e., higher mean intensity cueing higher levels of arousal) while mean F 0 was the main cue in the valence task (i.e., higher mean F 0 being interpreted as more negative). Even though there were no overall age group differences in arousal or valence ratings, compared to younger adults, older adults responded less strongly to mean intensity differences cueing arousal and responded more strongly to differences in mean F 0 cueing valence. Individual hearing sensitivity among the older adults did not modify the use of mean intensity as an arousal cue. However, individual hearing sensitivity generally affected valence ratings and modified the use of mean F 0. We conclude that age differences in the interpretation of mean F 0 as a cue for valence are likely due to age-related hearing loss, whereas age differences in rating arousal do not seem to be driven by hearing sensitivity differences between age groups (as measured by pure-tone audiometry).

Do Hearing Aids Improve Affect Perception?

Normal-hearing listeners use acoustic cues in speech to interpret a speaker's emotional state. This study investigates the effect of hearing aids on the perception of the emotion dimensions arousal (aroused/calm) and valence (positive/negative attitude) in older adults with hearing loss. More specifically, we investigate whether wearing a hearing aid improves the correlation between affect ratings and affect-related acoustic parameters. To that end, affect ratings by 23 hearing-aid users were compared for aided and unaided listening. Moreover, these ratings were compared to the ratings by an age-matched group of 22 participants with age-normal hearing.For arousal, hearing-aid users rated utterances as generally more aroused in the aided than in the unaided condition. Intensity differences were the strongest indictor of degree of arousal. Among the hearing-aid users, those with poorer hearing used additional prosodic cues (i.e., tempo and pitch) for their arousal ratings, compared to those with relatively good hearing. For valence, pitch was the only acoustic cue that was associated with valence. Neither listening condition nor hearing loss severity (differences among the hearing-aid users) influenced affect ratings or the use of affect-related acoustic parameters. Compared to the normal-hearing reference group, ratings of hearing-aid users in the aided condition did not generally differ in both emotion dimensions. However, hearing-aid users were more sensitive to intensity differences in their arousal ratings than the normal-hearing participants.We conclude that the use of hearing aids is important for the rehabilitation of affect perception and particularly influences the interpretation of arousal.

Age and hearing loss and the use of acoustic cues in fricative categorization.

This study examined the use of fricative noise information and coarticulatory cues for categorization of word-final fricatives [s] and [f] by younger and older Dutch listeners alike. Particularly, the effect of information loss in the higher frequencies on the use of these two cues for fricative categorization was investigated. If information in the higher frequencies is less strongly available, fricative identification may be impaired or listeners may learn to focus more on coarticulatory information. The present study investigates this second possibility. Phonetic categorization results showed that both younger and older Dutch listeners use the primary cue fricative noise and the secondary cue coarticulatory information to distinguish word-final [f] from [s]. Individual hearing sensitivity in the older listeners modified the use of fricative noise information, but did not modify the use of coarticulatory information. When high-frequency information was filtered out from the speech signal, fricative noise could no longer be used by the younger and older adults. Crucially, they also did not learn to rely more on coarticulatory information as a compensatory cue for fricative categorization. This suggests that listeners do not readily show compensatory use of this secondary cue to fricative identity when fricative categorization becomes difficult.

The role of attentional abilities in lexically guided perceptual learning by older listeners.

This study investigates two variables that may modify lexically guided perceptual learning: individual hearing sensitivity and attentional abilities. Older Dutch listeners (aged 60+ years, varying from good hearing to mild-to-moderate high-frequency hearing loss) were tested on a lexically guided perceptual learning task using the contrast [f]-[s]. This contrast mainly differentiates between the two consonants in the higher frequencies, and thus is supposedly challenging for listeners with hearing loss. The analyses showed that older listeners generally engage in lexically guided perceptual learning. Hearing loss and selective attention did not modify perceptual learning in our participant sample, while attention-switching control did: listeners with poorer attention-switching control showed a stronger perceptual learning effect. We postulate that listeners with better attention-switching control may, in general, rely more strongly on bottom-up acoustic information compared to listeners with poorer attention-switching control, making them in turn less susceptible to lexically guided perceptual learning. Our results, moreover, clearly show that lexically guided perceptual learning is not lost when acoustic processing is less accurate.

Phoneme categorization and discrimination in younger and older adults: a comparative analysis of perceptual, lexical, and attentional factors.

This study investigates the extent to which age-related language processing difficulties are due to a decline in sensory processes or to a deterioration of cognitive factors, specifically, attentional control. Two facets of attentional control were examined: inhibition of irrelevant information and divided attention. Younger and older adults were asked to categorize the initial phoneme of spoken syllables ("Was it m or n?"), trying to ignore the lexical status of the syllables. The phonemes were manipulated to range in eight steps from m to n. Participants also did a discrimination task on syllable pairs ("Were the initial sounds the same or different?"). Categorization and discrimination were performed under either divided attention (concurrent visual-search task) or focused attention (no visual task). The results showed that even when the younger and older adults were matched on their discrimination scores: (1) the older adults had more difficulty inhibiting lexical knowledge than did younger adults, (2) divided attention weakened lexical inhibition in both younger and older adults, and (3) divided attention impaired sound discrimination more in older than younger listeners. The results confirm the independent and combined contribution of sensory decline and deficit in attentional control to language processing difficulties associated with aging. The relative weight of these variables and their mechanisms of action are discussed in the context of theories of aging and language.

Phonological abstraction without phonemes in speech perception.

Recent evidence shows that listeners use abstract prelexical units in speech perception. Using the phenomenon of lexical retuning in speech processing, we ask whether those units are necessarily phonemic. Dutch listeners were exposed to a Dutch speaker producing ambiguous phones between the Dutch syllable-final allophones approximant [r] and dark [l]. These ambiguous phones replaced either final /r/ or final /l/ in words in a lexical-decision task. This differential exposure affected perception of ambiguous stimuli on the same allophone continuum in a subsequent phonetic-categorization test: Listeners exposed to ambiguous phones in /r/-final words were more likely to perceive test stimuli as /r/ than listeners with exposure in /l/-final words. This effect was not found for test stimuli on continua using other allophones of /r/ and /l/. These results confirm that listeners use phonological abstraction in speech perception. They also show that context-sensitive allophones can play a role in this process, and hence that context-insensitive phonemes are not necessary. We suggest there may be no one unit of perception.

Comparing lexically guided perceptual learning in younger and older listeners.

Numerous studies have shown that younger adults engage in lexically guided perceptual learning in speech perception. Here, we investigated whether older listeners are also able to retune their phonetic category boundaries. More specifically, in this research we tried to answer two questions. First, do older adults show perceptual-learning effects of similar size to those of younger adults? Second, do differences in lexical behavior predict the strength of the perceptual-learning effect? An age group comparison revealed that older listeners do engage in lexically guided perceptual learning, but there were two age-related differences: Younger listeners had a stronger learning effect right after exposure than did older listeners, but the effect was more stable for older than for younger listeners. Moreover, a clear link was shown to exist between individuals' lexical-decision performance during exposure and the magnitude of their perceptual-learning effects. A subsequent analysis on the results of the older participants revealed that, even within the older participant group, with increasing age the perceptual retuning effect became smaller but also more stable, mirroring the age group comparison results. These results could not be explained by differences in hearing loss. The age effect may be accounted for by decreased flexibility in the adjustment of phoneme categories or by age-related changes in the dynamics of spoken-word recognition, with older adults being more affected by competition from similar-sounding lexical competitors, resulting in less lexical guidance for perceptual retuning. In conclusion, our results clearly show that the speech perception system remains flexible over the life span.

Models of spoken-word recognition.

All words of the languages we know are stored in the mental lexicon. Psycholinguistic models describe in which format lexical knowledge is stored and how it is accessed when needed for language use. The present article summarizes key findings in spoken-word recognition by humans and describes how models of spoken-word recognition account for them. Although current models of spoken-word recognition differ considerably in the details of implementation, there is general consensus among them on at least three aspects: multiple word candidates are activated in parallel as a word is being heard, activation of word candidates varies with the degree of match between the speech signal and stored lexical representations, and activated candidate words compete for recognition. No consensus has been reached on other aspects such as the flow of information between different processing levels, and the format of stored prelexical and lexical representations. WIREs Cogn Sci 2012, 3:387-401. doi: 10.1002/wcs.1178 For further resources related to this article, please visit the WIREs website.

Modeling the use of durational information in human spoken-word recognition.

Evidence that listeners, at least in a laboratory environment, use durational cues to help resolve temporarily ambiguous speech input has accumulated over the past decades. This paper introduces Fine-Tracker, a computational model of word recognition specifically designed for "tracking" fine-phonetic information in the acoustic speech signal and using it during word recognition. Two simulations were carried out using real speech as input to the model. The simulations showed that the Fine-Tracker, as has been found for humans, benefits from durational information during word recognition, and uses it to disambiguate the incoming speech signal. The availability of durational information allows the computational model to distinguish embedded words from their matrix words (first simulation), and to distinguish word final realizations of [s] from word initial realizations (second simulation). Fine-Tracker thus provides the first computational model of human word recognition that is able to extract durational information from the speech signal and to use it to differentiate words.

Unsupervised speech segmentation: an analysis of the hypothesized phone boundaries.

Despite using different algorithms, most unsupervised automatic phone segmentation methods achieve similar performance in terms of percentage correct boundary detection. Nevertheless, unsupervised segmentation algorithms are not able to perfectly reproduce manually obtained reference transcriptions. This paper investigates fundamental problems for unsupervised segmentation algorithms by comparing a phone segmentation obtained using only the acoustic information present in the signal with a reference segmentation created by human transcribers. The analyses of the output of an unsupervised speech segmentation method that uses acoustic change to hypothesize boundaries showed that acoustic change is a fairly good indicator of segment boundaries: over two-thirds of the hypothesized boundaries coincide with segment boundaries. Statistical analyses showed that the errors are related to segment duration, sequences of similar segments, and inherently dynamic phones. In order to improve unsupervised automatic speech segmentation, current one-stage bottom-up segmentation methods should be expanded into two-stage segmentation methods that are able to use a mix of bottom-up information extracted from the speech signal and automatically derived top-down information. In this way, unsupervised methods can be improved while remaining flexible and language-independent.

How should a speech recognizer work?

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.