scholarly journals Combining predictive coding with neural oscillations optimizes on-line speech processing

2018 ◽  
Author(s):  
Sevada Hovsepyan ◽  
Itsaso Olasagasti ◽  
Anne-Lise Giraud
Keyword(s):  
Sensory Processing ◽  
Speech Processing ◽  
Predictive Coding ◽  
Theta Oscillations ◽  
Neural Oscillations ◽  
Speech Comprehension ◽  
Contextual Cues ◽  
Continuous Speech ◽  
Neural Representations ◽  
On Line

ABSTRACTSpeech comprehension requires segmenting continuous speech to connect it on-line with discrete linguistic neural representations. This process relies on theta-gamma oscillation coupling, which tracks syllables and encodes them in decipherable neural activity. Speech comprehension also strongly depends on contextual cues predicting speech structure and content. To explore the effects of theta-gamma coupling on bottom-up/top-down dynamics during on-line speech perception, we designed a generative model that can recognize syllable sequences in continuous speech. The model uses theta oscillations to detect syllable onsets and align both gamma-rate encoding activity with syllable boundaries and predictions with speech input. We observed that the model performed best when theta oscillations were used to align gamma units with input syllables, i.e. when bidirectional information flows were coordinated, and internal timing knowledge was exploited. This work demonstrates that notions of predictive coding and neural oscillations can usefully be brought together to account for dynamic on-line sensory processing.

The Role of Low-frequency Neural Oscillations in Speech Processing: Revisiting Delta Entrainment

2019 ◽  
Vol 31 (8) ◽  
pp. 1205-1215 ◽  
Author(s):  
Victor J. Boucher ◽  
Annie C. Gilbert ◽  
Boutheina Jemel
Keyword(s):  
Speech Processing ◽  
Low Frequency ◽  
Phase Coherence ◽  
Theta Oscillations ◽  
Neural Oscillations ◽  
Acoustic Cues ◽  
Low Frequency Oscillations ◽  
Temporal Grouping ◽  
Delta Oscillations

Studies that use measures of cerebro-acoustic coherence have shown that theta oscillations (3–10 Hz) entrain to syllable-size modulations in the energy envelope of speech. This entrainment creates sensory windows in processing acoustic cues. Recent reports submit that delta oscillations (<3 Hz) can be entrained by nonsensory content units like phrases and serve to process meaning—though such views face fundamental problems. Other studies suggest that delta underlies a sensory chunking linked to the processing of sequential attributes of speech sounds. This chunking associated with the “focus of attention” is commonly manifested by the temporal grouping of items in sequence recall. Similar grouping in speech may entrain delta. We investigate this view by examining how low-frequency oscillations entrain to three types of stimuli (tones, nonsense syllables, and utterances) having similar timing, pitch, and energy contours. Entrainment was indexed by “intertrial phase coherence” in the EEGs of 18 listeners. The results show that theta oscillations at central sites entrain to syllable-size elements in speech and tones. However, delta oscillations at frontotemporal sites specifically entrain to temporal groups in both meaningful utterances and meaningless syllables, which indicates that delta may support but does not directly bear on a processing of content. The findings overall suggest that, although theta entrainment relates to a processing of acoustic attributes, delta entrainment links to a sensory chunking that relates to a processing of properties of articulated sounds. The results also show that measures of intertrial phase coherence can be better suited than cerebro-acoustic coherence in revealing delta entrainment.

scholarly journals Sounding It Out: Auditory Stimulation and Overnight Memory Processing

2021 ◽  
Author(s):  
Marcus O. Harrington ◽  
Scott A. Cairney
Keyword(s):  
Rem Sleep ◽  
Memory Consolidation ◽  
Slow Wave Sleep ◽  
Auditory Stimulation ◽  
Theta Oscillations ◽  
Neural Oscillations ◽  
Healthy Children ◽  
Factors Affecting ◽  
Memory Processing ◽  
Eeg Recordings

Abstract Purpose of Review Auditory stimulation is a technique that can enhance neural oscillations linked to overnight memory consolidation. In this review, we evaluate the impacts of auditory stimulation on the neural oscillations of sleep and associated memory processes in a variety of populations. Recent Findings Cortical EEG recordings of slow-wave sleep (SWS) are characterised by two cardinal oscillations: slow oscillations (SOs) and sleep spindles. Auditory stimulation delivered in SWS enhances SOs and phase-coupled spindle activity in healthy children and adults, children with ADHD, adults with mild cognitive impairment and patients with major depression. Under certain conditions, auditory stimulation bolsters the benefits of SWS for memory consolidation, although further work is required to fully understand the factors affecting stimulation-related memory gains. Recent work has turned to rapid eye movement (REM) sleep, demonstrating that auditory stimulation can be used to manipulate REM sleep theta oscillations. Summary Auditory stimulation enhances oscillations linked to overnight memory processing and shows promise as a technique for enhancing the memory benefits of sleep.

Auditory–Articulatory Neural Alignment between Listener and Speaker during Verbal Communication

2019 ◽  
Vol 30 (3) ◽  
pp. 942-951 ◽  
Author(s):  
Lanfang Liu ◽  
Yuxuan Zhang ◽  
Qi Zhou ◽  
Douglas D Garrett ◽  
Chunming Lu ◽  
...  
Keyword(s):  
Speech Processing ◽  
Auditory Processing ◽  
Information Transfer ◽  
Temporal Cortex ◽  
Real Life ◽  
Hierarchical Organization ◽  
Speech Comprehension ◽  
Level Information ◽  
Motor Information ◽  
High Level

Abstract Whether auditory processing of speech relies on reference to the articulatory motor information of speaker remains elusive. Here, we addressed this issue under a two-brain framework. Functional magnetic resonance imaging was applied to record the brain activities of speakers when telling real-life stories and later of listeners when listening to the audio recordings of these stories. Based on between-brain seed-to-voxel correlation analyses, we revealed that neural dynamics in listeners’ auditory temporal cortex are temporally coupled with the dynamics in the speaker’s larynx/phonation area. Moreover, the coupling response in listener’s left auditory temporal cortex follows the hierarchical organization for speech processing, with response lags in A1+, STG/STS, and MTG increasing linearly. Further, listeners showing greater coupling responses understand the speech better. When comprehension fails, such interbrain auditory-articulation coupling vanishes substantially. These findings suggest that a listener’s auditory system and a speaker’s articulatory system are inherently aligned during naturalistic verbal interaction, and such alignment is associated with high-level information transfer from the speaker to the listener. Our study provides reliable evidence supporting that references to the articulatory motor information of speaker facilitate speech comprehension under a naturalistic scene.

scholarly journals Continuous speech processing

2020 ◽  
Vol 18 ◽  
pp. 25-31 ◽  
Author(s):  
Christian Brodbeck ◽  
Jonathan Z Simon
Keyword(s):  
Speech Processing ◽  
Continuous Speech

Phonetic Analysis and the Automatic Segmentation and Labeling of Speech Sounds

1990 ◽  
Vol 20 (1) ◽  
pp. 15-21 ◽  
Author(s):  
Peter Roach ◽  
Helen Roach ◽  
Andrea Dew ◽  
Paul Rowlands
Keyword(s):  
Speech Processing ◽  
Acoustic Analysis ◽  
Automatic Segmentation ◽  
Research Work ◽  
Speech Sounds ◽  
Computer Algorithms ◽  
Continuous Speech ◽  
Considerable Debate ◽  
Theoretical Issues

A fundamentally important practice in phonetics is the analysis of continuous speech into a sequence of discrete segments. There has been considerable debate about the theoretical validity of this practice within classical auditory/kinaesthetic phonetics and in phonology as well as in the context of the acoustic analysis of speech; in recent years the issue has become more widely important as research work in automatic speech processing has resulted in computer algorithms for segmenting speech and assigning phonetic labels to the segments. Work on the automatic segmentation and labeling of speech sounds has been carried on in our department since 1980. This paper begins by examining the theoretical issues involved in segmentation and labeling, then describes our own work.

scholarly journals High-frequency neural activity predicts word parsing in ambiguous speech streams

2016 ◽  
Vol 116 (6) ◽  
pp. 2497-2512 ◽  
Author(s):  
Anne Kösem ◽  
Anahita Basirat ◽  
Leila Azizi ◽  
Virginie van Wassenhove
Keyword(s):  
High Frequency ◽  
Low Frequency ◽  
Acoustic Properties ◽  
Neural Oscillations ◽  
Speech Comprehension ◽  
Top Down ◽  
Linguistic Representations ◽  
High Frequency Activity ◽  
Speech Percept ◽  
The Brain

During speech listening, the brain parses a continuous acoustic stream of information into computational units (e.g., syllables or words) necessary for speech comprehension. Recent neuroscientific hypotheses have proposed that neural oscillations contribute to speech parsing, but whether they do so on the basis of acoustic cues (bottom-up acoustic parsing) or as a function of available linguistic representations (top-down linguistic parsing) is unknown. In this magnetoencephalography study, we contrasted acoustic and linguistic parsing using bistable speech sequences. While listening to the speech sequences, participants were asked to maintain one of the two possible speech percepts through volitional control. We predicted that the tracking of speech dynamics by neural oscillations would not only follow the acoustic properties but also shift in time according to the participant's conscious speech percept. Our results show that the latency of high-frequency activity (specifically, beta and gamma bands) varied as a function of the perceptual report. In contrast, the phase of low-frequency oscillations was not strongly affected by top-down control. Whereas changes in low-frequency neural oscillations were compatible with the encoding of prelexical segmentation cues, high-frequency activity specifically informed on an individual's conscious speech percept.

scholarly journals Atypical cortical connectivity in autism spectrum disorder (ASD) as measured by magnetoencephalography (MEG)

2019 ◽  
Author(s):  
Robert A. Seymour ◽  
Paul Sowman ◽  
Klaus Kessler
Keyword(s):  
Autism Spectrum Disorder ◽  
Sensory Processing ◽  
Perspective Taking ◽  
Predictive Coding ◽  
Gamma Oscillations ◽  
Autism Spectrum ◽  
Spectrum Disorder ◽  
Alpha Oscillations ◽  
Oscillatory Power ◽  
Phase Amplitude

Autism Spectrum Disorder (ASD) is a neurodevelopmental condition, characterised by impairments in social interaction and communication, the presence of repetitive behaviours, and multisensory hyper- and hypo-sensitives. This thesis utilised magnetoencephalography, in combination with robust analysis techniques, to investigate the neural basis of ASD. Based on previous research, it was hypothesised that cortical activity in ASD would be associated with disruptions to oscillatory synchronisation during sensory processing, as well as during high-level perspective-taking. More specifically, a novel framework was introduced, based on local gamma-band dysregulation, global hypoconnectivity and deficient predictive-coding. To test this framework, data were collected from adolescents diagnosed with ASD and age-matched controls.Using a visual grating stimulus, it was found that in primary visual cortex, ASD participants had reduced coupling between the phase of alpha oscillations and the amplitude of gamma oscillations (i.e. phase amplitude coupling), suggesting dysregulated visual gamma in ASD. These findings were based on a robust analysis pipeline outlined in Chapter 2. Next, directed connectivity in the visual system was quantified using Granger causality. Compared with controls, ASD participants showed reductions in feedback connectivity, mediated by alpha oscillations, but no differences in inter-regional feedforward connectivity, mediated by gamma oscillations. In the auditory domain, it was found that ASD participants had reduced steady-state responses at 40Hz, in terms of oscillatory power and inter-trial coherence, again suggesting dysregulated gamma. Investigating predictive-coding theories of ASD using an auditory oddball paradigm, it was found that evoked responses to the omission of an expected tone were reduced for ASD participants. Finally, we found reductions in theta-band oscillatory power and connectivity for ASD participants, during embodied perspective-taking. Overall, these findings fit the proposed framework, and demonstrate that cortical activity in ASD is characterised by disruptions to oscillatory synchronisation, at the local and global scales, during both sensory processing and higher-level perspective-taking.Keywords: Autism Spectrum Disorder; Magnetoencephalography; Oscillations; Phase Amplitude Coupling; Connectivity.

Sign in / Sign up

Export Citation Format

Share Document