Non-linear processing of a linear speech stream: The influence of morphological structure on the recognition of spoken Arabic words

Gwilliams, Laura; Marantz, Alec

doi:10.1016/j.bandl.2015.04.006

Cited by 44 publications

(47 citation statements)

References 58 publications

(74 reference statements)

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Concretely, this work suggests that previous context sets up probabilistic expectations about upcoming information, and biases acoustic perception to be consistent with the predicted phonemes (Warren, 1970;Cole, 1973;Samuel, 1981). The left STG and HG appear to be involved in this process, and activity in both regions correlates with the extent to which an expectation is violated (Gagnepain et al, 2012;Ettinger et al, 2014;Gwilliams and Marantz, 2015).…”

Section: Significance Statementmentioning

confidence: 59%

See 1 more Smart Citation

In spoken word recognition the future predicts the past

Gwilliams

Linzen

Poeppel

et al. 2017

Preprint

View full text Add to dashboard Cite

Speech is an inherently noisy and ambiguous signal. In order to fluently derive meaning, a listener must integrate contextual information to guide interpretations of the sensory input. While many studies have demonstrated the influence of prior context on speech perception, the neural mechanisms supporting the integration of subsequent context remain unknown. Using magnetoencephalography, we analysed responses to spoken words with a varyingly ambiguous onset phoneme, the identity of which is later disambiguated at the lexical uniqueness point1. Our findings suggest that primmary auditory cortex is sensitive to phonological ambiguity very early during processing — at just 50 ms after onset. Subphonemic detail is preserved in auditory cortex over long timescales, and re-evoked at subsequent phoneme positions. Commitments to phonological categories occur in parallel, resolving on the shorter time-scale of ~450 ms. These findings provide evidence that future input determines the perception of earlier speech sounds by maintaining sensory features until they can be integrated with top down lexical information.Significance statementThe perception of a speech sound is determined by its surrounding context, in the form of words, sentences, and other speech sounds. Often, such contextual information becomes available later than the sensory input. The present study is the first to unveil how the brain uses this subsequent information to aid speech comprehension. Concretely, we find that the auditory system supports prolonged access to the transient acoustic signal, while concurrently making guesses about the identity of the words being said. Such a processing strategy allows the content of the message to be accessed quickly, while also permitting re-analysis of the acoustic signal to minimise parsing mistakes.

show abstract

Section: Significance Statementmentioning

confidence: 59%

“…This sample size was selected based on previous studies using the same MEG machine (e.g. (Gwilliams and Marantz, 2015;Gwilliams et al, 2016)). They were recruited from the New York University Abu Dhabi community and were compensated for their time.…”

Section: Materials Creation (Common To Both Experiments)mentioning

confidence: 99%

In spoken word recognition the future predicts the past

Gwilliams

Linzen

Poeppel

et al. 2017

Preprint

View full text Add to dashboard Cite

show abstract

“…Importantly for the examination of naturalistic stimuli, N400 amplitude is known to vary parametrically with modulations of these cues, thus making it well suited to modeling neural activity based on continuous predictors and activity fluctuations on a trial-by-trial basis (cf. Cummings et al, 2006; Roehm et al, 2013; Sassenhagen et al, 2014; Payne et al, 2015; for isolated written words, see Hauk et al, 2006; Solomyak and Marantz, 2010; for isolated spoken words, see Ettinger et al, 2014; Lewis and Poeppel, 2014; Gwilliams and Marantz, 2015; and for written words in a story, see Brennan and Pylkkänen, 2012; Brennan and Pylkkänen, 2016). …”

Section: The N400mentioning

confidence: 99%

Electrophysiology Reveals the Neural Dynamics of Naturalistic Auditory Language Processing: Event-Related Potentials Reflect Continuous Model Updates

Alday

Schlesewsky

2017

eNeuro

View full text Add to dashboard Cite

The recent trend away from ANOVA-based analyses places experimental investigations into the neurobiology of cognition in more naturalistic and ecologically valid designs within reach. Using mixed-effects models for epoch-based regression, we demonstrate the feasibility of examining event-related potentials (ERPs), and in particular the N400, to study the neural dynamics of human auditory language processing in a naturalistic setting. Despite the large variability between trials during naturalistic stimulation, we replicated previous findings from the literature: the effects of frequency, animacy, and word order and find previously unexplored interaction effects. This suggests a new perspective on ERPs, namely, as a continuous modulation reflecting continuous stimulation instead of a series of discrete and essentially sequential processes locked to discrete events.

show abstract

“…This processing stage may therefore be linked to the perceptual stage of speech processing. Neural responses later than 100 ms appear to be associated with post-perceptual processing, such as the integration of previous expectations (e.g., phoneme surprisal; Ettinger, Linzen & Marantz, 2014; Gwilliams & Marantz, 2015) and integration of lexical information (Gow, 2008).…”

Section: Introductionmentioning

confidence: 99%

Neural adaptation to accented speech: prefrontal cortex aids attunement in auditory cortices

Gwilliams

Blanco-Elorrieta

Marantz

et al. 2019

Preprint

View full text Add to dashboard Cite

Speech is a complex and ambiguous acoustic signal that varies significantly within and across speakers. A prevalent and ubiquitous example of such variation is accented speech, to which humans adapt extremely rapidly. The goal of this study is to uncover the neurobiological bases of the attunement process that enables such fluent comprehension. Twenty-four native English participants listened to words spoken by an unaccented "canonical" American talker and two "accented" talkers, and performed a word-picture matching task, while magnetoencephalography (MEG) was recorded. Accented speech was created by including systematic phonological substitutions within the word (e.g. [s] → [sh]). Activity in the auditory cortex (superior temporal gyrus) was greater for accented speech, but, critically, this was not attenuated by exposure. By contrast, prefrontal regions showed an interaction between the presence of an accent and amount of exposure: while activity decreased for canonical speech over time, responses to accented speech remained consistently elevated. Grainger causality analyses further revealed that prefrontal responses serve to modulate activity in auditory regions, suggesting the recruitment of top-down processing to decode accented signal. In sum, our results show that accented speech does not elicit the same prefrontal reduction in amplitude over time that unaccented speech does, and points to a dynamic exchange of information between the prefrontal and auditory cortices in order to recalculate phonetic classification and subsequent identification of lexical items.Significance statement: Human ability to adapt to different people's idiosyncratic pronunciations is a hallmark of speech comprehension. This study aims to address whether adaptation to speakers' accents manifests itself at the perceptual level (i.e. through adaptation of low-level neural responses in the auditory cortex) or at the post-perceptual level (i.e. higher-order regions correcting the sensory signal received from auditory cortex). Our results support the post-perceptual hypothesis: we found that responses in auditory cortex emitted an error-like signal that was invariant to exposure; whereas responses in the prefrontal cortex were modulated by exposure. These findings provide initial insight into accent adaptation, and illuminate the computational stages supporting speech comprehension more generally.

show abstract

Non-linear processing of a linear speech stream: The influence of morphological structure on the recognition of spoken Arabic words

Cited by 44 publications

References 58 publications

In spoken word recognition the future predicts the past

In spoken word recognition the future predicts the past

Electrophysiology Reveals the Neural Dynamics of Naturalistic Auditory Language Processing: Event-Related Potentials Reflect Continuous Model Updates

Neural adaptation to accented speech: prefrontal cortex aids attunement in auditory cortices

Contact Info

Product

Resources

About