Alain Soquet

Mismatch negativity evoked by the McGurk–MacDonald effect: a phonetic representation within short-term memory

OBJECTIVES The McGurk-MacDonald illusory percept is obtained by dubbing an incongruent articulatory movement on an auditory phoneme. This type of audiovisual speech perception contributes to the assessment of theories of speech perception. The mismatch negativity (MMN) reflects the detection of a deviant stimulus within the auditory short-term memory and besides an acoustic component, possesses, under certain conditions, a phonetic one. The present study assessed the existence of an MMN evoked by McGurk-MacDonald percepts elicited by audiovisual stimuli with constant auditory components. METHODS Cortical evoked potentials were recorded using the oddball paradigm on 8 adults in 3 experimental conditions: auditory alone, visual alone and audiovisual stimulation. Obtaining illusory percepts was confirmed in an additional psychophysical condition. RESULTS The auditory deviant syllables and the audiovisual incongruent syllables elicited a significant MMN at Fz. In the visual condition, no negativity was observed either at Fz, or at O(z). CONCLUSIONS An MMN can be evoked by visual articulatory deviants, provided they are presented in a suitable auditory context leading to a phonetically significant interaction. The recording of an MMN elicited by illusory McGurk percepts suggests that audiovisual integration mechanisms in speech take place rather early during the perceptual processes.

The Influence of Ambient Speech on Adult Speech Productions through Unintentional Imitation

This paper deals with the influence of ambient speech on individual speech productions. A methodological framework is defined to gather the experimental data necessary to feed computer models simulating self-organisation in phonological systems. Two experiments were carried out. Experiment 1 was run on French native speakers from two regiolects of Belgium: two from Liège and two from Brussels. When exposed to the way of speaking of the other regiolect via loudspeakers, the speakers of one regiolect produced vowels that were significantly different from their typical realisations, and significantly closer to the way of speaking specific of the other regiolect. Experiment 2 achieved a replication of the results for 8 Mons speakers hearing a Liège speaker. A significant part of the imitative effect remained up to 10 min after the end of the exposure to the other regiolect productions. As a whole, the results suggest that: (i) imitation occurs automatically and unintentionally, (ii) the modified realisations leave a memory trace, in which case the mechanism may be better defined as ‘mimesis’ than as ‘imitation’. The potential effects of multiple imitative speech interactions on sound change are discussed in this paper, as well as the implications for a general theory of phonetic implementation and phonetic representation.

Electrophysiology of spatial scene analysis: the mismatch negativity (MMN) is sensitive to the ventriloquism illusion.

OBJECTIVES The ventriloquism effect is the tendency to underestimate the spatial separation between synchronous auditory and visual signals moderately separated in space. If, as it is thought, this effect is pre-attentive, it could modulate the mismatch negativity (MMN) that indexes the automatic, pre-attentive detection of deviant auditory stimuli rarely occurring in a sequence of standard stimuli. We assessed the existence of an MMN evoked by auditory and visual signals made up of standard sounds coming from the same location as the visual signal and deviant sounds coming from lateral deviations (20 or 60 degrees). As first observed in a behavioral study, a ventriloquism effect occurred for 20 degrees spatial separation but not for 60 degrees. METHODS Cortical evoked potentials were recorded using the oddball paradigm on 8 adults in auditory alone and audiovisual conditions. RESULTS In the auditory alone condition, each spatial localization contrast elicited a significant MMN. In the audiovisual condition, a significant MMN was only evoked for the 60 degrees contrasts. CONCLUSIONS The MMN evoked by spatial separation contrasts (20 and 60 degrees) in the auditory alone condition was suppressed by the corresponding audiovisual condition only when the latter yielded a ventriloquism effect, suggesting that this effect occurs at an early perceptual level.

Real-time MRI and articulatory coordination in speech.

This paper describes the real-time MRI technique and its use for the study of speech production. The two major problems, (i) the simultaneous recording of the MR images and the speech signal and (ii) the synchronisation of the images and of the speech signal, are addressed. Measurement accuracy on real-time images is evaluated by comparison with similar measurements on static MR images.

Mid-sagittal cut to area function transformations: direct measurements of mid-sagittal distance and area with MRI

This paper presents a comparative study of transformations used to compute the area of cross-sections of the vocal tract from the mid-sagittal measurements of the vocal tract. MRI techniques have been used to obtain both mid-sagittal distances and cross-sections of the vocal tract for French oral vowels uttered by two subjects. The measured cross-sectional areas can thus be compared to the cross-sectional areas computed by the different transformations. The evaluation is performed with a jackknife method where the parameters of the transformation are estimated from all but one measurement of a speakers vocal tract region and evaluated on the remaining measurement. This procedure allows the study of both the performance of the different forms of transformation as a function of the vocal tract region and the stability of the transformation parameters for a given vocal tract region. Three different forms of transformation are compared: linear, polynomial and power function. The estimation performances are also compared with four existing transformations.

Three-dimensional measurement of the vocal tract by MRI

Recent studies have shown that MRI techniques are reliable at measuring mid-sagittal cuts of the vocal tract. Three dimensional images of the vocal tract have been proposed e.g. by A.V. Foldvick et al. (1991; 1995) but despite this advance, these images do not offer any accurate measurement of sagittal cuts or area functions. Advance in MRI performances allows us to collect data much faster and with a better accuracy. We propose to use this new technology to make measurements of sagital, coronal, coronal oblique and transversal cuts. The data collected allow accurate measurement of area at different points along the vocal tract. The different cuts obtained by this method are then used to make measurements of the vocal tract shape. The study focuses on French oral vowels.

The aerodynamics of nasalization in French

Abstract Nasalization in French involves a complex interplay between several phonetic and phonological factors that have been, for the most part, investigated separately over the last 40 years. The present study provides a detailed account of the aerodynamics of French nasalization from eight Belgian French speakers reading word lists. Patterns of tautosyllabic nasal coarticulation are investigated in CV˜, NV˜, CV˜C, CV˜.CV, CV, NV, (C)VN, and NVN items, comparing different vowel and consonant types. Dependent variables involve temporal measures of both the extent of nasalization and its starting point relative to the oral–nasal boundary, and average flow rates across the acoustically defined segments. Results confirm previous findings that carryover nasalization is more extensive than anticipatory nasalization in French for both vowels and consonants. We further show that the temporal extent of intra-syllabic nasal coarticulatory airflow varies across vowel height and consonant manner of articulation and voicing. Various factors are considered in accounting for this variation.

Determination of velum opening for French nasal vowels by magnetic resonance imaging.

MRI techniques have been used to describe velum opening of French vowels. Data based on 18 joined axial slices of 4 mm thickness were recorded with four subjects. Differences in velum opening are calculated from areas measured in the tract between the lowered velum and the back pharynx wall. Results show that for all subjects, the back vowel [symbol: see text] has the smallest opening, while some variations are observed for the other vowels.

The role of sound intensity and stop-consonant voicing on McGurk fusions and combinations

When presented with an auditory /b/ dubbed onto a visual /g/, listeners sometimes perceive a fused phoneme like /d/ while with the reverse presentation, they experience a combination such as /bg/. This phenomenon reported by McGurk and MacDonald (1976) is here investigated in French for both voiced and voiceless stop consonants, using two levels of auditory intensity (70 dB vs 40 dB). In a first experiment, audiovisual incongruent monosyllables (A/bi/ V/gi/, A/gi/ V/bi/, A/ki/ V/pi/, A/pi/ V/ki/) uttered by a man and by a woman speaker were recorded and dubbed, using an analogical technology. In a second experiment, the same syllables articulated by the man speaker were recorded and dubbed according to digital technology. In a third experiment, the same materials as in the second experiment were used but the presentation procedure of the experimental items was changed: Audiovisual incongruent trials were mixed up with congruent ones. In the three experiments, the role of voicing and of auditory intensity were investigated. Overall, combinations were much more numerous than fusions and both types of illusions tended to increase at low intensity. Voicing had a differential effect on both types of illusions. Combinations were more numerous with voiceless consonants but fusions tended to occur more often with voiced ones. The number of illusions was affected by the dubbing technique but not by the presentation procedure.

Phonetic characteristics of an unexploded palatal implosive in Hendo

This paper describes the phonetic characteristics of an unexploded palatal implosive in Hendo, a Bantu language spoken in the Democratic Republic of Congo. This sound is an allophone of a palatal affricate implosive. Consideration of acoustic, aerodynamic and articulatory data obtained from various methods give a precise description of this sound.